Dissertations / Theses: 'Intelligent recommendation system'

1

Thiengburanathum, Pree. "An intelligent destination recommendation system for tourists." Thesis, Bournemouth University, 2018. http://eprints.bournemouth.ac.uk/30571/.

Full text

Abstract:

Choosing a tourist destination from the information available is one of the most complex tasks for tourists when making travel plans, both before and during their travel. With the development of a recommendation system, tourists can select, compare and make decisions almost instantly. This involves the construction of decision models, the ability to predict user preferences, and interpretation of the results. This research aims to develop a Destination Recommendation System (DRS) focusing on the study of machine-learning techniques to improve both technical and practical aspects in DRS. First, to design an effective DRS, an intensive literature review was carried out on published studies of recommendation systems in the tourism domain. Second, the thesis proposes a model-based DRS, involving a two-step filtering feature selection method to remove irrelevant and redundant features and a Decision Tree (DT) classifier to offer interpretability, transparency and efficiency to tourists when they make decisions. To support high scalability, the system is evaluated with a huge body of real-world data collected from a case-study city. Destination choice models were developed and evaluated. Experimental results show that our proposed model-based DRS achieves good performance and can provide personalised recommendations with regard to tourist destinations that are satisfactory to intended users of the system. Third, the thesis proposes an ensemble-based DRS using weight hybrid and cascade hybrid. Three classification algorithms, DT, Support Vector Machines (SVMs) and Multi- Layer Perceptrons (MLPs), were investigated. Experimental results show that the bagging ensemble of MLP classifiers achieved promising results, outperforming baseline learners and other combiners. Lastly, the thesis also proposes an Adaptive, Responsive, Interactive Model-based User Interface (ARIM-UI) for DRS that allows tourists to interact with the recommended results easily. The proposed interface provides adaptive, informative and responsive information to tourists and improves the level of the user experience of the proposed system.

APA, Harvard, Vancouver, ISO, and other styles

2

Xu, Shuting. "Study and Design of an Intelligent Preconditioner Recommendation System." UKnowledge, 2005. http://uknowledge.uky.edu/gradschool_diss/327.

Full text

Abstract:

There are many scientific applications in which there is a need to solve very large linear systems. The preconditioned Krylove subspace methods are considered the preferred methods in this field. The preconditioners employed in the preconditioned iterative solvers usually determine the overall convergence rate. However, choosing a good preconditioner for a specific sparse linear system arising from a particular application is the combination of art and science, and presents a formidable challenge for many design engineers and application scientists who do not have much knowledge of preconditioned iterative methods. We tackled the problem of choosing suitable preconditioners for particular applications from a nontraditional point of view. We used the techniques and ideas in knowledge discovery and data mining to extract useful information and special features from unstructured sparse matrices and analyze the relationship between these features and the solving status of the spearse linear systems generated from these sparse matrices. We have designed an Intelligent Preconditioner Recommendation System, which can provide advice on choosing a high performance preconditioner as well as suitable parameters for a given sparse linear system. This work opened a new research direction for a very important topic in large scale high performance scientific computing. The performance of the various data mining algorithms applied in the recommendation system is directly related to the set of matrix features used in the system. We have extracted more than 60 features to represent a sparse matrix. We have proposed to use data mining techniques to predict some expensive matrix features like the condition number. We have also proposed to use the combination of the clustering and classification methods to predict the solving status of a sparse linear system. For the preconditioners with multiple parameters, we may predict the possible combinations of the values of the parameters with which a given sparse linear system may be successfully solved. Furthermore, we have proposed an algorithm to find out which preconditioners work best for a certain sparse linear system with what parameters.

APA, Harvard, Vancouver, ISO, and other styles

3

Zhang, Junjie. "Development of a consumer-oriented intelligent garment recommendation system." Thesis, Lille 1, 2017. http://www.theses.fr/2017LIL10026/document.

Full text

Abstract:

Maintenant, l’achat de vêtements sur l’Internet est devenu une tendance importante pour les consommateurs du monde entier. Pourtant, dans les différents systèmes de vente en ligne, il manque systématiquement de recommandations personnalisées, comme celles fournies par les vendeurs d’une boutique physique, afin de proposer les produits les mieux adaptés à des différents consommateurs selon leurs morphotypes et leurs attentes émotionnelles. Dans cette thèse doctorale, nous proposons un système de recommandation orienté vers les consommateurs, qui peut être utilisé, comme un vendeur virtuel, à l’intérieur d’un système de vente de vêtements en ligne. Ce système a été développé par intégration de connaissance professionnelle des créateurs et des vendeurs et la perception des consommateurs sur les produits. En s’appuyant sur la connaissance de vente de vêtements, ce système propose des produits aux consommateurs spécifiques par exécuter successivement les trois modules de recommandation suivants, comprenant 1) le Module de Base de Données pour les Cas de Succès ; 2) le Module de Prévision du Marché ; 3) le Module de Recommandation utilisant la Connaissance. De plus, un autre module, appelé le Module de Mise à Jour de la Connaissance. Cette thèse présente une méthode originale de prévision d’un ou plusieurs profils de produits bien adaptés à un consommateur spécifique. Elle peut aider effectivement les consommateurs à effectuer des achats de vêtements sur l’Internet. En comparant avec les autres méthodes de prévision, la méthode proposée est plus robuste et plus interprétable en raison de sa capacité de traitement de l’incertitude
Garment purchasing through the Internet has become an important trend for consumers of all parts of the world. However, in various garment e-shopping systems, it systematically lacks personalized recommendations, like sales advisors in classical shops, in order to propose the most relevant products to different consumers according to their body shapes and fashion requirements. In this thesis, we propose a consumer-oriented recommendation system, which can be used inside a garment online shopping system like a virtual sales advisor. This system has been developed by integrating the professional knowledge of designers and shoppers and taking into account consumers’ perception on products. Following the shopping knowledge on garments, the proposed system recommends garment products to specific consumers by successively executing three modules, namely 1) the Successful Cases Database Module; 2) the Market Forecasting Module; 3) the Knowledge-based Recommendation Module. Also, another module, called the Knowledge Updating Module.This thesis presents an original method for predicting one or several relevant product profiles from a specific consumer profile. It can effectively help consumers to choose garments from the Internet. Compared with other prediction methods, the proposed method is more robust and interpretable owing to its capacity of treating uncertainty

APA, Harvard, Vancouver, ISO, and other styles

4

Dong, Min. "Development of an intelligent recommendation system to garment designers for designing new personalized products." Thesis, Lille 1, 2017. http://www.theses.fr/2017LIL10025/document.

Full text

Abstract:

Durant mes travaux en thèse, nous avons imaginé et poser les briques d'un système de recommandation intelligent (DIRS) orienté vers les créateurs de vêtements afin de les aider à créer des nouveaux produits personnalisés. Pour développer ce système, nous avons dans un premier temps identifié les composants clés du processus de création, puis nous avons créé un ensemble de bases de données pour collecter les données pertinentes. Dans un deuxième temps, nous avons acquis des données anthropométriques, recueilli la perception du concepteur à partir de ces mêmes morphotypes en utilisant un body scanner 3D et une procédure d'évaluation sensorielle. A la suite, une expérience instrumentale est conduite pour capturer les paramètres techniques des matières, nécessaires à leur représentation virtuelle en lien avec les morphotypes. Enfin, cinq expériences sensorielles sont réalisées pour capitaliser les connaissances des créateurs. Les données acquises servent à classer les morphotypes, à modéliser les relations entre morphotypes et facteurs de la création. A partir de ces modèles, nous avons mis en place une base de connaissances de la création mettant en œuvre une ontologie. Cette base de connaissances est mise à jour par un apprentissage dynamique au travers de nouveaux cas présentés en création. Ce système est utilisé au sein d’un nouveau processus de création. Ce processus peut s’effectuer autant de fois que nécessaire jusqu'à la satisfaction du créateur. Le système de recommandation proposé a été validé à l'aide de plusieurs cas réels
In my PhD research project, we originally propose a Designer-oriented Intelligent Recommendation System (DIRS) for supporting the design of new personalized garment products. For developing this system, we first identify the key components of a garment design process, and then set up a number of relevant databases, from which each design scheme can be formed. Second, we acquire the anthropometric data and designer’s perception on body shapes by using a 3D body scanning system and a sensory evaluation procedure. Third, an instrumental experiment is conducted for measuring the technical parameters of fabrics, and five sensory experiments are carried out in order to acquire designers’ knowledge. The acquired data are used to classify body shapes and model the relations between human bodies and the design factors. From these models, we set up an ontology-based design knowledge base. This knowledge base can be updated by dynamically learning from new design cases. On this basis, we put forward the knowledge-based recommendation system. This system is used with a newly developed design process. This process can be performed repeatedly until the designer’s satisfaction. The proposed recommendation system has been validated through a number of successful real design cases

APA, Harvard, Vancouver, ISO, and other styles

5

Lohi, Abdolkhalil. "Investigation of an intelligent personalised service recommendation system in an IMS based cellular mobile network." Thesis, University of Westminster, 2013. https://westminsterresearch.westminster.ac.uk/item/99060/investigation-of-an-intelligent-personalised-service-recommendation-system-in-an-ims-based-cellular-mobile-network.

Full text

Abstract:

Success or failure of future information and communication services in general and mobile communications in particular is greatly dependent on the level of personalisations they can offer. While the provision of anytime, anywhere, anyhow services has been the focus of wireless telecommunications in recent years, personalisation however has gained more and more attention as the unique selling point of mobile devices. Smart phones should be intelligent enough to match user’s unique needs and preferences to provide a truly personalised service tailored for the individual user. In the first part of this thesis, the importance and role of personalisation in future mobile networks is studied. This is followed, by an agent based futuristic user scenario that addresses the provision of rich data services independent of location. Scenario analysis identifies the requirements and challenges to be solved for the realisation of a personalised service. An architecture based on IP Multimedia Subsystem is proposed for mobility and to provide service continuity whilst roaming between two different access standards. Another aspect of personalisation, which is user preference modelling, is investigated in the context of service selection in a multi 3rd party service provider environment. A model is proposed for the automatic acquisition of user preferences to assist in service selection decision-making. User preferences are modelled based on a two-level Bayesian Metanetwork. Personal agents incorporating the proposed model provide answers to preference related queries such as cost, QoS and service provider reputation. This allows users to have their preferences considered automatically.

APA, Harvard, Vancouver, ISO, and other styles

6

Chi, Cheng. "Personalized pattern recommendation system of men’s shirts based on precise body measurement." Electronic Thesis or Diss., Centrale Lille Institut, 2022. http://www.theses.fr/2022CLIL0003.

