To see the other types of publications on this topic, follow the link: RECOMMENDATION TECHNIQUES.
Dissertations / Theses on the topic 'RECOMMENDATION TECHNIQUES'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'RECOMMENDATION TECHNIQUES.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Martins, Diogo Marques. "Trajectory clustering techniques with application to route recommendation." Master's thesis, Universidade de Aveiro, 2015. http://hdl.handle.net/10773/18581.
Full textAbstract:
Mestrado em Engenharia de Computadores e TelemáticaO uso generalizado de dispositivos capazes de obter e transmitir dados sobre a localização de objetos ao longo do tempo tem permitido recolher grandes volumes de dados espácio-temporais. Por isso, tem-se assistido a uma procura crescente de técnicas e ferramentas para a análise de grandes volumes de dados espácio-temporais com o intuito de disponibilizar uma gama variada de serviços baseados na localização. Esta dissertação centra-se no desenvolvimento de um sistema para recomendaSr trajetos com base em dados históricos sobre a localização de objetos móveis ao longo do tempo. O principal problema estudado neste trabalho consiste no agrupamento de trajetórias e na extração de informação a partir dos grupos de trajetórias. Este estudo, não se restringe a dados provenientes apenas de veículos, podendo ser aplicado a outros tipos de trajetórias, por exemplo, percursos realizados por pessoas a pé ou de bicicleta. O agrupamento baseia-se numa medida de similaridade. A extração de informação consiste em criar uma trajetória representativa para cada grupo de trajetórias. As trajetórias representativas podem ser visualizadas usando uma aplicação web, sendo também possível configurar cada módulo do sistema com parâmetros desejáveis, na sua maioria distâncias limiares. Por fim, são apresentados casos de teste para avaliar o desempenho global do sistema desenvolvido.
The widespread use of devices to capture and transmit data about the location of objects over time allows collecting large volumes of spatio-temporal data. Consequently, there has been in recent years a growing demand for tools and techniques to analyze large volumes of spatio-temporal data aiming at providing a wide range of location-based services. This dissertation focuses on the development of a system for recommendation of trajectories based on historical data about the location of moving objects over time. The main issues covered in this work are trajectory clustering and extracting information from trajectory clusters. This study is not restricted to data from vehicles and can also be applied to other kinds of trajectories, for example, the movement of runners or bikes. The clustering is based on a similarity measure. The information extraction consists in creating a representative trajectory for the trajectories clusters. Finally, representative trajectories are displayed using a web application and it is also possible to configure each system module with desired parameters, mostly distance thresholds. Finally, case studies are presented to evaluate the developed system.
2
Vahabi, Hossein. "Recommendation techniques for Web search and social media." Thesis, IMT Alti Studi Lucca, 2012. http://e-theses.imtlucca.it/86/1/Vahabi_phdthesis.pdf.
Full textAbstract:
Recommenders aim to solve the information and interaction overload. A recommender system is a tool that identifies highly relevant items of interest for a user. In this thesis we aim to harness the information available through
Web to build recommender systems. In particularwe study on a number of different recommendation problems: "correct tag recommendation", "query recommendation", "tweet recommendation", and "tourist points of interest recommendation". For each problem, we describe and evaluate
effective solutions using novel efficient techniques.
3
Alabdulrahman, Rabaa. "Towards Personalized Recommendation Systems: Domain-Driven Machine Learning Techniques and Frameworks." Thesis, Université d'Ottawa / University of Ottawa, 2020. http://hdl.handle.net/10393/41012.
Full textAbstract:
Recommendation systems have been widely utilized in e-commerce settings to aid users through their shopping experiences. The principal advantage of these systems is their ability to narrow down the purchase options in addition to marketing items to customers. However, a number of challenges remain, notably those related to obtaining a clearer understanding of users, their profiles, and their preferences in terms of purchased items. Specifically, recommender systems based on collaborative filtering recommend items that have been rated by other users with preferences similar to those of the targeted users. Intuitively, the more information and ratings collected about the user, the more accurate are the recommendations such systems suggest.
In a typical recommender systems database, the data are sparse. Sparsity occurs when the number of ratings obtained by the users is much lower than the number required to build a prediction model. This usually occurs because of the users’ reluctance to share their reviews, either due to privacy issues or an unwillingness to make the extra effort. Grey-sheep users pose another challenge. These are users who shared their reviews and ratings yet disagree with the majority in the systems. The current state-of-the-art typically treats these users as outliers and removes them from the system. Our goal is to determine whether keeping these users in the system may benefit learning. Thirdly, cold-start problems refer to the scenario whereby a new item or user enters the system and is another area of active research. In this case, the system will have no information about the new user or item, making it problematic to find a correlation with others in the system. This thesis addresses the three above-mentioned research challenges through the development of machine learning methods for use within the recommendation system setting.
First, we focus on the label and data sparsity though the development of the Hybrid Cluster analysis and Classification learning (HCC-Learn) framework, combining supervised and unsupervised learning methods. We show that combining classification algorithms such as k-nearest neighbors and ensembles based on feature subspaces with cluster analysis algorithms such as expectation maximization, hierarchical clustering, canopy, k-means, and cascade k-means methods, generally produces high-quality results when applied to benchmark datasets. That is, cluster analysis clearly benefits the learning process, leading to high predictive accuracies for existing users.
Second, to address the cold-start problem, we present the Popular Users Personalized Predictions (PUPP-DA) framework. This framework combines cluster analysis and active learning, or so-called user-in-the-loop, to assign new customers to the most appropriate groups in our framework. Based on our findings from the HCC-Learn framework, we employ the expectation maximization soft clustering technique to create our user segmentations in the PUPP-DA framework, and we further incorporate Convolutional Neural Networks into our design. Our results show the benefits of user segmentation based on soft clustering and the use of active learning to improve predictions for new users. Furthermore, our findings show that focusing on frequent or popular users clearly improves classification accuracy. In addition, we demonstrate that deep learning outperforms machine learning techniques, notably resulting in more accurate predictions for individual users.
Thirdly, we address the grey-sheep problem in our Grey-sheep One-class Recommendations (GSOR) framework. The existence of grey-sheep users in the system results in a class imbalance whereby the majority of users will belong to one class and a small portion (grey-sheep users) will fall into the minority class. In this framework, we use one-class classification to provide a class structure for the training examples. As a pre-assessment stage, we assess the characteristics of grey-sheep users and study their impact on model accuracy. Next, as mentioned above, we utilize one-class learning, whereby we focus on the majority class to first learn the decision boundary in order to generate prediction lists for the grey-sheep (minority class). Our results indicate that including grey-sheep users in the training step, as opposed to treating them as outliers and removing them prior to learning, has a positive impact on the general predictive accuracy.
4
Aleksandra, Klašnja-Milićević. "Personalized Recommendation Based on Collaborative Tagging Techniques for an e‐Learning System." Phd thesis, Univerzitet u Novom Sadu, Prirodno-matematički fakultet u Novom Sadu, 2013. https://www.cris.uns.ac.rs/record.jsf?recordId=83535&source=NDLTD&language=en.
Full textAbstract:
The research topic involves personalization of an e‐learning system based oncollaborative tagging techniques integrated in a recommender system. Collaborative tagging systems allow users to upload their resources, and to label them with arbitrary words, so‐called tags. The systems can be distinguished according to what kind of resources are supported. Besides helping user to organize his or her personal collections, a tag also can be regarded as a user’s personal opinion expression. The increasing number of users providing information about themselves through social tagging activities caused the emergence of tag‐based profilingapproaches, which assume that users expose their preferences for certain contents through tag assignments. Thus, the tagging information can be used to make recommendations. Dissertation research aims to analyze and define an enhanced model to select tags that reveal the preferences and characteristics of users required to generate personalized recommendations. Options on the use of models for personalized tutoring system were also considered. Personalized learning occurs when e‐learning systems make deliberate efforts to design educational experiences that fit the needs, goals, talents, learning styles, interests of theirlearners and learners with similar characteristics. In practice, models defined in the dissertation were evaluated on tutoring system for teaching Java programming language.Predmet istraživanja disertacije obuhvata personalizaciju tutorskih sistema za elektronsko učenje primenom tehnika kolaborativnog tagovanja (collaborative tagging techniques) integrisanih u sisteme za generisanje preporuka (recommender systems). Tagovi, kao oblik meta podataka, predstavljaju proizvoljne ključne reči ili fraze koje korisnik može da upotrebi za označavanje različitih sadržaja. Pored toga što tagovi korisnicima pružaju pomoć u organizaciji sadržaja, oni su korisni i u izražavanju mišljenja korisnika. Veliki broj informacija koje korisnici pružaju o sebi kroz aktivnosti tagovanja otvorio je mogućnost primene tagova u generisanju preporuka. Istraživanje disertacije je usmereno na analizu i definisanje poboljšanih modela za odabir tagova koji otkrivaju sklonosti i osobine korisnika potrebne za generisanje personalizovanih preporuka. Razmatrane su i mogućnosti primene tako dobijenih modela za personalizaciju tutorskih sistema. Personalizovani tutorski sistemi korisniku pružaju optimalne putanje kretanja i adekvatne aktivnosti učenja na osnovu njegovih osobina, njegovog stila učenja, znanja koje on poseduje u toj oblasti, kao i prethodnog iskustva korisnika sistema koji imaju slične karakteristike. Modeli definisani u disertaciji u praksi su evaluirani na tutorskom sistemu za učenje programskog jezika Java.
5
Nagi, Mohamad. "Integrating Network Analysis and Data Mining Techniques into Effective Framework for Web Mining and Recommendation. A Framework for Web Mining and Recommendation." Thesis, University of Bradford, 2015. http://hdl.handle.net/10454/14200.
Full textAbstract:
The main motivation for the study described in this dissertation is to benefit from the development in technology and the huge amount of available data which can be easily captured, stored and maintained electronically. We concentrate on Web usage (i.e., log) mining and Web structure mining. Analysing Web log data will reveal valuable feedback reflecting how effective the current structure of a web site is and to help the owner of a web site in understanding the behaviour of the web site visitors. We developed a framework that integrates statistical analysis, frequent pattern mining, clustering, classification and network construction and analysis. We concentrated on the statistical data related to the visitors and how they surf and pass through the various pages of a given web site to land at some target pages. Further, the frequent pattern mining technique was used to study the relationship between the various pages constituting a given web site. Clustering is used to study the similarity of users and pages. Classification suggests a target class for a given new entity by comparing the characteristics of the new entity to those of the known classes. Network construction and analysis is also employed to identify and investigate the links between the various pages constituting a Web site by constructing a network based on the frequency of access to the Web pages such that pages get linked in the network if they are identified in the result of the frequent pattern mining process as frequently accessed together. The knowledge discovered by analysing a web site and its related data should be considered valuable for online shoppers and commercial web site owners. Benefitting from the outcome of the study, a recommendation system was developed to suggest pages to visitors based on their profiles as compared to similar profiles of other visitors. The conducted experiments using popular datasets demonstrate the applicability and effectiveness of the proposed framework for Web mining and recommendation. As a by product of the proposed method, we demonstrate how it is effective in another domain for feature reduction by concentrating on gene expression data analysis as an application with some interesting results reported in Chapter 5.