Full text

Abstract:

Les systèmes commerciaux de recommandation de vêtements ont été largement utilisés dans l'industrie de l'habillement. Cependant, les recherches existantes sur la conception de vêtements numériques se sont concentrées sur les évolutions techniques du processus de conception virtuelle, avec peu de retours de métier provenant des designers. La coupe d'un vêtement joue un rôle important dans l'achat de celui-ci par le client. Afin de développer un vêtement correctement ajusté, les stylistes et les modélistes doivent ajuster le patron du vêtement plusieurs fois jusqu'à ce que le client soit satisfait. Actuellement, le modélisme traditionnel présente trois inconvénients majeurs : 1) il est très long et inefficace, 2) il repose trop sur des concepteurs expérimentés, 3) la relation entre la forme du corps humain et le vêtement n'est pas pleinement explorée. Dans la pratique, le styliste joue un rôle clé dans la réussite du processus de conception. Il est nécessaire d'intégrer les connaissances et l'expérience du styliste dans les systèmes actuels de CAD de vêtements afin de fournir rapidement une solution de conception réalisable, centrée sur l'homme et à faible coût, pour chaque besoin personnalisé. En outre, les services basés sur les données, tels que les systèmes de recommandation, la classification des formes corporelles, la modélisation du corps en 3D et l'évaluation de l'ajustement des vêtements, devraient être intégrés dans le système de CAD de l'habillement afin d'améliorer l'efficacité du processus de conception.Sur la base de ces besoins, cette thèse propose un système de recommandation intelligent composé de modèles de vêtements ajustables pour conduire à la conception de vêtements personnalisés. Le système fonctionne en combinaison avec un nouveau processus de conception nouvellement développé, à savoir l'identification de la forme du corps humain - la recommandation d'une solution de conception - la représentation virtuelle 3D et l'évaluation - l'ajustement des paramètres de conception. Ce processus peut être répété jusqu'à ce que l'utilisateur soit satisfait. Le système de recommandation proposé a été validé par quelques cas pratiques de conception réussis
Commercial garment recommendation systems have been widely used in the apparel industry. However, existing research on digital garment design has focused on the technical development of the virtual design process, with little knowledge of traditional designers. The fit of a garment plays a significant role in whether a customer purchases that garment. In order to develop a well-fitting garment, designers and pattern makers should adjust the garment pattern several times until the customer is satisfied. Currently, there are three main disadvantages of traditional pattern-making: 1) it is very time-consuming and inefficient, 2) it relies too much on experienced designers, 3) the relationship between the human body shape and the garment is not fully explored. In practice, the designer plays a key role in a successful design process. There is a need to integrate the designer's knowledge and experience into current garment CAD systems to provide a feasible human-centered, low-cost design solution quickly for each personalized requirement. Also, data-based services such as recommendation systems, body shape classification, 3D body modelling, and garment fit assessment should be integrated into the apparel CAD system to improve the efficiency of the design process.Based on the above issues, in this thesis, a fit-oriented garment pattern intelligent recommendation system is proposed for supporting the design of personalized garment products. The system works in combination with a newly developed design process, i.e. body shape identification - design solution recommendation - 3D virtual presentation and evaluation - design parameter adjustment. This process can be repeated until the user is satisfied. The proposed recommendation system has been validated by some successful practical design cases

APA, Harvard, Vancouver, ISO, and other styles

7

Robles, Sebastian. "Business intelligence in Chile, recommendations to develop local applications." Thesis, Massachusetts Institute of Technology, 2010. http://hdl.handle.net/1721.1/70831.

Full text

Abstract:

Thesis (S.M. in Engineering and Management)--Massachusetts Institute of Technology, Engineering Systems Division, June 2011.
"February 2010." Cataloged from PDF version of thesis.
Includes bibliographical references (p. 60).
The volume of information generated from enterprise applications is growing exponentially, and the cost of storage is decreasing rapidly. In addition, cloud-based applications, mobile devices and social networks are becoming relevant sources of unstructured data that provide essential information for strategic decisions making. Therefore, with time, enterprise databases will become more valuable for business but also much harder to integrate, process and analyze. Business Intelligence software was instrumental in helping organizations to analyze information and provide reports to support business decision-making. Accordingly, BI applications evolved as enterprise information grew, hardware-processing capacities developed, and storage cost is being reduced significantly. In this paper, we will analyze the current BI world market and compare it with the Chilean market, in order to come up with business plan recommendations for local developers and systems integrators interested in capitalizing the opportunities generated by the global BI software market consolidation.
by Sebastian Robles.
S.M.in Engineering and Management

APA, Harvard, Vancouver, ISO, and other styles

8

Schröder, Anna Marie. "Unboxing The Algorithm : Understandability And Algorithmic Experience In Intelligent Music Recommendation Systems." Thesis, Malmö universitet, Institutionen för konst, kultur och kommunikation (K3), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-43841.

Full text

Abstract:

After decades of black-boxing the existence of algorithms in technologies of daily need, users lack confidence in handling them. This thesis study investigates the use situation of intelligent music recommendation systems and explores how understandability as a principle drawn from sociology, design, and computing can enhance the algorithmic experience. In a Research-Through-Design approach, the project conducted focus user sessions and an expert interview to explore first-hand insights. The analysis showed that users had limited mental models so far but brought curiosity to learn. Explorative prototyping revealed that explanations could improve the algorithmic experience in music recommendation systems. Users could comprehend information the best when it was easy to access and digest, directly related to user behavior, and gave control to correct the algorithm. Concluding, trusting users with more transparent handling of algorithmic workings might make authentic recommendations from intelligent systems applicable in the long run.

APA, Harvard, Vancouver, ISO, and other styles

9

Lagerqvist, Gustaf, and Anton Stålhandske. "Recommendation systems for recruitment within an educational context." Thesis, Malmö universitet, Fakulteten för teknik och samhälle (TS), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-42902.

Full text

Abstract:

Alongside the evolution of the recruitment process, different types of recommendation systems have been developed. The purpose of this study is to investigate recommendation systems within educational contexts, successful implementations of recommendation system architecture patterns, and alternatives to previous experience when evaluating candidates. The study is conducted through two separate methods; A literature review with a qualitative approach and design science research methodology focused on design and development, demonstration and evaluation. The literature review shows that, for recommendation systems, a layered architecture built within a microservice ecosystem is successfully utilized and has multiple beneficial aspects such as improved scalability, maintainability and security. Through design science research methodology, this study shows a suggested approach to implementing a layered architecture in combination with KNN and hybrid filtering. To avoid the lapse of suitable candidates, caused by demanding previous experience, this study shows an alternative approach to recruitment, within an educational context, through the use of soft skills. Within the study, this approach is successfully used to evaluate and compare students, but the same approach could possibly be applied to evaluate and compare companies. Moving forward, this study could be further expanded by looking into possible biases arising as a result of using AI and choices made during this study, as well as weighting of student-attributes.

APA, Harvard, Vancouver, ISO, and other styles

10

Sun, Runpu. "Using Social Media Intelligence to Support Business Knowledge Discovery and Decision Making." Diss., The University of Arizona, 2011. http://hdl.handle.net/10150/145394.

Full text

Abstract:

The new social media sites - blogs, micro-blogs, and social networking sites, among others - are gaining considerable momentum to facilitate collaboration and social interactions in general. These sites provide a tremendous asset for understanding social phenomena by providing a wide availability of novel data sources. Recent estimates suggest that social media sites are responsible for as much as one third of new Web content, in the forms of social networks, comments, trackbacks, advertisements, tags, etc. One critical and immediate challenge facing the MIS researchers then becomes - how to effectively utilize this huge wealth of social media data, to facilitate business knowledge discovery and decision making.Among these available data sources, social networks constitute the backbone of almost all social media sites. These network structures provide a rich description of the social scenes and contexts, which is helpful for us to address the above challenge. In this dissertation, I have primarily employed the probabilistic network models, to study various social network related problems arose from the use of social media services. In Chapter 2 and Chapter 3, I studied how information overload can affect the efficiency of information diffusion in online social networks (Delicious.com and Digg.com). Novel diffusion model were proposed to model the observed information overload. The models and their extensions are thoroughly evaluated by solving the Influence Maximization problem related to information diffusion and viral marketing applications. In Chapter 4, I studied the information overload in a micro-blogging application (Twitter.com) using a design science methodology. A content recommendation framework was proposed to help micro-blogging users to efficiently identify quality emergency news feeds. Chapter 5 presents a novel burst detection algorithm concerning identifying and analyzing correlated burst patterns by considering multiple inputs (data streams) that co-evolve over time. The algorithm was later used for discovering burst keywords/tag pairs from online social communities, which are strong indicators of emerging or changing user interests.Chapter 6 concludes this dissertation by highlighting major research contributions and future directions.

APA, Harvard, Vancouver, ISO, and other styles

11

Alsalama, Ahmed. "A Hybrid Recommendation System Based on Association Rules." TopSCHOLAR®, 2013. http://digitalcommons.wku.edu/theses/1250.

Full text

Abstract:

Recommendation systems are widely used in e-commerce applications. Theengine of a current recommendation system recommends items to a particular user based on user preferences and previous high ratings. Various recommendation schemes such as collaborative filtering and content-based approaches are used to build a recommendation system. Most of current recommendation systems were developed to fit a certain domain such as books, articles, and movies. We propose a hybrid framework recommendation system to be applied on two dimensional spaces (User × Item) with a large number of users and a small number of items. Moreover, our proposed framework makes use of both favorite and non-favorite items of a particular user. The proposed framework is built upon the integration of association rules mining and the content-based approach. The results of experiments show that our proposed framework can provide accurate recommendations to users.

APA, Harvard, Vancouver, ISO, and other styles

12

Khoshkangini, Reza. "Personalized Game Content Generation and Recommendation for Gamified Systems." Doctoral thesis, Università degli studi di Padova, 2018. http://hdl.handle.net/11577/3424854.