6
Guerrini, Gabriele. "Analysis, design and implementation of a parking recommendation system applying machine learning techniques." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2021. http://amslaurea.unibo.it/23162/.
Full textAbstract:
Nowadays, smart city and urban mobility topics are focusing growing interest for business and R\&D purposes both, and one of its trendy application regards the study and the realization of parking recommendation systems.
In our case, a parking recommendation system is going to be developed for the city of Bologna. After a brief revision of similar proposed solutions and their approaches and findings, a first data analysis phase is going to be executed to explore the data and to mine any useful information about the parking behavior. Then, once the adopted forecasting model has been exposed, the design and the development of the system are going to be described. In the end, the performances of the system are going to be evaluated under multiple points of view, considering some project-specific constraints too.
7
Cabir, Hassane Natu Hassane. "A Comparison Of Different Recommendation Techniques For A Hybrid Mobile Game Recommender System." Master's thesis, METU, 2012. http://etd.lib.metu.edu.tr/upload/12615173/index.pdf.
Full textAbstract:
As information continues to grow at a very fast pace, our ability to access this
information effectively does not, and we are often realize how harder is getting to
locate an object quickly and easily. The so-called personalization technology is one
of the best solutions to this information overload problem: by automatically learning
the user profile, personalized information services have the potential to offer users a
more proactive and intelligent form of information access that is designed to assist
us in finding interesting objects. Recommender systems, which have emerged as a
solution to minimize the problem of information overload, provide us with
recommendations of content suited to our needs. In order to provide
recommendations as close as possible to a user&rsquos taste, personalized recommender systems require accurate user models of characteristics, preferences and needs. Collaborative filtering is a widely accepted technique to provide recommendations based on ratings of similar users, But it suffers from several issues like data sparsity and cold start. In one-class collaborative filtering, a special type of collaborative filtering methods that aims to deal with datasets that lack counter-examples, the challenge is even greater, since these datasets are even sparser. In this thesis, we present a series of experiments conducted on a real-life customer purchase database from a major Turkish E-Commerce site. The sparsity problem is handled by the use of content-based technique combined with TFIDF weights, memory based collaborative filtering combined with different similarity measures and also hybrids approaches, and also model based collaborative filtering with the use of Singular Value Decomposition (SVD). Our study showed that the binary similarity measure and SVD outperform conventional measures in this OCCF dataset.
8
Videla, Cavieres Iván Fernando. "Improvement of recommendation system for a wholesale store chain using advanced data mining techniques." Tesis, Universidad de Chile, 2015. http://repositorio.uchile.cl/handle/2250/133522.
Full textAbstract:
Magíster en Gestión de OperacionesIngeniero Civil Industrial
En las empresas de Retail, las áreas de Customer Intelligence tienen muchas oportunidades de mejorar sus decisiones estratégicas a partir de la información que podrían obtener de los registros de interacciones con sus clientes. Sin embargo se ha convertido en un desafío poder procesar estos grandes volúmenes de datos. Uno de los problemas que se enfrentan día a día es segmentar o agrupar clientes. La mayoría de las empresas generan agrupaciones según nivel de gasto, no por similitud en sus canastas de compra, como propone la literatura. Otro desafío de estas empresas es aumentar las ventas en cada visita del cliente y fidelizar. Una de las técnicas utilizadas para lograrlo es usar sistemas de recomendación. En este trabajo se proceso ́ alrededor de medio billón de registros transaccionales de una cadena de supermercados mayorista. Al aplicar las técnicas tradicionales de Clustering y Market Basket Analysis los resultados son de baja calidad, haciendo muy difícil la interpretación, además no se logra identificar grupos que permitan clasificar a un cliente de acuerdo a sus compras históricas. Entendiendo que la presencia simultánea de dos productos en una misma boleta implica una relación entre ellos, se usó un método de graph mining basado en redes sociales que permitió obtener grupos de productos identificables que denominamos comunidades, a las que puede pertenecer un cliente. La robustez del modelo se comprueba por la estabilidad de los grupos generados en distintos periodos de tiempo. Bajo las mismas restricciones que la empresa exige, se generan recomendaciones basadas en las compras históricas y en la pertenencia de los clientes a los distintos grupos de productos. De esta manera, los clientes reciben recomendaciones mucho más pertinentes y no solo son basadas en los que otros clientes también compraron. La novedosa forma de resolver el problema de segmentar clientes ayuda a mejorar en un 140% el actual método de recomendaciones que utiliza la cadena Chilena de supermercados mayoristas. Esto se traduce en un aumento de más de 430% de los ingresos posibles.
9
Holländer, John. "Investigating the performance of matrix factorization techniques applied on purchase data for recommendation purposes." Thesis, Malmö högskola, Fakulteten för teknik och samhälle (TS), 2015. http://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-20624.
Full textAbstract:
Automated systems for producing product recommendations to users is a relatively new area within the field of machine learning. Matrix factorization techniques have been studied to a large extent on data consisting of explicit feedback such as ratings, but to a lesser extent on implicit feedback data consisting of for example purchases.The aim of this study is to investigate how well matrix factorization techniques perform compared to other techniques when used for producing recommendations based on purchase data. We conducted experiments on data from an online bookstore as well as an online fashion store, by running algorithms processing the data and using evaluation metrics to compare the results. We present results proving that for many types of implicit feedback data, matrix factorization techniques are inferior to various neighborhood- and association rules techniques for producing product recommendations. We also present a variant of a user-based neighborhood recommender system algorithm \textit{(UserNN)}, which in all tests we ran outperformed both the matrix factorization algorithms and the k-nearest neighbors algorithm regarding both accuracy and speed. Depending on what dataset was used, the UserNN achieved a precision approximately 2-22 percentage points higher than those of the matrix factorization algorithms, and 2 percentage points higher than the k-nearest neighbors algorithm. The UserNN also outperformed the other algorithms regarding speed, with time consumptions 3.5-5 less than those of the k-nearest neighbors algorithm, and several orders of magnitude less than those of the matrix factorization algorithms.
10
Bambia, Meriam. "Jointly integrating current context and social influence for improving recommendation." Thesis, Toulouse 3, 2017. http://www.theses.fr/2017TOU30110/document.
Full textAbstract:
La diversité des contenus recommandation et la variation des contextes des utilisateurs rendent la prédiction en temps réel des préférences des utilisateurs de plus en plus difficile mettre en place. Toutefois, la plupart des approches existantes n'utilisent que le temps et l'emplacement actuels séparément et ignorent d'autres informations contextuelles sur lesquelles dépendent incontestablement les préférences des utilisateurs (par exemple, la météo, l'occasion). En outre, ils ne parviennent pas considérer conjointement ces informations contextuelles avec les interactions sociales entre les utilisateurs. D'autre part, la résolution de problèmes classiques de recommandation (par exemple, aucun programme de télévision vu par un nouvel utilisateur connu sous le nom du problème de démarrage froid et pas assez d'items co-évalués par d'autres utilisateurs ayant des préférences similaires, connu sous le nom du problème de manque de donnes) est d'importance significative puisque sont attaqués par plusieurs travaux. Dans notre travail de thèse, nous proposons un modèle probabiliste qui permet exploiter conjointement les informations contextuelles actuelles et l'influence sociale afin d'améliorer la recommandation des items. En particulier, le modèle probabiliste vise prédire la pertinence de contenu pour un utilisateur en fonction de son contexte actuel et de son influence sociale. Nous avons considérer plusieurs éléments du contexte actuel des utilisateurs tels que l'occasion, le jour de la semaine, la localisation et la météo. Nous avons utilisé la technique de lissage Laplace afin d'éviter les fortes probabilités. D'autre part, nous supposons que l'information provenant des relations sociales a une influence potentielle sur les préférences des utilisateurs. Ainsi, nous supposons que l'influence sociale dépend non seulement des évaluations des amis mais aussi de la similarité sociale entre les utilisateurs. Les similarités sociales utilisateur-ami peuvent être établies en fonction des interactions sociales entre les utilisateurs et leurs amis (par exemple les recommandations, les tags, les commentaires). Nous proposons alors de prendre en compte l'influence sociale en fonction de la mesure de similarité utilisateur-ami afin d'estimer les préférences des utilisateurs. Nous avons mené une série d'expérimentations en utilisant un ensemble de donnes réelles issues de la plateforme de TV sociale Pinhole. Cet ensemble de donnes inclut les historiques d'accès des utilisateurs-vidéos et les réseaux sociaux des téléspectateurs. En outre, nous collectons des informations contextuelles pour chaque historique d'accès utilisateur-vidéo saisi par le système de formulaire plat. Le système de la plateforme capture et enregistre les dernières informations contextuelles auxquelles le spectateur est confronté en regardant une telle vidéo.Dans notre évaluation, nous adoptons le filtrage collaboratif axé sur le temps, le profil dépendant du temps et la factorisation de la matrice axe sur le réseau social comme tant des modèles de référence. L'évaluation a port sur deux tâches de recommandation. La première consiste sélectionner une liste trie de vidéos. La seconde est la tâche de prédiction de la cote vidéo. Nous avons évalué l'impact de chaque élément du contexte de visualisation dans la performance de prédiction. Nous testons ainsi la capacité de notre modèle résoudre le problème de manque de données et le problème de recommandation de démarrage froid du téléspectateur. Les résultats expérimentaux démontrent que notre modèle surpasse les approches de l'état de l'art fondes sur le facteur temps et sur les réseaux sociaux. Dans les tests des problèmes de manque de donnes et de démarrage froid, notre modèle renvoie des prédictions cohérentes différentes valeurs de manque de donnéesDue to the diversity of alternative contents to choose and the change of users' preferences, real-time prediction of users' preferences in certain users' circumstances becomes increasingly hard for recommender systems. However, most existing context-aware approaches use only current time and location separately, and ignore other contextual information on which users' preferences may undoubtedly depend (e.g. weather, occasion). Furthermore, they fail to jointly consider these contextual information with social interactions between users. On the other hand, solving classic recommender problems (e.g. no seen items by a new user known as cold start problem, and no enough co-rated items with other users with similar preference as sparsity problem) is of significance importance since it is drawn by several works. In our thesis work, we propose a context-based approach that leverages jointly current contextual information and social influence in order to improve items recommendation. In particular, we propose a probabilistic model that aims to predict the relevance of items in respect with the user's current context. We considered several current context elements such as time, location, occasion, week day, location and weather. In order to avoid strong probabilities which leads to sparsity problem, we used Laplace smoothing technique. On the other hand, we argue that information from social relationships has potential influence on users' preferences. Thus, we assume that social influence depends not only on friends' ratings but also on social similarity between users. We proposed a social-based model that estimates the relevance of an item in respect with the social influence around the user on the relevance of this item. The user-friend social similarity information may be established based on social interactions between users and their friends (e.g. recommendations, tags, comments). Therefore, we argue that social similarity could be integrated using a similarity measure. Social influence is then jointly integrated based on user-friend similarity measure in order to estimate users' preferences. We conducted a comprehensive effectiveness evaluation on real dataset crawled from Pinhole social TV platform. This dataset includes viewer-video accessing history and viewers' friendship networks. In addition, we collected contextual information for each viewer-video accessing history captured by the plat form system. The platform system captures and records the last contextual information to which the viewer is faced while watching such a video. In our evaluation, we adopt Time-aware Collaborative Filtering, Time-Dependent Profile and Social Network-aware Matrix Factorization as baseline models. The evaluation focused on two recommendation tasks. The first one is the video list recommendation task and the second one is video rating prediction task. We evaluated the impact of each viewing context element in prediction performance. We tested the ability of our model to solve data sparsity and viewer cold start recommendation problems. The experimental results highlighted the effectiveness of our model compared to the considered baselines. Experimental results demonstrate that our approach outperforms time-aware and social network-based approaches. In the sparsity and cold start tests, our approach returns consistently accurate predictions at different values of data sparsity
11
Goetz, Charity. "Textile dyes techniques and their effects on the environment with a recommendation for dyers concerning the Green effect /." Lynchburg, Va. : Liberty University, 2008. http://digitalcommons.liberty.edu.