Full text

Abstract:

Gamification, that is, the usage of game content in non-game contexts, has been successfully employed in several application domains to foster engagement, as well as to influence the behavior of end users. Although gamification is often effective in inducing behavioral changes in citizens, the difficulty in retaining players and sustaining the acquired behavior over time, shows some limitations of this technology. That is especially unfortunate, because changing players’ demeanor (which have been shaped for a long time), cannot be immediately internalized; rather, the gamification incentive must be reinforced to lead to stabilization. This issue could be sourced from utilizing static game content and a one-size-fits-all strategy in generating the content during the game. This reveals the need for dynamic personalization over the course of the game. Our research hypothesis is that we can overcome these limitations with Procedural Content Generation (PCG) of playable units that appeal to each individual player and make her user experience more varied and compelling. In this thesis, we propose a deep, large and long solution, deployed in two main phases of Design and Integration to tackle these limitations. To support the former phase, we present a “PCG and Recommender system” to automate the generation and recommendation of playable units, named “Challenges”, which are Personalized and Contextualized on the basis of players’ preferences, skills, etc., and the game ulterior objectives. To this end, we develop a multi-layered framework to generate the personalized game content to be assigned and recommended to the players involved in the gamified system. To support the latter phase, we integrate two modules into the system including Machine Learning (ML) and Player Modeling, in order to optimize the challenge selection process and learning players’ behavior to further improve the personalization, by deriving the style of the player, respectively. We have carried out the implementation and evaluation of the proposed framework and its integration in two different contexts. First, we assess our Automatic Procedural Content Generation and Recommendation (APCGR) system within a large-scale and long-running open field experiment promoting sustainable urban mobility that lasted twelve weeks and involved more than 400 active players. Then, we implement the “Player Modeling” module (in the integration phase) in an educational interactive game domain to assess the performance of the proposed play style extraction approach. The contributions of this dissertation are a first step toward the application of machine learning in automating the procedural content generation and recommendation in gamification systems.

APA, Harvard, Vancouver, ISO, and other styles

13

Drushku, Krista. "User intent based recommendation for modern BI systems." Thesis, Tours, 2019. http://www.theses.fr/2019TOUR4001/document.

Full text

Abstract:

Stocker de grandes quantités de données complexifie les interactions avec les systèmes de Business Intelligence (BI). Les systèmes de recommandation semblent un choix logique pour aider les utilisateurs dans leur analyse. Ils extraient des comportements de données historiques et suggèrent des actions personnalisées, potentiellement redondantes, via des scores de similarité. La diversité est essentielle pour améliorer la satisfaction des utilisateurs, d’où l’intérêt particulier accordé aux recommandations complémentaires. Nous avons étudié deux problèmes concrets d’exploration de données en BI et proposons de découvrir et exploiter les intentions utilisateur pour fournir deux recommandeurs de requête. Le premier, un recommandeur collaboratif réactif original basé sur l’intention, recommande des séquences de requêtes à l’utilisateur pour poursuivre son analyse. Le second propose proactivement un ensemble de requêtes pour compléter un rapport BI, en fonction di contexte utilisateur
The storage of big amounts of data may lead to a series of long questions towards the expected solution which complicates user interactions with Business Intelligence (BI) systems. Recommender systems appear as a natural solution to help the users complete their analysis. They try to discover user behaviors from the past logs and to suggest personalized actions by predicting lists of likeness scores, which may lead to redundant recommendations. Nowadays, diversity is becoming essential to improve users’ satisfaction, thus, a special interest is dedicated to complementary recommendation. We studied two concrete data exploration problems in BI and we propose to discover and leverage the user intents to provide two query recommenders. The first, an original reactive collaborative Intent-based Recommender, recommends sequences of queries for the user to pursue her analysis. The second one proactively proposes a bundle of queries to complete user BI report, based on the user intents

APA, Harvard, Vancouver, ISO, and other styles

14

Shapiro, Daniel. "Composing Recommendations Using Computer Screen Images: A Deep Learning Recommender System for PC Users." Thesis, Université d'Ottawa / University of Ottawa, 2017. http://hdl.handle.net/10393/36272.

Full text

Abstract:

A new way to train a virtual assistant with unsupervised learning is presented in this thesis. Rather than integrating with a particular set of programs and interfaces, this new approach involves shallow integration between the virtual assistant and computer through machine vision. In effect the assistant interprets the computer screen in order to produce helpful recommendations to assist the computer user. In developing this new approach, called AVRA, the following methods are described: an unsupervised learning algorithm which enables the system to watch and learn from user behavior, a method for fast filtering of the text displayed on the computer screen, a deep learning classifier used to recognize key onscreen text in the presence of OCR translation errors, and a recommendation filtering algorithm to triage the many possible action recommendations. AVRA is compared to a similar commercial state-of-the-art system, to highlight how this work adds to the state of the art. AVRA is a deep learning image processing and recommender system that can col- laborate with the computer user to accomplish various tasks. This document presents a comprehensive overview of the development and possible applications of this novel vir- tual assistant technology. It detects onscreen tasks based upon the context it perceives by analyzing successive computer screen images with neural networks. AVRA is a rec- ommender system, as it assists the user by producing action recommendations regarding onscreen tasks. In order to simplify the interaction between the user and AVRA, the system was designed to only produce action recommendations that can be accepted with a single mouse click. These action recommendations are produced without integration into each individual application executing on the computer. Furthermore, the action recommendations are personalized to the user’s interests utilizing a history of the user’s interaction.

APA, Harvard, Vancouver, ISO, and other styles

15

Casey, Walker Evan. "Scalable Collaborative Filtering Recommendation Algorithms on Apache Spark." Scholarship @ Claremont, 2014. http://scholarship.claremont.edu/cmc_theses/873.

Full text

Abstract:

Collaborative filtering based recommender systems use information about a user's preferences to make personalized predictions about content, such as topics, people, or products, that they might find relevant. As the volume of accessible information and active users on the Internet continues to grow, it becomes increasingly difficult to compute recommendations quickly and accurately over a large dataset. In this study, we will introduce an algorithmic framework built on top of Apache Spark for parallel computation of the neighborhood-based collaborative filtering problem, which allows the algorithm to scale linearly with a growing number of users. We also investigate several different variants of this technique including user and item-based recommendation approaches, correlation and vector-based similarity calculations, and selective down-sampling of user interactions. Finally, we provide an experimental comparison of these techniques on the MovieLens dataset consisting of 10 million movie ratings.

APA, Harvard, Vancouver, ISO, and other styles

16

Mittal, Nupur. "Data, learning and privacy in recommendation systems." Thesis, Rennes 1, 2016. http://www.theses.fr/2016REN1S084/document.

Full text

Abstract:

Les systèmes de recommandation sont devenus une partie indispensable des services et des applications d’internet, en particulier dû à la surcharge de données provenant de nombreuses sources. Quel que soit le type, chaque système de recommandation a des défis fondamentaux à traiter. Dans ce travail, nous identifions trois défis communs, rencontrés par tous les types de systèmes de recommandation: les données, les modèles d'apprentissage et la protection de la vie privée. Nous élaborons différents problèmes qui peuvent être créés par des données inappropriées en mettant l'accent sur sa qualité et sa quantité. De plus, nous mettons en évidence l'importance des réseaux sociaux dans la mise à disposition publique de systèmes de recommandation contenant des données sur ses utilisateurs, afin d'améliorer la qualité des recommandations. Nous fournissons également les capacités d'inférence de données publiques liées à des données relatives aux utilisateurs. Dans notre travail, nous exploitons cette capacité à améliorer la qualité des recommandations, mais nous soutenons également qu'il en résulte des menaces d'atteinte à la vie privée des utilisateurs sur la base de leurs informations. Pour notre second défi, nous proposons une nouvelle version de la méthode des k plus proches voisins (knn, de l'anglais k-nearest neighbors), qui est une des méthodes d'apprentissage parmi les plus populaires pour les systèmes de recommandation. Notre solution, conçue pour exploiter la nature bipartie des ensembles de données utilisateur-élément, est évolutive, rapide et efficace pour la construction d'un graphe knn et tire sa motivation de la grande quantité de ressources utilisées par des calculs de similarité dans les calculs de knn. Notre algorithme KIFF utilise des expériences sur des jeux de données réelles provenant de divers domaines, pour démontrer sa rapidité et son efficacité lorsqu'il est comparé à des approches issues de l'état de l'art. Pour notre dernière contribution, nous fournissons un mécanisme permettant aux utilisateurs de dissimuler leur opinion sur des réseaux sociaux sans pour autant dissimuler leur identité
Recommendation systems have gained tremendous popularity, both in academia and industry. They have evolved into many different varieties depending mostly on the techniques and ideas used in their implementation. This categorization also marks the boundary of their application domain. Regardless of the types of recommendation systems, they are complex and multi-disciplinary in nature, involving subjects like information retrieval, data cleansing and preprocessing, data mining etc. In our work, we identify three different challenges (among many possible) involved in the process of making recommendations and provide their solutions. We elaborate the challenges involved in obtaining user-demographic data, and processing it, to render it useful for making recommendations. The focus here is to make use of Online Social Networks to access publicly available user data, to help the recommendation systems. Using user-demographic data for the purpose of improving the personalized recommendations, has many other advantages, like dealing with the famous cold-start problem. It is also one of the founding pillars of hybrid recommendation systems. With the help of this work, we underline the importance of user’s publicly available information like tweets, posts, votes etc. to infer more private details about her. As the second challenge, we aim at improving the learning process of recommendation systems. Our goal is to provide a k-nearest neighbor method that deals with very large amount of datasets, surpassing billions of users. We propose a generic, fast and scalable k-NN graph construction algorithm that improves significantly the performance as compared to the state-of-the art approaches. Our idea is based on leveraging the bipartite nature of the underlying dataset, and use a preprocessing phase to reduce the number of similarity computations in later iterations. As a result, we gain a speed-up of 14 compared to other significant approaches from literature. Finally, we also consider the issue of privacy. Instead of directly viewing it under trivial recommendation systems, we analyze it on Online Social Networks. First, we reason how OSNs can be seen as a form of recommendation systems and how information dissemination is similar to broadcasting opinion/reviews in trivial recommendation systems. Following this parallelism, we identify privacy threat in information diffusion in OSNs and provide a privacy preserving algorithm for the same. Our algorithm Riposte quantifies the privacy in terms of differential privacy and with the help of experimental datasets, we demonstrate how Riposte maintains the desirable information diffusion properties of a network

APA, Harvard, Vancouver, ISO, and other styles

17

Olmucci, Poddubnyy Oleksandr. "Graph Neural Networks for Recommender Systems." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2022. http://amslaurea.unibo.it/25033/.

Full text

Abstract:

In recent years, a new type of deep learning models, Graph Neural Networks (GNNs), have demonstrated to be a powerful learning paradigm when applied to problems that can be described via graph data, due to their natural ability to integrate representations across nodes that are connected via some topological structure. One of such domains is Recommendation Systems, the majority of whose data can be naturally represented via graphs. For example, typical item recommendation datasets can be represented via user-item bipartite graphs, social recommendation datasets by social networks, and so on. The successful application of GNNs to the field of recommendation, is demonstrated by the state of the art results achieved on various datasets, making GNNs extremely appealing in this domain, also from a commercial perspective. However, the introduction of graph layers and their associated sampling techniques significantly affects the nature of the calculations that need to be performed on GPUs, the main computational accelerator used nowadays: something that hasn't been investigated so far by any of the architectures in the recommendation literature. This thesis aims to fill this gap by conducting the first systematic empirical investigation of GNN-based architectures for recommender systems, focusing on their multi-GPU scalability and precision speed-up properties, when using different types of hardware.

APA, Harvard, Vancouver, ISO, and other styles

18

Nyberg, Selma. "Video Recommendation Based on Object Detection." Thesis, Uppsala universitet, Avdelningen för systemteknik, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-351122.

Full text

Abstract:

In this thesis, various machine learning domains have been combined in order to build a video recommender system that is based on object detection. The work combines two extensively studied research fields, recommender systems and computer vision, that also are rapidly growing and popular techniques on commercial markets. To investigate the performance of the approach, three different content-based recommender systems have been implemented at Spotify, which are based on the following video features: object detections, titles and descriptions, and user preferences. These systems have then been evaluated and compared against each other together with their hybridized result. Two algorithms have been implemented, the prediction and the top-N algorithm, where the former is the more reliable source for evaluating the system's performance. The evaluation of the system shows that the overall performance scores for predicting values of the users' liked and disliked videos are in the range from about 40 % to 70 % for the prediction algorithm and from about 15 % to 70 % for the top-N algorithm. The approach based on object detection performs worse in comparison to the other approaches. Hence, there seems to be is a low correlation between the user preferences and the video contents in terms of object detection data. Therefore, this data is not very suitable for describing the content of videos and using it in the recommender system. However, the results of this study cannot be generalized to apply for other systems before the approach has been evaluated in other environments and for various data sets. Moreover, there are plenty of room for refinements and improvements to the system, as well as there are many interesting research areas for future work.

APA, Harvard, Vancouver, ISO, and other styles

19

Lisena, Pasquale. "Knowledge-based music recommendation : models, algorithms and exploratory search." Electronic Thesis or Diss., Sorbonne université, 2019. http://www.theses.fr/2019SORUS614.

Full text

Abstract:

Représenter l'information décrivant la musique est une activité complexe, qui implique différentes sous-tâches. Ce manuscrit de thèse porte principalement sur la musique classique et étudie comment représenter et exploiter ses informations. L'objectif principal est l'étude de stratégies de représentation et de découverte des connaissances appliquées à la musique classique, dans des domaines tels que la production de base de connaissances, la prédiction de métadonnées et les systèmes de recommandation. Nous proposons une architecture pour la gestion des métadonnées de musique à l'aide des technologies du Web Sémantique. Nous introduisons une ontologie spécialisée et un ensemble de vocabulaires contrôlés pour les différents concepts spécifiques à la musique. Ensuite, nous présentons une approche de conversion des données, afin d’aller au-delà de la pratique bibliothécaire actuellement utilisée, en s’appuyant sur des règles de mapping et sur l’interconnexion avec des vocabulaires contrôlés. Enfin, nous montrons comment ces données peuvent être exploitées. En particulier, nous étudions des approches basées sur des plongements calculés sur des métadonnées structurées, des titres et de la musique symbolique pour classer et recommander de la musique. Plusieurs applications de démonstration ont été réalisées pour tester les approches et les ressources précédentes
Representing the information about music is a complex activity that involves different sub-tasks. This thesis manuscript mostly focuses on classical music, researching how to represent and exploit its information. The main goal is the investigation of strategies of knowledge representation and discovery applied to classical music, involving subjects such as Knowledge-Base population, metadata prediction, and recommender systems. We propose a complete workflow for the management of music metadata using Semantic Web technologies. We introduce a specialised ontology and a set of controlled vocabularies for the different concepts specific to music. Then, we present an approach for converting data, in order to go beyond the librarian practice currently in use, relying on mapping rules and interlinking with controlled vocabularies. Finally, we show how these data can be exploited. In particular, we study approaches based on embeddings computed on structured metadata, titles, and symbolic music for ranking and recommending music. Several demo applications have been realised for testing the previous approaches and resources

APA, Harvard, Vancouver, ISO, and other styles

20

Hou, Hailong. "Computing with Granular Words." Digital Archive @ GSU, 2011. http://digitalarchive.gsu.edu/cs_theses/73.

Full text

Abstract:

Computational linguistics is a sub-field of artificial intelligence; it is an interdisciplinary field dealing with statistical and/or rule-based modeling of natural language from a computational perspective. Traditionally, fuzzy logic is used to deal with fuzziness among single linguistic terms in documents. However, linguistic terms may be related to other types of uncertainty. For instance, different users search ‘cheap hotel’ in a search engine, they may need distinct pieces of relevant hidden information such as shopping, transportation, weather, etc. Therefore, this research work focuses on studying granular words and developing new algorithms to process them to deal with uncertainty globally. To precisely describe the granular words, a new structure called Granular Information Hyper Tree (GIHT) is constructed. Furthermore, several technologies are developed to cooperate with computing with granular words in spam filtering and query recommendation. Based on simulation results, the GIHT-Bayesian algorithm can get more accurate spam filtering rate than conventional method Naive Bayesian and SVM; computing with granular word also generates better recommendation results based on users’ assessment when applied it to search engine.

APA, Harvard, Vancouver, ISO, and other styles

21

Gutowski, Nicolas. "Recommandation contextuelle de services : application à la recommandation d'évènements culturels dans la ville intelligente." Thesis, Angers, 2019. http://www.theses.fr/2019ANGE0030.

Full text

Abstract:

Les algorithmes de bandits-manchots pour les systèmes de recommandation sensibles au contexte font aujourd’hui l’objet de nombreuses études. Afin de répondre aux enjeux de cette thématique, les contributions de cette thèse sont organisées autour de 3 axes : 1) les systèmes de recommandation ; 2) les algorithmes de bandits-manchots (contextuels et non contextuels) ; 3) le contexte. La première partie de nos contributions a porté sur les algorithmes de bandits-manchots pour la recommandation. Elle aborde la diversification des recommandations visant à améliorer la précision individuelle. La seconde partie a porté sur la capture de contexte, le raisonnement contextuel pour les systèmes de recommandation d’événements culturels dans la ville intelligente, et l’enrichissement dynamique de contexte pour les algorithmes de bandits-manchots contextuels
Nowadays, Multi-Armed Bandit algorithms for context-aware recommendation systems are extensively studied. In order to meet challenges underlying this field of research, our works and contributions have been organised according to three research directions : 1) recommendation systems ; 2) Multi-Armed Bandit (MAB) and Contextual Multi-Armed Bandit algorithms (CMAB) ; 3) context.The first part of our contributions focuses on MAB and CMAB algorithms for recommendation. It particularly addresses diversification of recommendations for improving individual accuracy. The second part is focused on contextacquisition, on context reasoning for cultural events recommendation systems for Smart Cities, and on dynamic context enrichment for CMAB algorithms

APA, Harvard, Vancouver, ISO, and other styles

22

Tomasone, Marco Benito. "Pipeline per il Machine Learning: Analisi dei workflow e framework per l’orchestrazione i casi Recommendation System e Face2Face Traslation." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2021.

Find full text

Abstract:

Al giorno d’oggi tantissimi dei problemi che affrontiamo quotidianamente prevedono l’utilizzo di tecniche di Intelligenza Artificiale. Sono nate sempre più techinchè fino al machine learning è una materia in forte sviluppo, e i modelli, gli algoritmi si espandono a vista d’occhio. Spesso molti dei problemi vengono affrontati e risolti tramite pipeline di modelli di machine learning che sequenzializzati fra loro portano alla soluzione sperata. Nascono sempre più piattaforme come Ai4Eu che si pongono come centro di scambio di modelli, dataset e conoscenza. Nasce quindi l'esigenza di voler automatizzare su queste piattaforme la creazione delle pipeline di ml riutilizzando ove possibile il codice. Questo lavoro di tesi, sfruttando un approccio bottom-up, prevede un’attenta osservazione del workflow del Recommendation System di YouTube. Succesivamente si valutano gli approcci standard in letteratura, individuando due principali classi di Recommendation System in funzione del filtraggio applicato, il Collaborative filtering (classe di appartenenza del Recommendation System di YouTube) o l’Item-Based filtering. Si nota come la parte più importante di questo tipo di applicativi riguardi la gestione dei dati. Seguendo lo stesso metodo operativo viene studiata la pipeline di Face2Face Traslation, analizzando per ogni suo componente l’approccio ai dati e la struttura del modello e confrontando ogni componente con i suoi corrispettivi in letteratura, per valutarne invarianti e versalità mostrando come alcuni modelli si presentino con modelli standard e approcci ai dati diversi, mentre altri presentino approcci standard ai dati ma una grande varietà nei modelli a disposizione. Vengono infine presentati tre framework per l’orchestrazione di pipeline di Machine Learning: MLRun, ZenML e Kale, scelti poichè permettono il deployment e la riusabilità del codice. Si osserva come, escluse piccole differenze, questi tre framework si presentano molto equivalenti fra loro.

APA, Harvard, Vancouver, ISO, and other styles

23

Kaufman, Jaime C. "A Hybrid Approach to Music Recommendation: Exploiting Collaborative Music Tags and Acoustic Features." UNF Digital Commons, 2014. http://digitalcommons.unf.edu/etd/540.

Full text

Abstract:

Recommendation systems make it easier for an individual to navigate through large datasets by recommending information relevant to the user. Companies such as Facebook, LinkedIn, Twitter, Netflix, Amazon, Pandora, and others utilize these types of systems in order to increase revenue by providing personalized recommendations. Recommendation systems generally use one of the two techniques: collaborative filtering (i.e., collective intelligence) and content-based filtering. Systems using collaborative filtering recommend items based on a community of users, their preferences, and their browsing or shopping behavior. Examples include Netflix, Amazon shopping, and Last.fm. This approach has been proven effective due to increased popularity, and its accuracy improves as its pool of users expands. However, the weakness with this approach is the Cold Start problem. It is difficult to recommend items that are either brand new or have no user activity. Systems that use content-based filtering recommend items based on extracted information from the actual content. A popular example of this approach is Pandora Internet Radio. This approach overcomes the Cold Start problem. However, the main issue with this approach is its heavy demand on computational power. Also, the semantic meaning of an item may not be taken into account when producing recommendations. In this thesis, a hybrid approach is proposed by utilizing the strengths of both collaborative and content-based filtering techniques. As proof-of-concept, a hybrid music recommendation system was developed and evaluated by users. The results show that this system effectively tackles the Cold Start problem and provides more variation on what is recommended.

APA, Harvard, Vancouver, ISO, and other styles

24

Gonard, François. "Cold-start recommendation : from Algorithm Portfolios to Job Applicant Matching." Thesis, Université Paris-Saclay (ComUE), 2018. http://www.theses.fr/2018SACLS121/document.