Full text
12
Costa, Arthur Fortes da. "Recomendação de conteúdo baseada em interações multimodais." Universidade de São Paulo, 2015. http://www.teses.usp.br/teses/disponiveis/55/55134/tde-09042015-153225/.
Full textAbstract:
A oferta de produtos,informação e serviços a partir de perfis de usuários tem tornado os sistemas de recomendação cada vez mais presentes na Web, aumentando a facilidade de escolha e de permanência dos usuários nestes sistemas. Entretanto, existem otimizações a serem feitas principalmente com relação à modelagem do perfil do usuário. Geralmente, suas preferências são modeladas de modo superficial, devido à escassez das informações coletadas,como notas ou comentários, ou devido a informações indutivas que estão suscetíveis a erros. Esta dissertação propõe uma ferramenta de recomendação baseado em interações multimodais, capaz de combinar informações de usuários processadas individualmente por algoritmos de recomendação tradicionais. Nesta ferramenta desenvolveram-se quatro técnicas de combinação afim fornecer aos sistemas de recomendação, subsídios para melhoria na qualidade das predições em diversos domínios.Providing products, information and services from user profiles has made the recommendation systems to be increasingly present, increasing the ease of selection and retention of users in Webservices. However, there are optimizations to be made in these systems mainly with respect to modeling the user profile. Generally, the preferences are modeled superficially, due to the scarcity of information collected, as notes or comments, or because of inductive information that is susceptible to errors. This work proposes are commendation tool based on multimodal interactions that combines users\' interactions, wich are processed individually by traditional recommendation algorithms. In this tool developed four combination of techniques in order to provide recommendation systems subsidies to improve the quality of predictions.
13
Embarak, Ossama Hashem Khamis. "A new technique for intelligent web personal recommendation." Thesis, Heriot-Watt University, 2011. http://hdl.handle.net/10399/2503.
Full textAbstract:
Personal recommendation systems nowadays are very important in web applications because of the available huge volume of information on the World Wide Web, and the necessity to save users’ time, and provide appropriate desired information, knowledge, items, etc. The most popular recommendation systems are collaborative filtering systems, which suffer from certain problems such as cold-start, privacy, user identification, and scalability. In this thesis, we suggest a new method to solve the cold start problem taking into consideration the privacy issue. The method is shown to perform very well in comparison with alternative methods, while having better properties regarding user privacy. The cold start problem covers the situation when recommendation systems have not sufficient information about a new user’s preferences (the user cold start problem), as well as the case of newly added items to the system (the item cold start problem), in which case the system will not be able to provide recommendations. Some systems use users’ demographical data as a basis for generating recommendations in such cases (e.g. the Triadic Aspect method), but this solves only the user cold start problem and enforces user’s privacy. Some systems use users’ ’stereotypes’ to generate recommendations, but stereotypes often do not reflect the actual preferences of individual users. While some other systems use user’s ’filterbots’ by injecting pseudo users or bots into the system and consider these as existing ones, but this leads to poor accuracy. We propose the active node method, that uses previous and recent users’ browsing targets and browsing patterns to infer preferences and generate recommendations (node recommendations, in which a single suggestion is given, and batch recommendations, in which a set of possible target nodes are shown to the user at once). We compare the active node method with three alternative methods (Triadic Aspect Method, Naïve Filterbots Method, and MediaScout Stereotype Method), and we used a dataset collected from online web news to generate recommendations based on our method and based on the three alternative methods. We calculated the levels of novelty, coverage, and precision in these experiments, and we found that our method achieves higher levels of novelty in batch recommendation while achieving higher levels of coverage and precision in node recommendations comparing to these alternative methods. Further, we develop a variant of the active node method that incorporates semantic structure elements. A further experimental evaluation with real data and users showed that semantic node recommendation with the active node method achieved higher levels of novelty than nonsemantic node recommendation, and semantic-batch recommendation achieved higher levels of coverage and precision than non-semantic batch recommendation.
14
Ammari, Ahmad N. "Transforming user data into user value by novel mining techniques for extraction of web content, structure and usage patterns. The Development and Evaluation of New Web Mining Methods that enhance Information Retrieval and improve the Understanding of User¿s Web Behavior in Websites and Social Blogs." Thesis, University of Bradford, 2010. http://hdl.handle.net/10454/5269.
Full textAbstract:
The rapid growth of the World Wide Web in the last decade makes it the largest publicly accessible data source in the world, which has become one of the most significant and influential information revolution of modern times. The influence of the Web has impacted almost every aspect of humans' life, activities and fields, causing paradigm shifts and transformational changes in business, governance, and education. Moreover, the rapid evolution of Web 2.0 and the Social Web in the past few years, such as social blogs and friendship networking sites, has dramatically transformed the Web from a raw environment for information consumption to a dynamic and rich platform for information production and sharing worldwide. However, this growth and transformation of the Web has resulted in an uncontrollable explosion and abundance of the textual contents, creating a serious challenge for any user to find and retrieve the relevant information that he truly seeks to find on the Web. The process of finding a relevant Web page in a website easily and efficiently has become very difficult to achieve. This has created many challenges for researchers to develop new mining techniques in order to improve the user experience on the Web, as well as for organizations to understand the true informational interests and needs of their customers in order to improve their targeted services accordingly by providing the products, services and information that truly match the requirements of every online customer.
With these challenges in mind, Web mining aims to extract hidden patterns and discover useful knowledge from Web page contents, Web hyperlinks, and Web usage logs. Based on the primary kinds of Web data used in the mining process, Web mining tasks can be categorized into three main types: Web content mining, which extracts knowledge from Web page contents using text mining techniques, Web structure mining, which extracts patterns from the hyperlinks that represent the structure of the website, and Web usage mining, which mines user's Web navigational patterns from Web server logs that record the Web page access made by every user, representing the interactional activities between the users and the Web pages in a website. The main goal of this thesis is to contribute toward addressing the challenges that have been resulted from the information explosion and overload on the Web, by proposing and developing novel Web mining-based approaches. Toward achieving this goal, the thesis presents, analyzes, and evaluates three major contributions. First, the development of an integrated Web structure and usage mining approach that recommends a collection of hyperlinks for the surfers of a website to be placed at the homepage of that website. Second, the development of an integrated Web content and usage mining approach to improve the understanding of the user's Web behavior and discover the user group interests in a website. Third, the development of a supervised classification model based on recent Social Web concepts, such as Tag Clouds, in order to improve the retrieval of relevant articles and posts from Web social blogs.
15
Ammari, Ahmad N. "Transforming user data into user value by novel mining techniques for extraction of web content, structure and usage patterns : the development and evaluation of new Web mining methods that enhance information retrieval and improve the understanding of users' Web behavior in websites and social blogs." Thesis, University of Bradford, 2010. http://hdl.handle.net/10454/5269.
Full textAbstract:
The rapid growth of the World Wide Web in the last decade makes it the largest publicly accessible data source in the world, which has become one of the most significant and influential information revolution of modern times. The influence of the Web has impacted almost every aspect of humans' life, activities and fields, causing paradigm shifts and transformational changes in business, governance, and education. Moreover, the rapid evolution of Web 2.0 and the Social Web in the past few years, such as social blogs and friendship networking sites, has dramatically transformed the Web from a raw environment for information consumption to a dynamic and rich platform for information production and sharing worldwide. However, this growth and transformation of the Web has resulted in an uncontrollable explosion and abundance of the textual contents, creating a serious challenge for any user to find and retrieve the relevant information that he truly seeks to find on the Web. The process of finding a relevant Web page in a website easily and efficiently has become very difficult to achieve. This has created many challenges for researchers to develop new mining techniques in order to improve the user experience on the Web, as well as for organizations to understand the true informational interests and needs of their customers in order to improve their targeted services accordingly by providing the products, services and information that truly match the requirements of every online customer. With these challenges in mind, Web mining aims to extract hidden patterns and discover useful knowledge from Web page contents, Web hyperlinks, and Web usage logs. Based on the primary kinds of Web data used in the mining process, Web mining tasks can be categorized into three main types: Web content mining, which extracts knowledge from Web page contents using text mining techniques, Web structure mining, which extracts patterns from the hyperlinks that represent the structure of the website, and Web usage mining, which mines user's Web navigational patterns from Web server logs that record the Web page access made by every user, representing the interactional activities between the users and the Web pages in a website. The main goal of this thesis is to contribute toward addressing the challenges that have been resulted from the information explosion and overload on the Web, by proposing and developing novel Web mining-based approaches. Toward achieving this goal, the thesis presents, analyzes, and evaluates three major contributions. First, the development of an integrated Web structure and usage mining approach that recommends a collection of hyperlinks for the surfers of a website to be placed at the homepage of that website. Second, the development of an integrated Web content and usage mining approach to improve the understanding of the user's Web behavior and discover the user group interests in a website. Third, the development of a supervised classification model based on recent Social Web concepts, such as Tag Clouds, in order to improve the retrieval of relevant articles and posts from Web social blogs.