Full text

Abstract:

La quantité d'informations, de produits et de relations potentielles dans les réseaux sociaux a rendu indispensable la mise à disposition de recommandations personnalisées. L'activité d'un utilisateur est enregistrée et utilisée par des systèmes de recommandation pour apprendre ses centres d'intérêt. Les recommandations sont également utiles lorsqu'estimer la pertinence d'un objet est complexe et repose sur l'expérience. L'apprentissage automatique offre d'excellents moyens de simuler l'expérience par l'emploi de grandes quantités de données.Cette thèse examine le démarrage à froid en recommandation, situation dans laquelle soit un tout nouvel utilisateur désire des recommandations, soit un tout nouvel objet est proposé à la recommandation. En l'absence de données d'intéraction, les recommandations reposent sur des descriptions externes. Deux problèmes de recommandation de ce type sont étudiés ici, pour lesquels des systèmes de recommandation spécialisés pour le démarrage à froid sont présentés.En optimisation, il est possible d'aborder le choix d'algorithme dans un portfolio d'algorithmes comme un problème de recommandation. Notre première contribution concerne un système à deux composants, un sélecteur et un ordonnanceur d'algorithmes, qui vise à réduire le coût de l'optimisation d'une nouvelle instance d'optimisation tout en limitant le risque d'un échec de l'optimisation. Les deux composants sont entrainés sur les données du passé afin de simuler l'expérience, et sont alternativement optimisés afin de les faire coopérer. Ce système a remporté l'Open Algorithm Selection Challenge 2017.L'appariement automatique de chercheurs d'emploi et d'offres est un problème de recommandation très suivi par les plateformes de recrutement en ligne. Une seconde contribution concerne le développement de techniques spécifiques pour la modélisation du langage naturel et leur combinaison avec des techniques de recommandation classiques afin de tirer profit à la fois des intéractions passées des utilisateurs et des descriptions textuelles des annonces. Le problème d'appariement d'offres et de chercheurs d'emploi est étudié à travers le prisme du langage naturel et de la recommandation sur deux jeux de données tirés de contextes réels. Une discussion sur la pertinence des différents systèmes de recommandations pour des applications similaires est proposée
The need for personalized recommendations is motivated by the overabundance of online information, products, social connections. This typically tackled by recommender systems (RS) that learn users interests from past recorded activities. Another context where recommendation is desirable is when estimating the relevance of an item requires complex reasoning based on experience. Machine learning techniques are good candidates to simulate experience with large amounts of data.The present thesis focuses on the cold-start context in recommendation, i.e. the situation where either a new user desires recommendations or a brand-new item is to be recommended. Since no past interaction is available, RSs have to base their reasoning on side descriptions to form recommendations. Two of such recommendation problems are investigated in this work. Recommender systems designed for the cold-start context are designed.The problem of choosing an optimization algorithm in a portfolio can be cast as a recommendation problem. We propose a two components system combining a per-instance algorithm selector and a sequential scheduler to reduce the optimization cost of a brand-new problem instance and mitigate the risk of optimization failure. Both components are trained with past data to simulate experience, and alternatively optimized to enforce their cooperation. The final system won the Open Algorithm Challenge 2017.Automatic job-applicant matching (JAM) has recently received considerable attention in the recommendation community for applications in online recruitment platforms. We develop specific natural language (NL) modeling techniques and combine them with standard recommendation procedures to leverage past user interactions and the textual descriptions of job positions. The NL and recommendation aspects of the JAM problem are studied on two real-world datasets. The appropriateness of various RSs on applications similar to the JAM problem are discussed

APA, Harvard, Vancouver, ISO, and other styles

25

Falih, Issam. "Attributed Network Clustering : Application to recommender systems." Thesis, Sorbonne Paris Cité, 2018. http://www.theses.fr/2018USPCD011/document.

Full text

Abstract:

Au cours de la dernière décennie, les réseaux (les graphes) se sont révélés être un outil efficace pour modéliser des systèmes complexes. La problématique de détection de communautés est une tâche centrale dans l’analyse des réseaux complexes. La majeur partie des travaux dans ce domaine s’intéresse à la structure topologique des réseaux. Cependant, dans plusieurs cas réels, les réseaux complexes ont un ensemble d’attributs associés aux nœuds et/ou aux liens. Ces réseaux sont dites : réseaux attribués. Mes activités de recherche sont basées principalement sur la détection des communautés dans les réseaux attribués. Pour aborder ce problème, on s’est intéressé dans un premier temps aux attributs relatifs aux liens, qui sont un cas particulier des réseaux multiplexes. Un multiplex est un modèle de graphe multi-relationnel. Il est souvent représenté par un graphe multi-couches. Chaque couche contient le même ensemble de nœuds mais encode une relation différente. Dans mes travaux de recherche, nous proposons une étude comparative des différentes approches de détection de communautés dans les réseaux multiplexes. Cette étude est faite sur des réseaux réels. Nous proposons une nouvelle approche centrée "graine" pour la détection de communautés dans les graphes multiplexes qui a nécessité la redéfinition des métriques de bases des réseaux complexes au cas multiplex. Puis, nous proposons une approche de clustering dans les réseaux attribués qui prend en considération à la fois les attributs sur les nœuds et sur les liens. La validation de mes approches a été faite avec des indices internes et externes, mais aussi par une validation guidée par un système de recommandation que nous avons proposé et dont la détection de communautés est sa tâche principale. Les résultats obtenus sur ces approches permettent d’améliorer la qualité des communautés détectées en prenant en compte les informations sur les attributs du réseaux. De plus, nous offrons des outils d’analyse des réseaux attribués sous le langage de programmation R
In complex networks analysis field, much effort has been focused on identifying graphs communities of related nodes with dense internal connections and few external connections. In addition to node connectivity information that are mostly composed by different types of links, most real-world networks contains also node and/or edge associated attributes which can be very relevant during the learning process to find out the groups of nodes i.e. communities. In this case, two types of information are available : graph data to represent the relationship between objects and attributes information to characterize the objects i.e nodes. Classic community detection and data clustering techniques handle either one of the two types but not both. Consequently, the resultant clustering may not only miss important information but also lead to inaccurate findings. Therefore, various methods have been developed to uncover communities in networks by combining structural and attribute information such that nodes in a community are not only densely connected, but also share similar attribute values. Such graph-shape data is often referred to as attributed graph.This thesis focuses on developing algorithms and models for attributed graphs. Specifically, I focus in the first part on the different types of edges which represent different types of relations between vertices. I proposed a new clustering algorithms and I also present a redefinition of principal metrics that deals with this type of networks.Then, I tackle the problem of clustering using the node attribute information by describing a new original community detection algorithm that uncover communities in node attributed networks which use structural and attribute information simultaneously. At last, I proposed a collaborative filtering model in which I applied the proposed clustering algorithms

APA, Harvard, Vancouver, ISO, and other styles

26

Bahceci, Oktay. "Deep Neural Networks for Context Aware Personalized Music Recommendation : A Vector of Curation." Thesis, KTH, Skolan för datavetenskap och kommunikation (CSC), 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-210252.

Full text

Abstract:

Information Filtering and Recommender Systems have been used and has been implemented in various ways from various entities since the dawn of the Internet, and state-of-the-art approaches rely on Machine Learning and Deep Learning in order to create accurate and personalized recommendations for users in a given context. These models require big amounts of data with a variety of features such as time, location and user data in order to find correlations and patterns that other classical models such as matrix factorization and collaborative filtering cannot. This thesis researches, implements and compares a variety of models with the primary focus of Machine Learning and Deep Learning for the task of music recommendation and do so successfully by representing the task of recommendation as a multi-class extreme classification task with 100 000 distinct labels. By comparing fourteen different experiments, all implemented models successfully learn features such as time, location, user features and previous listening history in order to create context-aware personalized music predictions, and solves the cold start problem by using user demographic information, where the best model being capable of capturing the intended label in its top 100 list of recommended items for more than 1/3 of the unseen data in an offine evaluation, when evaluating on randomly selected examples from the unseen following week.
Informationsfiltrering och rekommendationssystem har använts och implementeratspå flera olika sätt från olika enheter sedan gryningen avInternet, och moderna tillvägagångssätt beror påMaskininlärrning samtDjupinlärningför att kunna skapa precisa och personliga rekommendationerför användare i en given kontext. Dessa modeller kräver data i storamängder med en varians av kännetecken såsom tid, plats och användardataför att kunna hitta korrelationer samt mönster som klassiska modellersåsom matris faktorisering samt samverkande filtrering inte kan. Dettaexamensarbete forskar, implementerar och jämför en mängd av modellermed fokus påMaskininlärning samt Djupinlärning för musikrekommendationoch gör det med succé genom att representera rekommendationsproblemetsom ett extremt multi-klass klassifikationsproblem med 100000 unika klasser att välja utav. Genom att jämföra fjorton olika experiment,så lär alla modeller sig kännetäcken såsomtid, plats, användarkänneteckenoch lyssningshistorik för att kunna skapa kontextberoendepersonaliserade musikprediktioner, och löser kallstartsproblemet genomanvändning av användares demografiska kännetäcken, där den bästa modellenklarar av att fånga målklassen i sin rekommendationslista medlängd 100 för mer än 1/3 av det osedda datat under en offline evaluering,när slumpmässigt valda exempel från den osedda kommande veckanevalueras.

APA, Harvard, Vancouver, ISO, and other styles

27

Smith, Matthew Scott. "Implicit Affinity Networks." BYU ScholarsArchive, 2007. https://scholarsarchive.byu.edu/etd/1112.

Full text

Abstract:

Although they clearly exist, affinities among individuals are not all easily identified. Yet, they offer unique opportunities to discover new social networks, strengthen ties among individuals, and provide recommendations. We propose the idea of Implicit Affinity Networks (IANs) to build, visualize, and track affinities among groups of individuals. IANs are simple, interactive graphical representations that users may navigate to uncover interesting patterns. This thesis describes a system supporting the construction of IANs and evaluates it in the context of family history and online communities.

APA, Harvard, Vancouver, ISO, and other styles

28

Langelaar, Johannes, and Mattsson Adam Strömme. "Federated Neural Collaborative Filtering for privacy-preserving recommender systems." Thesis, Uppsala universitet, Avdelningen för systemteknik, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-446913.

Full text

Abstract:

In this thesis a number of models for recommender systems are explored, all using collaborative filtering to produce their recommendations. Extra focus is put on two models: Matrix Factorization, which is a linear model and Multi-Layer Perceptron, which is a non-linear model. With an additional purpose of training the models without collecting any sensitive data from the users, both models were implemented with a learning technique that does not require the server's knowledge of the users' data, called federated learning. The federated version of Matrix Factorization is already well-researched, and has proven not to protect the users' data at all; the data is derivable from the information that the users communicate to the server that is necessary for the learning of the model. However, on the federated Multi-Layer Perceptron model, no research could be found. In this thesis, such a model is therefore designed and presented. Arguments are put forth in support of the privacy preservability of the model, along with a proof of the user data not being analytically derivable for the central server. In addition, new ways to further put the protection of the users' data on the test are discussed. All models are evaluated on two different data sets. The first data set contains data on ratings of movies and is called MovieLens 1M. The second is a data set that consists of anonymized fund transactions, provided by the Swedish bank SEB for this thesis. Test results suggest that the federated versions of the models can achieve similar recommendation performance as their non-federated counterparts.

APA, Harvard, Vancouver, ISO, and other styles

29

Sidana, Sumit. "Systèmes de recommandation pour la publicité en ligne." Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAM061/document.

Full text

Abstract:

Cette thèse est consacrée à l’étude des systèmes de recommandation basés sur des réseaux de neurones artificiels appris pour faire de l'ordonnancement de produits avec des retours implicites (sous forme de clics). Dans ce sens, nous proposons un nouveau modèle neuronal qui apprend conjointement la représentation des utilisateurs et des produits dans un espace latent, ainsi que la relation de préférence des utilisateurs sur les produits. Nous montrons que le modèle proposé est apprenable au sens du principe de la minimisation du risque empirique et performant par rapport aux autres modèles de l'état de l'art sur plusieurs collections. En outre, nous contribuons à la création de deux nouvelles collections, produites grâce aux enregistrements des comportements de clients de Kelkoo (https://www.kelkoo.com/); le leader européen de la publicité programmatique et de Purch (http://www.purch.com/). Les deux jeux de données recueillent des retours implicites des utilisateurs sur des produits, ainsi qu’un grand nombre d'informations contextuelles concernant à la fois les clients et les produits. La collections de données de Purch contient en plus une information sur la popularité des produits ainsi que des commentaires textuelles associés. Nous proposons, une stratégie simple et efficace sur la manière de prendre en compte le biais de la popularité ainsi qu'un modèle probabiliste latent temporel pour extraire automatiquement les thèmes des textes des commentaires.Mots clés. Systèmes de recommandation, apprentissage d'ordonnancement, réseaux de neurones, recommandations avec des retours implicites, Modèles probabilistes latents temporels
This thesis is dedicated to the study of Recommendation Systems for implicit feedback (clicks) mostly using Learning-to-rank and neural network based approaches. In this line, we derive a novel Neural-Network model that jointly learns a new representation of users and items in an embedded space as well as the preference relation of users over the pairs of items and give theoretical analysis. In addition we contribute to the creation of two novel, publicly available, collections for recommendations that record the behavior of customers of European Leaders in eCommerce advertising, Kelkoofootnote{url{https://www.kelkoo.com/}} and Purchfootnote{label{purch}url{http://www.purch.com/}}. Both datasets gather implicit feedback, in form of clicks, of users, along with a rich set of contextual features regarding both customers and offers. Purch's dataset, is affected by popularity bias. Therefore, we propose a simple yet effective strategy on how to overcome the popularity bias introduced while designing an efficient and scalable recommendation algorithm by introducing diversity based on an appropriate representation of items. Further, this collection contains contextual information about offers in form of text. We make use of this textual information in novel time-aware topic models and show the use of topics as contextual information in Factorization Machines that improves performance. In this vein and in conjunction with a detailed description of the datasets, we show the performance of six state-of-the-art recommender models.Keywords. Recommendation Systems, Data Sets, Learning-to-Rank, Neural Network, Popularity Bias, Diverse Recommendations, Contextual information, Topic Model

APA, Harvard, Vancouver, ISO, and other styles

30

Sahay, Saurav. "Socio-semantic conversational information access." Diss., Georgia Institute of Technology, 2011. http://hdl.handle.net/1853/42855.

Full text

Abstract:

The main contributions of this thesis revolve around development of an integrated conversational recommendation system, combining data and information models with community network and interactions to leverage multi-modal information access. We have developed a real time conversational information access community agent that leverages community knowledge by pushing relevant recommendations to users of the community. The recommendations are delivered in the form of web resources, past conversation and people to connect to. The information agent (cobot, for community/ collaborative bot) monitors the community conversations, and is 'aware' of users' preferences by implicitly capturing their short term and long term knowledge models from conversations. The agent leverages from health and medical domain knowledge to extract concepts, associations and relationships between concepts; formulates queries for semantic search and provides socio-semantic recommendations in the conversation after applying various relevance filters to the candidate results. The agent also takes into account users' verbal intentions in conversations while making recommendation decision. One of the goals of this thesis is to develop an innovative approach to delivering relevant information using a combination of social networking, information aggregation, semantic search and recommendation techniques. The idea is to facilitate timely and relevant social information access by mixing past community specific conversational knowledge and web information access to recommend and connect users with relevant information. Language and interaction creates usable memories, useful for making decisions about what actions to take and what information to retain. Cobot leverages these interactions to maintain users' episodic and long term semantic models. The agent analyzes these memory structures to match and recommend users in conversations by matching with the contextual information need. The social feedback on the recommendations is registered in the system for the algorithms to promote community preferred, contextually relevant resources. The nodes of the semantic memory are frequent concepts extracted from user's interactions. The concepts are connected with associations that develop when concepts co-occur frequently. Over a period of time when the user participates in more interactions, new concepts are added to the semantic memory. Different conversational facets are matched with episodic memories and a spreading activation search on the semantic net is performed for generating the top candidate user recommendations for the conversation. The tying themes in this thesis revolve around informational and social aspects of a unified information access architecture that integrates semantic extraction and indexing with user modeling and recommendations.

APA, Harvard, Vancouver, ISO, and other styles

31

Yang, Cheng-Kun, and 楊政錕. "Intelligent Recommendation System For Weight Control." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/58808999379688830836.

Full text

Abstract:

碩士
育達商業科技大學
資訊管理所
98
Taiwan in recent years, set off to lose weight boom, major celebrity publicly admitted losing weight. Many people lose weight because obesity does not look good. But in fact, the appearance of unsightly fat is one of the issues. Health threat of obesity is most worrying. There were many medical studies confirm that obesity increases disease morbidity and mortality. Health hazards of obesity are very broad, from head to toe will be affected by obesity. Medical research found that obesity may increase the incidence of cancer, such as female breast cancer, endometrial cancer, men suffering from prostate cancer, and obesity may be related to the rising incidence of cancer. The rising incidence of cancer and obesity can not avoid eating habits. Therefore, diet control and weight control must be in order to achieve the most effective weight control, and can minimize the burden on the body. In this study is to design an intelligent recommendation system. The method is based on the environmental conditions owned by the user, immediately and accurately recommends the best weight control for the user.

APA, Harvard, Vancouver, ISO, and other styles

32

Huang, Jhen-Gang, and 黃振綱. "Implementation of Intelligent Recommendation Learning Service System." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/37703725679706553554.

Full text

Abstract:

碩士
國立臺中教育大學
數位內容科技學系碩士班
97
As information overload becomes more severe, search agent can no longer fulfill the needs of all users, and this is the beginning of the emergence of recommendation mechanism. In the early days, recommendation mechanism was mostly applied to commercial behavior. Under the same background of information overload, e-Learning expands gradually, and e-Learning materials prevail on the Internet. This is a time when recommendation learning comes into exist. As technology matures, a large amount of various teaching materials have been digitalized and uploaded to the Internet in order for learners to convenient gaining knowledge of other fields. Past studies on recommendation learning mainly discussed how learners could access more accurate recommendations. This type of single direction recommendation cannot satisfy the diversified learning environments at present. Therefore, this research proposes an intelligent recommendation learning service system for learners to learn in different fields, which can provide fast and accurate recommendations to learners searching for information in their new fields or interests of learning. There are three features in our study. First, we implement the online recommendation system used the technique which combine the explicit rating and implicit rating for learner characteristics. The novel mechanism called Intelligent Recommendation Learning Service was proposed and improved the TOP-N recommendation rule and association recommendation rule. Secondly, the system used the standard SCORM and integrated relative valued functions. Third, we did an experiment to prove our hypotheses. There were three phases in this experiment, and then using questionnaire was processed for learners. After analyzing, the results of the study showed that learners accepted Intelligent Recommendation Learning Service System. The factor of recommendation accuracy was effected system stability and operation interface satisfaction, and the relevance of teaching materials was effected system stability and operation interface satisfaction, and system acceptance was effected recommendation accuracy and the relevance of teaching materials, and the factor of recommendation accuracy was effected the relevance of teaching materials.

APA, Harvard, Vancouver, ISO, and other styles

33

Wu, Wei Ying, and 吳威穎. "An Intelligent Recommendation System for Personalization Hairstyle." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/12051275344496767239.

Full text

Abstract:

碩士
國立屏東科技大學
資訊管理系所
101
In pace with the rapid development of information technology, image recognition technology are widely used in all walks of life, in beauty design-related industries order enhance their own competitiveness to a new status. Therefore, we propose a system structure that according user’s experience to design intuitive operating user interface. We can use computer to achieve virtual hairstyles replace and hairstyles recommended by image recognition technology that can reduce time and cost spend. The study module build a recommended hairstyle system to assist users select the most suitable hairstyle, the system provides functions to user：(1) Module of create a hairstyle samples—the hairstyle part is training images by skin-color detection algorithm, then extracted the component belongs the part of hairstyle only, provides users a virtual hairstyle replace. (2) Module of Create a facial shape samples—facial part is the use of active shape models algorithm training face images, then extract facial to component. The module matching with the template, then we completed the facial classification. (3)Module of Create a recommended system—using the data by users returned, after data analysis, the data can be used to a personal recommendation rules, allows users to according different hairstyle conditions to find the most suitable hairstyle. Finally, this study can setting facial and hairstyle components edge feature, for facial and hairstyle components setting synthetic match to achieve virtual synthesis technology. For facial and hairstyle components conduct synthesis match so as to achieve the effect of virtual synthesis

APA, Harvard, Vancouver, ISO, and other styles

34

林苡良. "Intelligent Recommendation Methodology and System for Patent Search." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/22003738133300676070.

Full text

Abstract:

碩士
國立清華大學
工業工程與工程管理學系
99
Due to the intangible assets have been concerned by enterprise continuously year by year, the importance of patent management also has increased. The patent search is the initial process for the patent management. The engineer of the enterprise search and collect the patents from the related patent databases, to study them for drawing out the R&D policy. However, this work causes great human-cost consuming and time-exhausted. It is not easy for users to grasp the core technology key phrases and correct search method. Therefore, this research develops an intelligent patent search and recommendation system, which provides a basic searching engine and management modules to analyze the behavior records of users and conclude their operating habits. This research collects relative patents filtering by patents’ bibliography data. Afterward, this research clusters the users and finds each user’s neighbors based on collaborative filtering mechanism. When user searches, the proposed system recommends user some appropriate patents which are inferred by his/her neighbors’ operation records of different patents to help user get more potential information. When user is drawing out their R&D policy or doing relative patent analysis, the recommend module can give user more comprehensive and useful information, helping them reduce the R&D cost. Finally, the research uses Copper Indium Gallium Selenide (CIGS) thin-film solar cell as the case study. We discuss the industry researchers or analysts’ patent management operating mode by collecting and analyzing their behavior records and provide the service for patent recommendation. The case study shows and verifies the practical value of the proposed methodology and system.