16
Zou, Hai Tao. "Local topology of social networks in supporting recommendations and diversity identification of reviews." Thesis, University of Macau, 2015. http://umaclib3.umac.mo/record=b3335434.
Full text
17
Wu, Kevin (Kevin L. ). "DeepTuner : a system for search technique recommendation in program autotuning." Thesis, Massachusetts Institute of Technology, 2015. http://hdl.handle.net/1721.1/115462.
Full textAbstract:
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, September 2015."July 2015." Cataloged from PDF version of thesis.
Includes bibliographical references (pages 86-89).
OpenTuner can help users achieve better or more portable performance in their specific domain through program autotuning. A key challenge for users seeking good autotuning performance in OpenTuner is selecting a search approach appropriate for problem. However, not only are current in-situ learning search approaches not robust enough to handle all search spaces, but there are also too many possible search approaches for a user to examine manually after factoring in composable techniques. In this thesis, we introduce DeepTuner, a system for search approach recommendation operating across OpenTuner autotuning sessions to facilitate development of robust transfer learning search approaches. By utilizing historical autotuning data via DeepTuner's technique recommendation endpoints, the new search approaches can efficiently explore the space of possible search approaches and the autotuning space simultaneously, resulting in an adaptive, self-improving search approach. We demonstrate the robustness that recommendation brings on nine problems spread over three domains for a variety of initial technique sets. In particular, we show that the new Database Initialized Recommendation Bandit Meta-technique is highly robust, performing on par or significantly better than various old in-situ search approaches in OpenTuner. We achieve up to 3.7x performance improvement over the old default in-situ search approach for OpenTuner in the TSP domain.
by Kevin Wu.
M. Eng.
18
Almuhaimeed, Abdullah. "Enhancing recommendations in specialist search through semantic-based techniques and multiple resources." Thesis, University of Essex, 2016. http://repository.essex.ac.uk/17584/.
Full textAbstract:
Information resources abound on the Internet, but mining these resources is a non-trivial task. Such abundance has raised the need to enhance services provided to users, such as recommendations. The purpose of this work is to explore how better recommendations can be provided to specialists in specific domains such as bioinformatics by introducing semantic techniques that reason through different resources and using specialist search techniques. Such techniques exploit semantic relations and hidden associations that occur as a result of the information overlapping among various concepts in multiple bioinformatics resources such as ontologies, websites and corpora. Thus, this work introduces a new method that reasons over different bioinformatics resources and then discovers and exploits different relations and information that may not exist in the original resources. Such relations may be discovered as a consequence of the information overlapping, such as the sibling and semantic similarity relations, to enhance the accuracy of the recommendations provided on bioinformatics content (e.g. articles). In addition, this research introduces a set of semantic rules that are able to extract different semantic information and relations inferred among various bioinformatics resources. This project introduces these semantic-based methods as part of a recommendation service within a content-based system. Moreover, it uses specialists' interests to enhance the provided recommendations by employing a method that is collecting user data implicitly. Then, it represents the data as adaptive ontological user profiles for each user based on his/her preferences, which contributes to more accurate recommendations provided to each specialist in the field of bioinformatics.
19
Rammal, Dina. "Thermo-mechanical behaviour of geothermal structures : numerical modelling and recommendations." Thesis, Lille 1, 2017. http://www.theses.fr/2017LIL10172/document.
Full textAbstract:
Les structures géothermiques agissent comme éléments échangeurs de chaleur en plus de leur rôle majeur en tant que structures porteuses. De ce fait, ils sont soumis à des sollicitations thermiques en plus des charges mécaniques. Cependant, leurs méthodes de dimensionnement ne sont pas encore clairement définies. Ce travail est divisé en deux parties principales qui couvrent le dimensionnement thermique et mécanique des pieux géothermiques et des parois moulées. En ce qui concerne les performances thermiques des structures géothermiques, deux stratégies sont introduites qui sont capables d'évaluer les énergies conductives et advectives échangées admissibles. Ils permettent de distinguer les différentes formes d'énergies échangées et montrent comment elles peuvent varier en cas de chargement thermique cyclique. Des modèles numériques couplés thermo-hydraulique bidimensionnels et tridimensionnels ont été réalisés et la performance thermique des structures géothermiques a été évaluée en fonction des deux approches présentées. En ce qui concerne le dimensionnement mécanique, ce travail couvre les problèmes liés au choix de la sollicitation thermique que le concepteur doit considérer pour la conception mécanique des structures géothermiques telles que le nombre de cycles, l'amplitude thermique cyclique et l'influence de l'ordre de chargement thermique. Ce travail traite ces problèmes dans le but de faciliter la conception des structures géothermiques. Des recommandations sont données pour la conception mécanique des pieux géothermiques et des parois moulées basées sur les résultats obtenus à partir des analyses numériques thermo-mécaniqueGeothermal structures act as heat exchanger elements in addition to their major role as bearing structures. Thus, they are subjected to thermal solicitations as well as to mechanical loading. However, their design methods are not clearly defined yet. This work is divided into two main parts that cover the thermal and mechanical design of thermo-active piles and diaphragm walls. Regarding the thermal performance of geothermal structures, two strategies are introduced that are capable to evaluate the allowable exchanged conductive and advective energies. They help to distinguish between different forms of exchanged energies and show how they may vary under cyclic thermal loading. Two and three dimensional hydro-thermal numerical models have been conducted and the thermal performance of geothermal structures has been evaluated based on the two presented approaches. Regarding the mechanical design, this work covers the issues related to the choice of the thermal solicitation that the designer has to consider for the mechanical design of geothermal structures such as the number of cycles, cyclic thermal amplitude, and influence of the thermal loading order. This work deals with these issues with the aim to facilitate the design of geothermal structures. Recommendations are given for the mechanical design of both thermo-active piles and diaphragm walls based on the results obtained from the thermo-mechanical numerical analyses
20
Armstrong, Norma B. "Software estimating : a description and analysis of current methodologies with recommendations on appropriate techniques for estimating RIT Research Corporatin software projects /." Online version of thesis, 1987. http://hdl.handle.net/1850/10296.
Full text
21
ALCANTARA, MARCELA C. "Avaliacao dos criterios de qualidade de imagem e estudo das doses em um departamento de mamografia." reponame:Repositório Institucional do IPEN, 2009. http://repositorio.ipen.br:8080/xmlui/handle/123456789/9470.
Full textAbstract:
Made available in DSpace on 2014-10-09T12:27:06Z (GMT). No. of bitstreams: 0Made available in DSpace on 2014-10-09T13:56:46Z (GMT). No. of bitstreams: 0
Dissertacao (Mestrado)
IPEN/D
Instituto de Pesquisas Energeticas e Nucleares - IPEN-CNEN/SP
22
Mkpong, Offiong Etim. "The effects of variable moisture levels on extractable Bray-l P, Bray-l Al, Bray-l Fe, Bray-l Mn : fertilizer P recommendation based on quicktest technique /." The Ohio State University, 1987. http://rave.ohiolink.edu/etdc/view?acc_num=osu1487329662144707.
Full text
23
Zomahoun, Damien Esse. "Emergsem : une approche d'annotation collaborative et de recherche d'images basée sur les sémantiques émergentes." Thesis, Dijon, 2015. http://www.theses.fr/2015DIJOS019/document.
Full textAbstract:
L’extraction de la sémantique d’une image est un processus qui nécessite une analyse profonde du contenu de l’image. Elle se réfère à leur interprétation à partir d’un point de vuehumain. Dans ce dernier cas, la sémantique d’une image pourrait être générique (par exemple un véhicule) ou spécifique (par exemple une bicyclette). Elle consiste à extraire une sémantique simple ou multiple de l’image afin de faciliter sa récupération. Ces objectifs indiquent clairement que l’extraction de la sémantique n’est pas un nouveau domaine de recherche. Cette thèse traite d’une approche d’annotation collaborative et de recherche d’images baséesur les sémantiques émergentes. Il aborde d’une part, la façon dont les annotateurs pourraient décrire et représenter le contenu des images en se basant sur les informations visuelles, et d’autre part comment la recherche des images pourrait être considérablement améliorée grâce aux récentes techniques, notamment le clustering et la recommandation. Pour atteindre ces objectifs, l’exploitation des outils de description implicite du contenu des images, des interactions des annotateurs qui décrivent la sémantique des images et celles des utilisateurs qui utilisent la sémantique produite pour rechercher les images seraient indispensables.Dans cette thèse, nous nous sommes penchés vers les outils duWeb Sémantique, notamment les ontologies pour décrire les images de façon structurée. L’ontologie permet de représenter les objets présents dans une image ainsi que les relations entre ces objets (les scènes d’image). Autrement dit, elle permet de représenter de façon formelle les différents types d’objets et leurs relations. L’ontologie code la structure relationnelle des concepts que l’on peut utiliser pour décrire et raisonner. Cela la rend éminemment adaptée à de nombreux problèmes comme la description sémantique des images qui nécessite une connaissance préalable et une capacité descriptive et normative.La contribution de cette thèse est focalisée sur trois points essentiels : La représentationsémantique, l’annotation sémantique collaborative et la recherche sémantique des images.La représentation sémantique permet de proposer un outil capable de représenter la sémantique des images. Pour capturer la sémantique des images, nous avons proposé une ontologie d’application dérivée d’une ontologie générique.L’annotation sémantique collaborative que nous proposons consiste à faire émerger la sémantique des images à partir des sémantiques proposées par une communauté d’annotateurs.La recherche sémantique permet de rechercher les images avec les sémantiques fournies par l’annotation sémantique collaborative. Elle est basée sur deux techniques : le clustering et la recommandation. Le clustering permet de regrouper les images similaires à la requête d’utilisateur et la recommandation a pour objectif de proposer des sémantiques aux utilisateurs en se basant sur leurs profils statiques et dynamiques. Elle est composée de trois étapes à savoir : la formation de la communauté des utilisateurs, l’acquisition des profils d’utilisateurs et la classification des profils d’utilisateurs avec l’algèbre de Galois. Des expérimentations ont été menées pour valider les différentes approches proposées dans ce travailThe extraction of images semantic is a process that requires deep analysis of the image content. It refers to their interpretation from a human point of view. In this lastest case, the image semantic may be generic (e.g., a vehicle) or specific (e.g., a bicycle). It consists in extracting single or multiple images semantic in order to facilitate its retrieval. These objectives clearly show that the extraction of semantic is not a new research field. This thesis deals with the semantic collaborative annotation of images and their retrieval. Firstly, it discusses how annotators could describe and represent images content based on visual information, and secondly how images retrieval could be greatly improved thank to latest techniques, such as clustering and recommendation. To achieve these purposes, the use of implicit image content description tools, interactions of annotators that describe the semantics of images and those of users that use generated semantics to retrieve the images, would be essential. In this thesis, we focus our research on the use of Semantic Web tools, in particular ontologies to produce structured descriptions of images. Ontology is used to represent image objects and the relationships between these objects. In other words, it allows to formally represent the different types of objects and their relationships. Ontology encodes the relational structure of concepts that can be used to describe and reason. This makes them eminently adapted to many problems such as semantic description of images that requires prior knowledge as well as descriptive and normative capacity. The contribution of this thesis is focused on three main points : semantic representation, collaborative semantic annotation and semantic retrieval of images.Semantic representation allows to offer a tool for the capturing semantics of images. To capture the semantics of images, we propose an application ontology derived from a generic ontology. Collaborative semantic annotation that we define, provides emergent semantics through the fusion of semantics proposed by the annotators.Semantic retrieval allows to look for images with semantics provided by collaborative semantic annotation. It is based on clustering and recommendation. Clustering is used to group similar images corresponding to the user’s query and recommendation aims to propose semantics to users based on their profiles. It consists of three steps : creation of users community, acquiring of user profiles and classification of user profiles with Galois algebra. Experiments were conducted to validate the approaches proposed in this work
24
Lima, Alexandre da Silva. "Form filling recommendation using machine learning techniques." Master's thesis, 2019. https://hdl.handle.net/10216/119419.
Full text
25
Chen, Yu-Hsuan, and 陳宇軒. "QA Document Recommendation Techniques for Knowledge Communities." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/56260803619245722369.
Full textAbstract:
博士國立交通大學
資訊管理研究所
102
With the emergence of Social Media, Social Question-Answering (SQA) websites have become common knowledge production and sharing platforms. This platform provides knowledge community services where users with common interests, needs or expertise can form a knowledge community. Community members can collect and share QA knowledge (documents) regarding their interests. However, due to the massive amount of QAs created every day, information overload can become a major problem. Consequently, a recommender system is needed to suggest QA documents for communities of SQA websites. In this thesis, we propose several novel methods, called GTPR-based approaches, to recommend related QA documents for knowledge communities of SQA sites. The proposed methods recommend QA documents by considering community-specific features, the relationships between knowledge documents, and documents’ relevance to the communities. In addition, due to the robustness problem of the existing topic grouping method using user-defined tags in Social Media, this study further propose a novel method, called GPTLR, incorporating the community’s latent topics of interest and collection weights based on members’ topical reputations to improve content-based recommendation models. This research evaluates and compares the proposed methods using a real-world dataset collected from a SQA website. Experimental results show that the proposed methods outperform other traditional methods, providing a more effective and accurate recommendations of Q&;A documents to knowledge communities.
26
Guo, X. "Personalized government online services with recommendation techniques." Thesis, 2006. http://hdl.handle.net/10453/37083.
Full textAbstract:
University of Technology, Sydney. Faculty of Information Technology.With the integration of information from different government agencies, a vast resource of information and services may be gathered in one portal. Many businesses have difficulty locating the required information and services. In such a situation of vast information overload, one of the difficulties facing governments is how to provide businesses with information specific to their needs, rather than an undifferentiated mass of information. One way to do this is through the development of personalized government online services. Indeed, the recent Accenture e-government study indicates that personalization techniques in e-government are beginning to emerge. However, existing personalization with recommendation techniques focuses on text document retrieval and e-commerce product recommendation domain. Personalization and recommendation applications in e-government have paid relatively little research attention. Many mechanisms have been developed to deliver only relevant information to web users and prevent information overload. The most popular recent developments in the e- commerce domain are the user-preference based personalization and recommendation techniques. The existing techniques have a major drawback: they are difficulty to generate recommendation on one-and-only items, because most of them do not understand the item’s semantic features and attributes. Therefore, this study aims to: (1) propose a novel approach, semantic product relevance model and its attendant personalized recommendation technique, to handle the one-and-only item recommendation problem; (2) develop a recommender system prototype, called Smart Trade Exhibition Finder, to tailor the relevant trade exhibition information to each particular business user, and to assist export business selecting the right trade exhibitions for market promotion. Smart Trade Exhibition Finder may reduce significantly the time, cost and risk faced by exporters in selecting, entering and developing international markets. In particular, the proposed approach can be used to overcome the drawback of existing recommendation techniques and enable recommender systems to work within a much wider range of problems which cannot currently be handled. The outcome of this study will solve the rating data lacking and new item problem, and significantly improve the performance compared to existing recommendation techniques.
27
Xu, Guandong. "Web mining techniques for recommendation and personalization." Thesis, 2008. https://vuir.vu.edu.au/1422/.
Full textAbstract:
Nowadays Web users are facing the problems of information overload and drowning due to the significant and rapid growth in the amount of information and the number of users. As a result, how to provide Web users with more exactly needed information is becoming a critical issue in web-based information retrieval and Web applications. In this work, we aim to address improving the performance of Web information retrieval and Web presentation through developing and employing Web data mining paradigms. Web data mining is a process that discovers the intrinsic relationships among Web data, which are expressed in the forms of textual, linkage or usage information, via analysing the features of the Web and web-based data using data mining techniques. Particularly, we concentrate on discovering Web usage pattern via Web usage mining, and then utilize the discovered usage knowledge for presenting Web users with more personalized Web contents, i.e. Web recommendation. For analysing Web user behaviour, we first establish a mathematical framework, called the usage data analysis model, to characterise the observed co-occurrence of Web log files. In this mathematical model, the relationships between Web users and pages are expressed by a matrix-based usage data schema. On the basis of this data model, we aim to devise algorithms to discover mutual associations between Web pages and user sessions hidden in the collected Web log data, and in turn, to use this kind of knowledge to uncover user access patterns. To reveal the underlying relationships among Web objects, such as Web pages or user sessions, and find the Web page categories and usage patterns from Web log files, we have proposed three kinds of latent semantic analytical techniques based on three statistical models, namely traditional Latent Semantic Indexing, Probabilistic Latent Semantic Analysis and Latent Dirichlet Allocation model. In comparison to conventional Web usage mining approaches, the main strengths of latent semantic based analysis are their capabilities that can not only, capture the mutual correlations hidden in the observed objects explicitly, but also reveal the unseen latent factors/tasks associated with the discovered knowledge implicitly. In the traditional Latent Semantic Indexing, a specific matrix operation, i.e. Singular Value Decomposition algorithm, is employed on the usage data to discover the Web user behaviour pattern over a transformed latent Web page space, which contains the maximum approximation of the original Web page space. Then, a k-means clustering algorithm is applied to the transformed usage data to partition user sessions. The discovered Web user session group is eventually treated as a user session aggregation, in which all users share like-minded access task or intention. The centroids of the discovered user session clusters are, then, constructed as user profiles. In addition to intuitive latent semantic analysis, Probabilistic Latent Semantic Analysis and Latent Dirichlet Allocation approaches are also introduced into Web usage mining for Web page grouping and usage profiling via a probability inference approach. Meanwhile, the latent task space is captured by interpreting the contents of prominent Web pages, which significantly contribute to the user access preference. In contrast to traditional latent semantic analysis, the latter two approaches are capable of not only revealing the underlying associations between Web pages and users, but also capturing the latent task space, which is corresponding to user navigational patterns and Web site functionality. Experiments are performed to discover user access patterns, reveal the latent task space and evaluate the proposed techniques in terms of quality of clustering. The discovered user profiles, which are represented by the centroids of the Web user session clusters, are then used to make usage-based collaborative recommendation via a top-N weighted scoring scheme algorithm. In this scheme, the generated user profiles are learned from usage data in an offline stage using above described methods, and are considered as a usage pattern knowledge base. When a new active user session is coming, a matching operation is carried out to find the most matched/closest usage pattern/user profile by measuring the similarity between the active user session and the learned user profiles. The user profile with the largest similarity is selected as the most matched usage profile, which reflects the most similar access interest to the active user session. Then, the pages in the most matched usage profile are ranked in a descending order by examining the normalized page weights, which are corresponding to how likely it is that the pages will be visited in near future. Finally, the top-N pages in the ranked list are recommended to the user as the recommendation pages that are very likely to be visited in the coming period. To evaluate the effectiveness and efficiency of the recommendation, experiments are conducted in terms of the proposed recommendation accuracy metric. The experimental results have demonstrated that the proposed latent semantic analysis models and related algorithms are able to efficiently extract needed usage knowledge and to accurately make Web recommendations. Data mining techniques have been widely used in many other domains recently due to the powerful capability of non-linear learning from a wide range of data sources. In this study, we also extend the proposed methodologies and technologies to a biomechanical data mining application, namely gait pattern mining. Likewise in the context of Web mining, various clustering-based learning approaches are performed on the constructed gait variable data model, which is expressed as a feature vector of kinematic variables, to discover the subject gait classes. The centroids of the partitioned gait clusters are used to represent different specific walking characteristics. The data analysis on two gait datasets corresponding to various specific populations is carried out to demonstrate the feasibility and applicability of gait pattern mining. The results have shown the discovered gait pattern knowledge can be used as a useful means for human movement research and clinical applications.
28
Xu, Guandong. "Web mining techniques for recommendation and personalization." 2008. http://eprints.vu.edu.au/1422/1/xu.pdf.
Full textAbstract:
Nowadays Web users are facing the problems of information overload and drowning due to the significant and rapid growth in the amount of information and the number of users. As a result, how to provide Web users with more exactly needed information is becoming a critical issue in web-based information retrieval and Web applications. In this work, we aim to address improving the performance of Web information retrieval and Web presentation through developing and employing Web data mining paradigms. Web data mining is a process that discovers the intrinsic relationships among Web data, which are expressed in the forms of textual, linkage or usage information, via analysing the features of the Web and web-based data using data mining techniques. Particularly, we concentrate on discovering Web usage pattern via Web usage mining, and then utilize the discovered usage knowledge for presenting Web users with more personalized Web contents, i.e. Web recommendation. For analysing Web user behaviour, we first establish a mathematical framework, called the usage data analysis model, to characterise the observed co-occurrence of Web log files. In this mathematical model, the relationships between Web users and pages are expressed by a matrix-based usage data schema. On the basis of this data model, we aim to devise algorithms to discover mutual associations between Web pages and user sessions hidden in the collected Web log data, and in turn, to use this kind of knowledge to uncover user access patterns. To reveal the underlying relationships among Web objects, such as Web pages or user sessions, and find the Web page categories and usage patterns from Web log files, we have proposed three kinds of latent semantic analytical techniques based on three statistical models, namely traditional Latent Semantic Indexing, Probabilistic Latent Semantic Analysis and Latent Dirichlet Allocation model. In comparison to conventional Web usage mining approaches, the main strengths of latent semantic based analysis are their capabilities that can not only, capture the mutual correlations hidden in the observed objects explicitly, but also reveal the unseen latent factors/tasks associated with the discovered knowledge implicitly. In the traditional Latent Semantic Indexing, a specific matrix operation, i.e. Singular Value Decomposition algorithm, is employed on the usage data to discover the Web user behaviour pattern over a transformed latent Web page space, which contains the maximum approximation of the original Web page space. Then, a k-means clustering algorithm is applied to the transformed usage data to partition user sessions. The discovered Web user session group is eventually treated as a user session aggregation, in which all users share like-minded access task or intention. The centroids of the discovered user session clusters are, then, constructed as user profiles. In addition to intuitive latent semantic analysis, Probabilistic Latent Semantic Analysis and Latent Dirichlet Allocation approaches are also introduced into Web usage mining for Web page grouping and usage profiling via a probability inference approach. Meanwhile, the latent task space is captured by interpreting the contents of prominent Web pages, which significantly contribute to the user access preference. In contrast to traditional latent semantic analysis, the latter two approaches are capable of not only revealing the underlying associations between Web pages and users, but also capturing the latent task space, which is corresponding to user navigational patterns and Web site functionality. Experiments are performed to discover user access patterns, reveal the latent task space and evaluate the proposed techniques in terms of quality of clustering. The discovered user profiles, which are represented by the centroids of the Web user session clusters, are then used to make usage-based collaborative recommendation via a top-N weighted scoring scheme algorithm. In this scheme, the generated user profiles are learned from usage data in an offline stage using above described methods, and are considered as a usage pattern knowledge base. When a new active user session is coming, a matching operation is carried out to find the most matched/closest usage pattern/user profile by measuring the similarity between the active user session and the learned user profiles. The user profile with the largest similarity is selected as the most matched usage profile, which reflects the most similar access interest to the active user session. Then, the pages in the most matched usage profile are ranked in a descending order by examining the normalized page weights, which are corresponding to how likely it is that the pages will be visited in near future. Finally, the top-N pages in the ranked list are recommended to the user as the recommendation pages that are very likely to be visited in the coming period. To evaluate the effectiveness and efficiency of the recommendation, experiments are conducted in terms of the proposed recommendation accuracy metric. The experimental results have demonstrated that the proposed latent semantic analysis models and related algorithms are able to efficiently extract needed usage knowledge and to accurately make Web recommendations. Data mining techniques have been widely used in many other domains recently due to the powerful capability of non-linear learning from a wide range of data sources. In this study, we also extend the proposed methodologies and technologies to a biomechanical data mining application, namely gait pattern mining. Likewise in the context of Web mining, various clustering-based learning approaches are performed on the constructed gait variable data model, which is expressed as a feature vector of kinematic variables, to discover the subject gait classes. The centroids of the partitioned gait clusters are used to represent different specific walking characteristics. The data analysis on two gait datasets corresponding to various specific populations is carried out to demonstrate the feasibility and applicability of gait pattern mining. The results have shown the discovered gait pattern knowledge can be used as a useful means for human movement research and clinical applications.
29
LEE, YEU TUNG, and 李雨瞳. "Personalized Video Recommendation by Using Data Mining Techniques." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/92127895633566878321.
Full textAbstract:
碩士大葉大學
工業工程與科技管理學系碩士在職專班
95
Due to film piracy and digital television, the visual-audio rental industrial are in keen competitions. Nowadays, most of studies focus on images and customer satisfaction of visual-audio rental store, and only a few studies discuss customer preferences and consumer behavior. The purpose of this study is to realize the association between the customer preferences and the consumer rental behavior, by means of improving profits of visual-audio rental industrial. In this study, we collect data (customers’ personal informations and their consuming records) by using questionnaires and continue to analyze the primary data in terms of data mining algorithm from association rule and the CHAID Decision Tree. This study discovers a correlation between individualized interests and the categories of rented videos, and applies recommender of content-base in the process of seeking the best matches between individualized interests and videos. This result of study has found eleven rules. For example, the consumers of loving travel will choose the comedy film and the science fiction film at the same time; the consumers being fond of sports will choose the crime, comedy and action films at the same time. So the result can be offered to the industries, the business strategy in terms of the film combination to expand the business opportunity. Consequently, how to make a good marketing strategy that let consumers rent their favorite films is the direction that the visual-audio sales have to try hard.
30
Chen, Hung-Chen, and 陳宏鎮. "Techniques of Music Analysis, Recommendation, and Retrieval for Music Services." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/48586024338205626628.
Full textAbstract:
博士國立清華大學
資訊工程學系
96
With the growth of Internet, a large amount of music data is available for many music-related applications. It is almost impractical for these applications to satisfy the user requirements manually. To provide efficient services within these applications, the techniques developed for automatic music analysis, recommendation, and retrieval are urgently necessary. In this paper, we consider the applications of interactive music tutorials and distance education at music school. In the two applications, we need to integrate several techniques to achieve the educational purposes. This demand motives us to develop the advanced techniques for the performances of music services. In the area of music analysis, the music structure usually needs to be analyzed manually by experts, which is time-consuming and impractical. Therefore, we propose an approach for automatic music segmentation to extract the phrases and sentences of the musical structure. In addition to the rhythmic features, the melodic shape is first-ever used to improve the effectiveness of the music segmentation. Concerning a large number of music objects available in the databases, the systems that provide the services for users to look for their favorite music objects are urgently needed. One of the most important services for the users to escape from this information-overloading problem is the recommendation service. Due to the complex semantics of the music objects and the difficult derivation of user interests and behaviors, we propose an alternative way of music recommendation, which overcomes the limitations of the previous works. The music objects are first grouped based on the automatically extracted features. Moreover, the user access histories are analyzed to derive the profiles of user interests and behaviors for user grouping. The content-based, collaborative, and statistics-based recommendation methods are proposed based on the favorite degrees of the users to the music groups, and the user groups they belong to. Many interesting applications based on music streams, such as interactive music tutorials, distance music education, and similar theme searching, make the research of content-based retrieval over music streams much important. Therefore, we consider multiple queries with error tolerances over music streams and address the issue of approximate matching in this environment. To satisfy this demand, we propose a novel approach to continuously process multiple queries over the music streams for finding all the music segments that are similar to the queries. Our approach is based on the concept of n-grams and two mechanisms are designed to reduce the heavy computation of approximate matching. One mechanism uses the clustering of query n-grams to prune the query n-grams that are irrelevant to the incoming data n-gram. The other mechanism records the data n-gram that matches a query n-gram as a partial answer and incrementally merges the partial answers of the same query. A series of experiments are performed to demonstrate the effectiveness and efficiency of our approaches by comparing with other related works.
31
Ja-HwungSu and 蘇家輝. "Multimedia Data Mining Techniques for Semantic Annotation, Retrieval and Recommendation." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/05323447331505634288.
Full textAbstract:
博士國立成功大學
資訊工程學系碩博士班
98
In recent years, the advance of digital capturing technologies lead to the rapid growth of multimedia data in various formats, such as image, music, video and so on. Moreover, the modern telecommunication systems make multimedia data widespread and extremely large. Hence, how to conceptualize, retrieve and recommend the multimedia data from such massive multimedia data resources has been becoming an attractive and challenging issue over the past few years. To deal with this issue, the primary aim of this dissertation is to develop effective multimedia data mining techniques for discovering the valuable knowledge from multimedia data, so as to achieve the high quality of multimedia annotation, retrieval and recommendation. Nowadays, a considerable number of studies in the field of multimedia annotations incur the difficulties of diverse relationships between human concepts and visual contents, namely diverse visual-concept associations. So-called visual-concept diversity indicates that, a set of different concepts share with very similar visual features. To alleviate the problems of diverse visual-concept associations, this dissertation presents the integrated mining of visual, speech and text features for semantic image/video annotation. For image annotation, we propose a visual-based annotation method to disambiguate the image sense while a number of senses are shared by a number of images. Additionally, a textual-based annotation method, which attempts to discover the affinities of image captions and web-page keywords, is also proposed to attack the lack of visual-based annotations. For video annotation, with considering the temporal continuity, the frequent visual, textual and visual-textual patterns can be mined to support semantic video annotation by proposed video annotation models. Based on the image annotation, the user’s interest and visual images can be bridged semantically for further textual-based image retrieval. However, little work has highlighted the conceptual retrieval from textual annotations to visual images in the last few years. To this end, the second intention in this dissertation is to retrieve the images by proposed image annotation, concept matching and fuzzy ranking techniques. In addition to textual-based image retrieval, the textual-based video retrieval cannot earn the user’s satisfaction either due to the problems of diverse query concepts. To supplement the weakness of textual-based video retrieval, we propose an innovative method to mine the temporal patterns from the video contents for supporting content-based video retrieval. On the basis of discovered temporal visual patterns, an efficient indexing technique and an effective sequence matching technique are integrated to reduce the computation cost and to raise the retrieval accuracy, respectively. In contrast to passive image/video retrieval, music recommendation is the final concentration in this dissertation to actively provide the users with the preferred music pieces. In this work, we design a novel music recommender that integrates music content mining and collaborative filtering to help the users find what she/he prefers from a huge amount of music collections. By discovering preferable perceptual-patterns from music pieces, the user’s listening interest and music can be associated effectively. Also the traditional rating diversity problem can be alleviated. For each proposed approach above, the experimental results in this dissertation reveal that, our proposed multimedia data mining methods are beneficial for better multimedia annotation, retrieval and recommendation so as to apply to some real multimedia applications, such as mobile multimedia retrieval and recommendation.
32
Chen, JIN-WEI, and 陳勁瑋. "Preference Analysis and Articles Recommendation based on Text Mining Techniques." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/yk8b2n.
Full textAbstract:
碩士國立雲林科技大學
資訊管理系
106
IN THE INFORMATION AGE, THERE ARE VARIOUS TYPES OF NEW ARTICLES POSTED ON THE INTERNET EVERY DAY. THE NUMBER OF ARTICLES GROWS AT A RAPID SPEED SO THAT READERS HARDLY KEEP TRACK OF THE ARTICLES THEY MIGHT BE INTERESTED. IN THIS PAPER, WE PROPOSED AN APPROACH TO IDENTIFY READERS’ PREFERENCE AND THEN RECOMMEND NEW COMING ARTICLES ACCORDINGLY. SPECIFICALLY, WE USE INFORMATION RETRIEVAL TECHNIQUES TO CALCULATE TF-IDF OF WORDS IN THE ARTICLES, SELECT REPRESENTATIVE WORDS AS KEYWORDS, AND THEN BUILD A KEYWORDS DICTIONARY. READER’S PREFERENCE IS CONSTRUCTED AS A KEYWORD VECTOR BASED ON THE ARTICLES APPEARING IN HIS READING LOG. FOR NEW ARTICLES, TOPICS OF THE ARTICLES ARE REPRESENTED BY CORRESPONDING KEYWORD VECTORS AND THE DEGREE OF MATCH BETWEEN READER’S PREFERENCE AND THE TOPICS OF THE ARTICLES IS MEASURED BY COSINE SIMILARITY WHICH THE RECOMMENDATION IS BASED ON. EXPERIMENTS ARE DIVIDED INTO TWO PARTS. THE FIRST USES THE STATISTICAL INDICATOR F_1 "-" MEASURE TO COMPARE THE RESULTS OF DIFFERENT DECAY PARAMETERS. THE SECOND SELECTS A REPRESENTATIVE USER, ANALYZES THE NUMBER OF ARTICLES READ IN EACH MONTH, AND PRESENTS THE RECOMMENDATION RESULTS ACCORDING TO VARIOUS DECAY PARAMETERS.
33
陳銘憲. "On Query Recommendation Techniques for the Criminal Case Knowledge Base." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/fr66p3.
Full textAbstract:
碩士中央警察大學
資訊管理研究所
106
In order to enhance the effectiveness of criminal investigation, the Criminal Investigation Bureau of the National Police Agency under the Ministry of the Interior has built a Criminal Case Knowledge Base to meet the needs and characteristics of criminal investigators. The Criminal Case Knowledge Base collects relevant data on suspects, such as their identification card numbers, names, license plate numbers, offenses charged and case numbers. So far, the Criminal Case Knowledge Base has become one of the most important tools for supporting and assisting criminal investigators. In addition, in response to the requirements of internal management and statistical analysis, the Criminal Case Knowledge Base also records a user’s searching trajectory, such as user code, case number, searching time, items searched (also called search-items) and searching results. This research explores and analyzes the data on searching trajectories in the Criminal Case Knowledge Base. The relationships among investigative searching trajectories, which might implicitly contain the investigative intelligence of criminal investigators, are extracted by using data mining techniques. This information is then used to develop the recommendation algorithms for the Criminal Case Knowledge Base. This research developed three recommended algorithms for the Criminal Case Knowledge Base including an item-based recommendation algorithm, a co-occurrence nearest neighbor-based recommendation algorithm, and an item attribute-based recommendation algorithm. The item-based recommendation algorithm uses a search-item in the test dataset to find matched search-item(s) in the training dataset. The search-item(s) following the matched search-item(s) are then used as the recommendation items for the search-item (in the test dataset). In addition to the recommendation items generated by the item-based recommendation algorithm, the co-occurrence nearest neighbor-based recommendation algorithm includes the co-occurrence nearest neighbors of the recommendation items in the recommendation item set. Although the item attribute-based recommendation algorithm is also based on the item-based recommendation algorithm, it recommends the search direction(s) for the search-item instead. This research collected searching trajectories of 4,793 criminal cases between 2013 and 2016 from the Criminal Case Knowledge Base of the New Taipei City Police Department as its experiment dataset. Seventy percent of the experiment dataset was randomly selected and used as the training dataset, while the other thirty percent was used as the test dataset. The average value of 10 experiments was used as the experiment result. The experiment results show that the average successful recommendation rate of the item-based recommendation algorithm is 4.67%, while the average successful recommendation rate of random recommendation is only 0.02%. Compared with the latter, the successful recommendation rate of the former increases approximately 233 folds. In addition, the experiment results suggest that the average successful recommendation rate of the co-occurrence nearest neighbor-based recommendation algorithm is 5.13%, which is 9.85% growth from the item-based recommendation algorithm. Moreover, the average successful recommendation rate of the item attribute-based recommendation algorithm is 24.58%, while the average successful recommendation rate of random recommendation is only 22.41%. Compared with the latter, the successful rate of the former increases by 9.68%. In summary, the recommendation algorithms proposed in this research can be used to improve the efficiency of the investigative searching in the Criminal Case Knowledge Base. Keywords: Data Mining, Investigative Intelligence, Recommendation Techniques, Criminal Case Knowledge Base, Searching Trajectory.
34
Yeh, Hsin-Ho, and 葉信和. "Intelligent Music Recommendation Techniques by Mining Context Information and Musical Contents." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/24420779600126900281.
Full textAbstract:
碩士國立成功大學
資訊工程學系碩博士班
97
Advanced networking and telecommunication technologies make the ease of mobile information retrieval. However, it is hard for the users to find what she/he prefers under large music databases and variant context conditions. To solve this problem, some previous music recommenders have been proposed for providing preferable music automatically. Unfortunately, current music recommenders only based on user’s rating log cannot earn the user’s satisfaction in finding the preferred music due to the lacked consideration of context information and sufficient rating data. To solve above problems, in this thesis, we propose a novel Music Recommendation technique by integrating Context information and musical Contents (called MRCC). For context information, the users in similar context conditions are grouped to mine the similar preference patterns. For musical contents, we propose two-stage clustering to convert musical contents into perceptual patterns with considering acoustical and temporal features simultaneously. Finally, the user’s preference can be accurately predicted by integrated mining of context information and musical contents. Experimental results show that our proposed MRCC can capture the user’s preference effectively and outperform existing collaborative filtering approaches in terms of accuracy on context-aware recommendation.
35
Li, Pei-shan, and 李佩珊. "The Recommendation of Elective Courses Using Data Mining and Collaborative Filtering Techniques." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/98121240770913385308.
Full textAbstract:
碩士雲林科技大學
資訊管理系碩士班
96
In the field of education, students play the role of customer. How to actively offer information and assist them in choosing suitable elective courses are agonizing problems in many universities. Although various recommender systems have been proposed to solve these problems, many studies focus on the arrangement of the learning path between different chapters in the same course, such as e-learning, but only few studies for recommending different courses are proposed. In this study, we proposed a methodology to provide personalized recommendations on elective courses based on Data Mining and Collaborative Filtering techniques. The research objects are the 4-year and 2-year technological program students in the Department of Information Management in a National University of Science and Technology. Firstly, students were classified into two groups: freshmen and seniors. Secondly, for freshmen, we used the Collaborative Filtering approach to find the top-N recommendation courses of each cluster of students. For seniors, according to students’ enrolling records and scores, we used the Sequential Patten approach to find the preferred sequential courses and then used the Association Rule approach to find the related courses based on the recommendation results of the Sequential Patten approach. Finally, we used the F1-measure to measures the quality of our recommendation results.
36
Li, Kun-Hung, and 李坤宏. "An Evaluation Study of Applying Data Mining Techniques in Developing Personalized Recommendation." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/58534521164950627360.
Full textAbstract:
碩士國立中央大學
企業管理研究所
91
It is recognized that customer relationship management (CRM) is the key point to benefit business, and the first task is to segment customers for providing customized products and services. However, the ways of market segmentation are of variety, and how to design the proper way to separate customers is more significant relatively. Also, recommender systems are used to suggest products to their customers and to provide consumers with information to help them purchase. If the recommendation is specifically designed for individuals, it will be more suitable and concise. The aims of our study are to suggest a method of personalized product recommendations to recommend appropriate products at appropriate time, and to evaluate the difference of recommendations based on demographic and behavioral segmentations. We employed two techniques of data mining: IBM Intelligent Miner to cluster the customers, and I-PrefixSpan algorithm to discover time-interval sequential patterns in every cluster. Results indicated the recommendation based on behavioral segmentation is more accurate than that based on demographic segmentation.
37
Bo-WenWang and 王博文. "Data Pre-Processing and Data Mining Techniques for Video Retrieval and Recommendation." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/24023618016292954357.
Full text
38
Huang, Jerry, and 黃俊榮. "Using Clustering Techniques to Discover the Recommendation Services of Borrowing Books for Libraries." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/83041893230825617362.
Full textAbstract:
碩士南台科技大學
資訊管理系
94
The fast development in economy of cause in recent years, the collected books amount of the library is increasing fast too, find the books which accord with reader's demand in these a large amount of collected books, it is more and more difficult. So, how can offer the information of the collected books according to reader's need, have already become an important subject of the library at present. On the other hand, the library hopes to offer more information and service to readers too, hope that the collection in the library can try one's best to be utilized effectively , in order to enable reader to obtain benefit most from the book borrowed. In this thesis, we according to borrowing database of library of Southern Taiwan University of Technology offered, and regard borrowing the data of readers as the data source mined, every one borrows data includes books and one degree of value of interest that readers once borrowed, look for adaptive books for readers,and look for adaptive readers for the books. Utilize clustering of the data mining approach come the books or readers, through calculation of the borrowing similar degree, look for adaptive reader or books of group and recommend it . Finally, according to the method put forward, we design and build a recommend system of adaptive books and reader. The result of mining, while planning reader's personalised service to the library, can offer very useful reference information .
39
Mota, Maria Dulce Fernandes. "A model for teaching-learning techniques recommendation to support teaching-learning activities design." Doctoral thesis, 2018. https://repositorio-aberto.up.pt/handle/10216/113819.
Full text
40
Mota, Maria Dulce Fernandes. "A model for teaching-learning techniques recommendation to support teaching-learning activities design." Tese, 2018. https://repositorio-aberto.up.pt/handle/10216/113819.
Full text
41
Tu, Yi-Ning, and 杜逸寧. "Integrating CBR and Clustering techniques in Collaborative Classification-An application of journal papers recommendation system." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/59064634372046793848.
Full textAbstract:
碩士國立彰化師範大學
資訊管理學系所
93
This paper tries to apply the clustering approach and collaborative recommendation for using the present experience to make the new users in indexing or searching papers fast and correctly. Besides the user’s experience is very important for the recommendation system, because the past cases could help the new user more quickly to index the paper. We suggest the CBR(Case-Based Reasoning) method for the system remaining the past indexed cases to help new user to index the paper based on the cases which had grouped by the past uses straightly. This research applied a system for the user to training their searching behavior and record the information for recommendation of the new user to index the paper. We used the information and applied two kinds of the collaborative recommendation. One of the methods is the keyword collaborative recommendation and the other is the CBR collaborative recommendation. We find out that CBR approach is better than keyword’s since it tries to find out the most similar person to make the recommendation. But it also has the high risk opposite to the keyword approach. Since the keyword approach use the users who have the same keyword would suggest the new users together, it could have the stable precision and the lower risk to get the worse recommendation. Key words: Collaborative recommendation, Clustering, Case-based Reasoning.
42
Kolbert, András Péter. "A scalable recommender system : using latent topics and alternating least squares techniques." Master's thesis, 2018. http://hdl.handle.net/10362/34383.
Full textAbstract:
Dissertation presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced AnalyticsA recommender system is one of the major techniques that handles information overload problem of Information Retrieval. Improves access and proactively recommends relevant information to each user, based on preferences and objectives. During the implementation and planning phases, designers have to cope with several issues and challenges that need proper attention. This thesis aims to show the issues and challenges in developing high-quality recommender systems. A paper solves a current research problem in the field of job recommendations using a distributed algorithmic framework built on top of Spark for parallel computation which allows the algorithm to scale linearly with the growing number of users. The final solution consists of two different recommenders which could be utilised for different purposes. The first method is mainly driven by latent topics among users, meanwhile the second technique utilises a latent factor algorithm that directly addresses the preference-confidence paradigm.
43
Wu, Tzer-Min, and 吳澤民. "A study of applying data mining techniques to maintenance knowledge recommendation -Aircraft maintenance database as example." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/75862413932268429462.
Full textAbstract:
碩士中原大學
資訊管理研究所
90
Due to the development of information techniques in recent years. Lots of companies have setup information systems to collect the daily operation data and stored these data in the database. In order to enforce the competitive advantage for company. They want to retrieve some useful information and knowledge from data. The maintenance engineering arises from professional field and experience accumulated. Since the development of information techniques, we use information systems to store the knowledge and experience of people in the database. If we can’t transfer these data to useful knowledge for people, then we can’t transfer knowledge to others from these data. The researches of data mining techniques have been developed very well in many fields. The application of data mining techniques in maintenance field being in paid much attention in these years. The purpose of this study is to apply data mining techniques to maintenance knowledge retrieval and to construct a maintenance knowledge recommendation system. This system can help maintenance engineer to finish their work more quickly. This study applying traditional memory based reasoning technique (RMBR) to obtain a preliminary result and analysis the data character. Then we propose a class-oriented memory based reasoning technique (CMBR) to improve the accuracy and performance of classification. The experimental results of CMBR show a higher accuracy than traditional MBR (RMBR). This study applying CMBR to construct a classified model and use this model to construct a maintenance knowledge recommendation system. This recommendation system can help maintenance engineer to find technical document and solution to complete his work. Because line maintenance must be completed in the shorten time. Helping by the maintenance knowledge recommendation system, then the maintenance engineer will do job well and quickly. In addition, he can acquire the experience from others.
44
Hsu, Pei-Ying, and 許珮瑩. "A Novel Explainable Mutual Fund Recommendation System Based on Deep Learning Techniques with Knowledge Graph Embeddings." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/wur49w.
Full textAbstract:
碩士國立交通大學
資訊管理研究所
107
Since deep learning based models have gained success in various fields during recent years, many recommendation systems also start to take advantage of the deep learning techniques. However, while the deep learning based recommendation systems have achieved high recommendation performance, the lack of interpretability may reduce users' trust and satisfaction, while limiting the model to wide adoption in the real world. As a result, to strike a balance between high accuracy and interpretability, or even obtain both of them at the same time, has become a popular issue among the researches of recommendation systems. In this thesis, we would like to predict and recommend the funds that would be purchased by the customers in the next month, while providing explanations simultaneously. To achieve the goal, we leverage the structure of knowledge graph, and take advantage of deep learning techniques to embed customers and funds features to a unified latent space. We fully utilize the structure knowledge which cannot be learned by the traditional deep learning models, and get the personalized recommendations and explanations. Moreover, we extend the explanations to more complex ones by changing the training procedure of the model, and proposed a measure to rate for the customized explanations while considering strength and uniqueness of the explanations at the same time. Finally, we regard that the knowledge graph based structure could be extended to other applications, and proposed some possible special recommendations accordingly. By evaluating on the dataset of mutual fund transaction records, we verify the effectiveness of our model to provide precise recommendations, and also evaluate the assumptions that our model could utilize the structure knowledge well. Last but not least, we conduct some case study of explanations to demonstrate the effectiveness of our model to provide usual explanations, complex explanations, and other special recommendations.
45
Tseng, Hsin Hui, and 曾馨慧. "Applying Data Mining Techniques and Hopfield Neural Network to Recommendation Application（Take Waste Application System as the Example）." Thesis, 2004. http://ndltd.ncl.edu.tw/handle/65293837250677845795.
Full textAbstract:
碩士中華大學
資訊管理學系
92
In recent years,development of the World Wide Web made fast progress.Information unit in enterprises take development of the Web as the main target. Due to the advantages of Web,IT of departments of various software companies have transfered their application systems to be operated in web interface. The main goal is to achieve an interface that can be operated with simple steps. That is,operations with complex steps should be avoided as much as passible. The main goal of this research is to achieve an application system that can be operated with simple steps by exploring the technologies of data maining and neural network.We take 「Internet Waste System」 as an example.We apply the two technologies mentioned above to create an interface that can be used as a modle for developing a good interactive interface to provide a good interaction pattern.By using the interface,efficient operation of an information system may reduce operation overload degree of the system. Keyword：Data Mining、Neural Network、Information Overload.
46
Chiu, Yung-Hsiang, and 邱永祥. "The Study of Applying Neural Network and Data Mining Techniques to Course Recommendation Base on E-learning Environment." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/82886359874632835710.
Full textAbstract:
碩士朝陽科技大學
資訊管理系碩士班
91
As for the various applications of Internet, E-learning changes the traditional learning style and elaborates the characteristics of Internet, such as speed, convenience, the ignorance of distance, the enhancement of teaching quality, and the reduced cost. There are many studies has been done regarding E-learning, however, these studies focus more on system unstruction and in curriculeum organization. These does not help the E-learner to select potential courses according to his or her interest. It is, therefore, the goal of this research to propose a better recommendation model for on-line E-learners. This research first utilizes artificial neural network to find out the clusters of E-learners. Based on these E-learner’s groups, user can obtain course recommendation from the group’s opinion. When groups of related interests have been established, the Rough Set and the Apriori algorithm will be used to find out the rules of “ learner groups vs. course ” and the “course vs. course” Through this recommending system, an inclusive curriculum may be suggested to a group of learners with same interest. It is ideal for this system to stimulate learners’ motivation and interest, moreover, to serve as a reference when learners are choosing between classes. The course recommendation base on E-learning environment presented in this research can be categorized into two stages: (1)preprocessing stage of E-learning courses and (2)on-line courses recommending stage. This research will also demonstrate a live example of a financial organization using this curriculum-recommending system. The demonstration, indeed, will confirm that this research is feasible and practicable. The advantages of the presented method are (1)providing community features, (2)supporting relative features, and (3)performing with on-line efficiency.
47
Cheng, Yu-Ling, and 鄭玉玲. "Apply Data Mining Techniques to Implement Personalized Retrieval and Recommendation on Digital Library － the Library of NHU for example." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/07289752881735063109.
Full textAbstract:
碩士南華大學
資訊管理學研究所
91
Due to the rapid advance of computer-related technology, the large quantities of electronic data produced everyday has been explored and analyzed to discover meaningful information so as to enhance the service quality of digital library. The objective of this study is to enable personalized retrieval and recommendation services on digital library. We utilize the well-defined “New Classification Scheme For Chinese Libraries” to support audience to retrieve the books that they really want. Besides, memory-based reasoning was applied to assign the unlabeled user to the cluster of its nearest labeled neighbors based on some predefined measures of users’ characteristics. Association rules discovered from the books borrowed by the readers in the same cluster are used as the basis of book recommendation. A simple clustering algorithm was exploited to speed up the processing time of recommendation. The experimental results show that the proposed approaches are effective in promoting the searching efficiency and accuracy of the system.
48
Li, Tzu-Chi, and 黎子頎. "A Case Study of Using Data Mining Techniques in Service Satisfaction and Customer Recommendation - Use Brand’s Health Museum as An Example." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/htnm5j.
Full textAbstract:
碩士國立彰化師範大學
企業管理學系
101
This study used the 2011 internal customers’ questionnaires from Brand’s Health Museum, and used IBM SPSS Modeler 15.0 for analyses by considering seven satisfaction items as input variables and customer recommendation as a target variable. First , we used classification and regression tree (CART) to classify and predict the customers’ behaviors. Second, dimension reduction and feature selection were individually performed to identify the variables, and these variables become the input variables for CART. Third, Bayesian network (BN) was applied to repeat the prior processes. Besides, we subjectively identify the critical variables from BN graph, and these variables become the input variables for BN. Finally, we evaluated the performance among CART and BN models. In the original CART model, tree depth was five, and six rules were generated. Nevertheless, the CART model with dimension reduction could not generate the tree and any rule, while the CART model with feature selection found all input variables as critical variables. By using the same input variables as the original CART model did, the results of using CART model with feature selection and original CART model were the same. As for BN, only the TAN (Tree Augmented Naive Bayes) structure in all models could generate BN graph. Hence, we computed the probabilities of customer recommendation while consumer satisfaction is given. In model evaluation, the performances of BN models are better than those of CART models, and the best model is BN model with feature selection. In the end, we suggested managerial implications for each model.
49
Indriana, Marcelli, and 麥瑟莉. "Applying Data Mining Techniques for Tourist Spot Recommendations." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/44860577782092342084.
Full textAbstract:
碩士中國文化大學
資訊管理學系
102
Recommender systems have become an important research area in past few years. They have been developed for a variety of domains, especially e-commerce. Recommender systems also can be applied in tourism industry to help tourists organizing their travel plans. Recommender systems can be developed by a variety of different techniques such as Content-Based filtering (CB), Collaborative Filtering (CF), and Demographic Filtering (DF). However, each method has its own advantages and disadvantages. For this reason, many previous researches used several mixed methods with an aim to reduce the disadvantages of using a single method and get more accurate recommendations. In this research, we proposed a hybrid recommender system that combines the results of different recommendation methods using data mining techniques. Data mining technique is a method to dig out hidden knowledge and rules among the various items from large number of information and establish the relationship between model data attributes and categories in order to get more effective relationship model predictions. The experimental results showed that the proposed hybrid recommendation method outperforms each individual recommendation method in terms of five evaluation metrics.
50
Hung, Shu-Feng, and 洪淑芬. "A Collaborative Filtering Recommendation Technique Based on Overlapped Clusters." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/51684312031537488659.
Full text