APA, Harvard, Vancouver, ISO, and other styles

35

Liu, Chi-Yun, and 劉季昀. "Effective Parking Recommendation Service for Intelligent Vehicular Guiding System." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/68620940219609724605.

Full text

Abstract:

碩士
國立中央大學
資訊工程研究所
99
In this paper, an effective parking recommendation service for vehicular intelligent guiding system is provided with real-time parking lot guiding service for future green city. The system shows drivers a best parking lot recommendation sequence and saves drivers’ time circling around by the accurate prediction of parking probability in each parking lot. The cost model containing parking probability as a factor for the best recommendation sequence is constructed according to the main conditions for parking. Through the collection and analysis of real data from parking lots in Taipei city, an integrated algorithm is developed to estimate the parking probability by the condition of current available spaces and parking popularity. Comparative experiments are performed to verify the improvement of prediction algorithm.

APA, Harvard, Vancouver, ISO, and other styles

36

Chih-LunChou and 周智倫. "Intelligent Multimedia Content Sharing and Recommendation System in Mobile Environments." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/16317623815215890529.

Full text

Abstract:

博士
國立成功大學
資訊工程學系碩博士班
100
This thesis proposes a novel approach to clustering the interests of mobile users, increasing the lifetime of interest groups, and increasing the throughput in mobile user-to-mobile user (M2M) environments for mobile IPTV (MOTV) societies. This thesis develops an interest ontology of cellular automata (CA) clustering using the zone of interest (ZOI) for mobicast communications in mobile ad hoc network (MANET) environments. The key to the proposed method is to integrate CA clustering with the ontology of users’ interests. This thesis proposes that both an interest profile (ontology) of users and information about mobile devices can help form a group of MANET-related interests. The current study evaluates the performance of the approach by conducting computer simulations. Simulation results reveal the strengths of the proposed CA-clustering algorithm in terms of increased group lifetime and increased ZOI throughput for MANETs. IEEE 802.16 WiMAX is a rapidly developing technology for broadband wireless access systems. The IEEE 802.16 MAC layer defines two operational modes, point-to-multipoint (PMP) mode and mesh mode. In the centralized protocol, all resources are controlled by base station (BS). In this thesis, we propose a novel two-stage scheme for constructing an effective multicast tree. The first stage applies a significance-based algorithm to find suitable multicast points and construct effective multicast sub-trees. The second stage applies an interference-aware Steiner tree to connect the source to each multicast sub-tree. Finally, an algorithm generates the final multicast tree topology. Simulation results reveal that the proposed approach outperforms others in the construction of a multicast tree and significantly reduces the interference of a mesh network. Recommender systems provide strategies that help users search or make decisions within the overwhelming information spaces nowadays. They have played an important role in various areas such as e-commerce and e-learning. In this thesis, we propose a hybrid recommendation strategy of content-based and knowledge-based methods that are flexible for any field to apply. By analyzing the past rating records of every user, the system learns the user’s preferences. After acquiring users’ preferences, the semantic search-and-discovery procedure takes place starting from a highly rated item. For every found item, the system evaluates the Interest Intensity indicating to what degree the user might like it. Recommender systems train a personalized estimating module using a genetic algorithm for each user, and the personalized estimating model helps improve the precision of the estimated scores. With the recommendation strategies and personalization strategies, users may have better recommendations that are closer to their preferences. In the latter part of this thesis, a real-world case, a movie-recommender system adopting proposed recommendation strategies, is implemented.

APA, Harvard, Vancouver, ISO, and other styles

37

Ou, Po-Wen, and 歐博文. "MOOCIRS: A MOOC Intelligent Recommendation System Based on Learning Diagnosis." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/9eyxru.

Full text

APA, Harvard, Vancouver, ISO, and other styles

38

Yang, Ming Jui, and 楊明瑞. "An Intelligent Mobile Ordering Recommendation System - A Case Study of Breakfast Store." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/ha2zh2.

Full text

Abstract:

碩士
景文科技大學
電子工程系電腦與通訊碩士班
104
Breakfast Store is one of the competitiveness career in Taiwan. There are about twenty four breakfast brand franchisees, and over 10 thousands stores so far. Their number and density are more than convenience store (e.g. 7-11, Welcome, Family). Recently, technology develop fast, and order the meal not only use human but also use the computer. Until Now, the mobile and online service are becoming popular, and more convenience than computer, so it will be a potential system in the feature. The thesis is design an outline system. For customer who using Smart Mobile Devices to order the meal in online and through the Hybrid Filtering system to provide personal favor, interesting food to achieve personalize and fast method to order the meal. For store, can use APP replace POS system to receive the order from Smart Mobile Device. This method can create many benefits which include decease equipment, easy arrange the working time, customer information, and decrease the mistake. In the summary, we use Intelligent Mobile Ordering Recommendation System to provide people a convenience method to enjoy the life.

APA, Harvard, Vancouver, ISO, and other styles

39

Eis, Martin Leroyce. "An intelligent chemical recommendation and applicator control system for site-specific crop management." 1989. http://hdl.handle.net/2097/22479.

Full text

APA, Harvard, Vancouver, ISO, and other styles

40

Chiu, Chuan-Feng, and 邱川峰. "An Integrated Negotiation and Recommendation System based on Intelligent Software Agent in Electronic Commerce." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/23629652347487748222.

Full text

Abstract:

博士
淡江大學
資訊工程學系
90
The Internet has become a popular medium for information exchange and knowledge delivery. Traditional buying and selling activities also follow the trend, therefore the commerce activities have moved to electronic commerce. However, even with the advents of the World Wide Web, online merchants must know what users want and to increase the sales revenue. So, providing recommendation services is an important strategy for the merchants. We analyze users’ on-line behavior and interests, and recommend to them new or potential products and the analysis mechanism is based on the correlation among customers, product items, and product features. On the other hand the negotiation is another issue in electronic commerce. When parties participate the business activities, they would negotiate with each other to make the best gain with their purpose. We use the multi-attribute utility theory to design the negotiation process. Hence, we combine the negotiation and recommendation service and propose the integrated system to serve users. But these kinds of task are hard to process for users, so we take advantage of intelligent software agent technology to develop the system and the integrated system will play an important application in the future commerce activities.

APA, Harvard, Vancouver, ISO, and other styles

41

Lee, Yi-Chi, and 李宜錡. "Integration of Game and Multi-Agent Theory on Intelligent Recommendation System of Organic Vegetables Planting." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/90514166662004915888.

Full text

Abstract:

碩士
育達商業技術學院
資訊管理所
96
Because the people are fastidious more and more regarding the diet healthy demand, therefore has accomplished the organic agriculture starting. But the majority farmers when are engaged in the organic planter, does the regular session faced with - what need to plant to a question to be only then good. Therefore, how effectiveness, and chooses the appropriate planter crops precisely, is direction which is worth discussing. In addition, in all appropriate planter's crops, considered that its relative economic value, takes in various seasons should basis of the priority selection, achieves goal of the entire year planter most greatly economic profit, is also another ponder topic. Taking organic vegetables farming as an example, this research uses knowledge-based and rule-based methods, while applying the game theory and multi-agent theory, this study develops a set of graphic intellectual suggestion mechanism with ASP.NET and MS-SQL. In the first stage of the study, we apply the knowledge base and the rule base composed for this study, we filter the suitable crops for each season, and order the list of crops in the order of suitability before we propose the planting suggestion for the entire year. Next, we design a realistic game theory and multi-agent theory to operate a negotiation process for a more effective system, which considers the organic plantations’ affect between each crop and the limitation of the system, as well as the crop shifting cost. In the end, we construct a multi-agent game theory of negotiation in order to analyze the maximum profit and propose a one year with a maximum profit. In this system, a merge of game theory and multi-agent system has been tested and verified to give suggestions that are 84.25% as effective, compared to the suggestions provided by human professionals. Other than this, the system’s greatest contribution is that, the mechanism may act as a front system of e-learning application, thus increase the level of the organic farming techniques. This research because of knowledge library and pattern union breaks original carries on the appraisal, the decision scheme recommendation system model purely. Besides domain knowledge knowledge library and model let the user in face several possibilities in the choices, provides is more objective a more effective suggestion. In the furture, it is expected to be applied to other crops planting suggestion.

APA, Harvard, Vancouver, ISO, and other styles

42

Yun-TzLee and 李韻紫. "ITRS: An Intelligent Touring Recommendation System over the Mobile Social Network Using the Cloud Computing Technique." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/89277506530310971514.

Full text

Abstract:

碩士
國立成功大學
資訊工程學系
102
Although many touring recommendation systems work with mobile social network, the delivered posts among users are still not to be utilized well for estimating user’s travel preferences. Besides, with the increasing of users in touring recommendation systems, using the parallel cloud computing can help save executing time and computing resources. However, the processing of estimating user’s travel preferences results in the severe redundancy problem [7][8], because when one user updates his/her travel experiences, the social relationship among all users will be changed synchronously. In this thesis, a cloud-based touring recommendation system with mobile social network, i.e., Intelligent Touring Recommending System (ITRS) is proposed, to recommend users suitable Point of Interests (PoIs) according to users’ travel preferences. Users are classified into meta-groups by analyzing the blogs through the proposed semi-structured user interface and ontology. Besides, the meta-groups mechanism is combined with the cloud computing model to reduce the redundancy problem existed in executing. In the experiment, results demonstrates the trend of classifying meta-groups results. Moreover, the experiment results also have shown the better efficiency of the proposed cloud model, for which the proposed cloud model still keeps the group’s internal similarity as good as the traditional ones.

APA, Harvard, Vancouver, ISO, and other styles

43

Kuo, Ting-Huan, and 郭庭歡. "The Design and Implementation of an Intelligent Personalization Food Service Recommendation System Based on Semantic Sensor Web - The Case of Hypertension, Hyperglycemia, and Hyperlipemia." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/19161411350855835013.

Full text

Abstract:

碩士
國立屏東科技大學
資訊管理系所
98
With the change of diet habit and lifestyle in Taiwan, the quantity of chronic disease patients is becoming more and more, especially of hypertension, hyperglycemia and hyperlipemia patients. However, the Food Service Recommendation (FSR) mechanism based on user’s vital data and health records has not been investigated. In this paper, we propose the Intelligent Personalization Food Service Recommendation System (IPFSRS) which contains Vital Sensor Web Layer (VSWL), Semantic Medical Web Layer (SMWL), and Medical Service Present Layer (MSPL). The vital sensors in VSWL can sense and transfer user’s vital data based on Sensor Web Enablement (SWE). The SMWL uses Rule-Based Reasoning (RBR) and Domain Ontologies (DO) based on Semantic Web (SW) to infer user’s health state according to user’s vital data from VSWL. Furthermore, we use Bayesian Classification (BC) to predict the future user’s health state. Finally, the FSR is generated according to current and future user’s health state and showed in MSPL for healthcare.

APA, Harvard, Vancouver, ISO, and other styles

44

Chiu, Ching-Yueh, and 邱敬越. "An Intelligent System for POIs Recommendations on Mobile Devices." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/48063743491665265389.

Full text

Abstract:

碩士
淡江大學
電機工程學系碩士班
103
Due to the increasing computational capability of mobile device, mature cloud service and prevalent internet infrastructure, mobile application service is now become a part of our convenient life. However the big data issue will come from the increasing data with these services. For example, user usually spend lots of time searching the information of attraction which match their demand with mobile device during traveling. If the interface on mobile device is in small size, how to provide the information of attraction to user accurately becomes a significant issue. Therefore, in this research, we collect the information of attraction from Facebook and store the information into database with preformed metadata format. Considering the processing efficiency issue caused by increasing information of attraction, we build our database on distributed computing platform, i.e., cloud. In this research, we analyze, select and sort the user’s information which was sent from mobile device and finally we send back the information of attraction which are suitable for the current scenario to mobile user for browsing.

APA, Harvard, Vancouver, ISO, and other styles

45

Kuo, Tai-Liang, and 郭泰良. "A Reviewer Recommendation System based on Collaborative Intelligence." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/62225427013974464416.

Full text

Abstract:

碩士
國立臺灣科技大學
資訊工程系
97
In this thesis, expert-finding problem is transformed to a classification issue. We build a knowledge database to represent the expertise characteristic of domain from web information constructed by collaborative intelligence, and an incremental learning method is proposed to update the database. Furthermore, results are ranked by measuring the correlation in the concept network from online encyclopedia. In our experiments, we use the real world dataset which comprise 2,701 experts who are categorized into 8 expertise domains. Our experimental results show that the expertise knowledge extracted from collaborative intelligence can improve efficiency and effect of classification and increase the precision of ranking expert at least 20%.

APA, Harvard, Vancouver, ISO, and other styles

46

Wu, Wei-chih, and 吳偉誌. "The Study of Mobile Intelligence Tourism Recommendation System." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/54060474423201903207.

Full text

Abstract:

碩士
國立高雄第一科技大學
資訊管理研究所
100
Due to circumstance of the two days holiday, people gradually pay more attention to the leisure time and tourism. Besides the indoor entertainments, more and more people are willing to stay with the natural beauty. The public pays more attention to the demand and the quality of tourism day by day. Relatively, public and private departments’ approaches to manage tourism are also changing and evolving. In tradition, most of countrymen’s ways of touring and information applying are the combination of plan maps, maps of direct mails, and GPS guiding systems which help to search tourist routes. The analysis of this study will base on current tourism systems as well as perform the most parity Android mobile platform with Google Map and GPS systems to design a complete and solid intelligent mobile tourism recommendation system which allows users to plan their topical or regional itineraries. The innovative system functions would also provide considerate services during tours to let users achieve tourist goals safely and successfully. The main point of this study will also build a database of tourist attractions and cultural orientation of the east coast of Taiwan to introduce the beauty of East coast. It allows the public to sense the spectacular scenery of coastal and the cultures much deeper. The implementation of the intelligent mobile tourism recommendation system combines IT technologies and Tourism, which not only promotes the willingness of the people to join tours and the quality of tours, but also propels the prosperity of Taiwan tourism industry.

APA, Harvard, Vancouver, ISO, and other styles

47

Bhaduri, Sashmit B. "sALERT : an intelligent information alerting and notification web service." Thesis, 2012. http://hdl.handle.net/2152/ETD-UT-2012-05-5870.

Full text

Abstract:

Web services increasingly serve as large repositories and conduits of information. However, they do not always allow for the efficient dissemination of this information, particularly in a reactive way. In this report, I describe sALERT, a web-based application that allows for targeted information from various web services to be combined and cross-referenced in order to produce a system that is more convenient and more efficient in reactively disseminating information. This dissemination is performed using mobile notification mechanisms such as text messages, and information targeting is performed using data from social networks and geolocation sources. I present the design, implementation, and plans for future improvement for this service within this report.
text

APA, Harvard, Vancouver, ISO, and other styles

48

Κόρδαρης, Ιωάννης. "Η αντιμετώπιση της πληροφοριακής υπερφόρτωσης ενός οργανισμού με χρήση ευφυών πρακτόρων." Thesis, 2014. http://hdl.handle.net/10889/7965.

Full text

Abstract:

Η πληροφοριακή υπερφόρτωση των χρηστών αποτελεί βασικό πρόβλημα ενός οργανισμού. Η συσσώρευση μεγάλου όγκου πληροφορίας στα πληροφοριακά συστήματα, προκαλεί στους χρήστες άγχος και υπερένταση, με αποτέλεσμα να δυσχεραίνει την ικανότητά τους για λήψη αποφάσεων. Λόγω αυτού, η επίδραση της πληροφοριακής υπερφόρτωσης στους οργανισμούς είναι καταστροφική και απαιτείται η αντιμετώπισή της. Υπάρχουν διάφοροι τρόποι αντιμετώπισης της πληροφοριακής υπερφόρτωσης όπως τα συστήματα υποστήριξης λήψης αποφάσεων, τα συστήματα φιλτραρίσματος πληροφορίας, οι αποθήκες δεδομένων και άλλες τεχνικές της εξόρυξης δεδομένων και της τεχνητής νοημοσύνης, όπως είναι οι ευφυείς πράκτορες. Οι ευφυείς πράκτορες αποτελούν εφαρμογές που εφάπτονται της τεχνικής νοημοσύνης, οι οποίες έχουν την ικανότητα να δρουν αυτόνομα, συλλέγοντας πληροφορίες, εκπαιδεύοντας τον εαυτό τους και επικοινωνώντας με τον χρήστη και μεταξύ τους. Συχνά, υλοποιούνται πολυπρακτορικά συστήματα προκει-μένου να επιλυθεί ένα πρόβλημα του οργανισμού. Στόχος τους είναι να διευκολύνουν τη λήψη αποφάσεων των χρηστών, προτείνοντας πληροφορίες βάσει των προτιμήσεών τους. Ο σκοπός της παρούσας διπλωματικής εργασίας είναι να αναλύσει σε βάθος τους ευφυείς πράκτορες, σαν μία αποτελεσματική μέθοδο αντιμετώπισης της πληροφοριακής υπερφόρτωσης, να προτείνει πειραματικούς πράκτορες προτά-σεων και να εξετάσει επιτυχημένες υλοποιήσεις. Συγκεκριμένα, παρουσιάζεται ένα ευφυές σύστημα διδασκαλίας για την ενίσχυση του e-Learning/e-Teaching, προτείνεται ένα σύστημα πρακτόρων για τον οργανισμό Flickr, ενώ εξετάζεται το σύστημα προτάσεων του Last.fm και ο αλγόριθμος προτάσεων του Amazon. Τέλος, αναλύεται μια πειραματική έρευνα ενός ευφυούς πράκτορα προτάσεων, ο οποίος αντιμετώπισε με επιτυχία την αντιληπτή πληροφοριακή υπερφόρτωση των χρηστών ενός θεωρητικού ηλεκτρονικού καταστήματος. Τα αποτελέσματα του πειράματος παρουσίασαν την επίδραση της αντιληπτής πληροφοριακής υπερφόρτωσης και του φορτίου πληροφορίας στην ποιότητα επιλογής, στην εμπιστοσύνη επιλογής και στην αντιληπτή αλληλεπίδραση μεταξύ ηλεκτρονικού καταστήματος και χρήστη, ενώ παρατηρήθηκε η καθοριστική συμβολή της χρήσης των ευφυών πρακτόρων στην αντιμετώπιση της πληροφοριακής υπερφόρτωσης.
-

APA, Harvard, Vancouver, ISO, and other styles

49

Rahmaniazad, Emad. "Community Recommendation in Social Networks with Sparse Data." Thesis, 2020. http://hdl.handle.net/1805/24760.

Full text

Abstract:

Indiana University-Purdue University Indianapolis (IUPUI)
Recommender systems are widely used in many domains. In this work, the importance of a recommender system in an online learning platform is discussed. After explaining the concept of adding an intelligent agent to online education systems, some features of the Course Networking (CN) website are demonstrated. Finally, the relation between CN, the intelligent agent (Rumi), and the recommender system is presented. Along with the argument of three different approaches for building a community recommendation system. The result shows that the Neighboring Collaborative Filtering (NCF) outperforms both the transfer learning method and the Continuous bag-of-words approach. The NCF algorithm has a general format with two various implementations that can be used for other recommendations, such as course, skill, major, and book recommendations.

APA, Harvard, Vancouver, ISO, and other styles

50

Guo, Song-Jie, and 郭松杰. "The Application of Artificial Intelligence in the Field of Talent Selection — Construction of Resume Recommendation System." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/4bz2ay.

Full text

Abstract:

碩士
國立臺灣科技大學
企業管理系
107
This research employs the Natural Language Processing (NLP) technology, which involves the application of text mining. The experimental process uses python for programming. This technique aims to analyze the resume data and construct the resume recommendation system. The entire research divides into two experiments. The pilot experiment simulates the actual resume selection process. The pilot experiment involves comparing the results of ‘Simulated Supervisor’ with the resume recommendation system and ‘Simulated HR’ respectively. The formal experiment focuses on comparing the results of the automated selection of the system's resume with the actual recruitment status and calculating the inter-rater agreement. In the pilot experiment, 83.36% of the ‘System Review’ results are close to the order of ‘Simulated Supervisor’ than the ‘Simulated HR’. In the formal experiment, the result of resume recommendation system tallies with the actual enterprise's manual resume selection status on average up to 80.8%. This research focuses on personal preference in the way of resume selection and recommendation, which is quite different from the previous related research. The system will satisfy many different selection conditions in the enterprise flexibly and quickly. In the selection process, candidates can be selected for different job vacancies. This is a person-job fit recommendation, it can achieve the right fit and assist candidates to match more suitable job choices.

APA, Harvard, Vancouver, ISO, and other styles

Dissertations / Theses on the topic 'Intelligent recommendation system'

Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles