Rozprawy doktorskie na temat „CONTEXT-AWARE RECOMMENDER”

Recommender systems are software tools and techniques providing suggestions and recommendations for items to be of use to a user. These sug- gestions can help users make better decisions on choosing products or services, such as which film to watch, what music to listen to or which travel insurance to buy. When making suggestions, many recommender systems do not consider contextual information, such as location or time [5]. Recommender systems that make use of contextual information are called context-aware recommender systems. Many context-aware recommender systems can not generate reliable rec- ommendations on sparse data. Besides, in most context-aware recommender systems, the contexts are pre-defined and not personalised. These limitations of existing methods usually lead to inaccurate recommendations. In this thesis, new context-aware recommendation methods are presented. In these methods, personalised contexts are defined based on users’ activity patterns. The underlying associations between contexts are analysed, and similar contexts are combined so that the system can make use of existing data collected in similar contexts. Experimental results from two datasets show that the proposed methods can achieve significantly higher recommendation accuracy than existing context-aware recommendation methods.

Les systèmes de recommandation se sont révélés être des outils efficaces pour aider les utilisateurs à faire face à la surcharge informationnelle. D’importants progrès ont été réalisés dans le domaine durant les deux dernières décennies, menant en particulier à l’exploitation de l’information contextuelle pour modéliser l’aspect dynamique des utilisateurs et des articles. La définition traditionnelle du contexte, adoptée dans la plupart des systèmes de recommandation contextuels, ne répond pas à plusieurs contraintes rencontrées dans les applications du monde réel. Dans cette thèse, nous abordons les problèmes de recommandation en présence d’informations contextuelles partiellement observables et d’informations contextuelles non observables dans deux applications particulières, la recommandation d’hôtels et la recommandation en ligne, remettant en question plusieurs aspects de la définition traditionnelle du contexte, notamment l'accessibilité, la pertinence, l'acquisition et la modélisation.La première partie de la thèse étudie le problème de recommandation d’hôtels qui souffre du démarrage à froid continu, limitant la performance des approches classiques de recommandation. Le voyage n’est pas une activité fréquente et les utilisateurs ont tendance à adopter des comportements diversifiés en fonction de leurs situations spécifiques. Après une analyse du comportement des utilisateurs dans ce domaine, nous proposons de nouvelles approches de recommandation intégrant des informations contextuelles partiellement observables affectant les utilisateurs. Nous montrons comment cela contribue à améliorer la qualité des recommandations.La deuxième partie de la thèse aborde le problème de recommandation en ligne en présence de flux de données où les observations apparaissent continûment à haute fréquence. Nous considérons que les utilisateurs et les articles reposent sur des informations contextuelles non observables par le système et évoluent de façons différentes à des rythmes différents. Nous proposons alors d’effectuer de la détection active de changements et d’assurer la mise à jour des modèles en temps réel. Nous concevons de nouvelles méthodes qui s’adaptent aux changements qui apparaissent au niveau des préférences des utilisateurs et des perceptions et descriptions des articles, et montrons l’importance de la recommandation adaptative en ligne pour garantir de bonnes performances au cours du temps
Recommender systems have proven to be valuable tools to help users overcome the information overload, and significant advances have been made in the field over the last two decades. In particular, contextual information has been leveraged to model the dynamics occurring within users and items. Context is a complex notion and its traditional definition, which is adopted in most recommender systems, fails to cope with several issues occurring in real-world applications. In this thesis, we address the problems of partially observable and unobservable contexts in two particular applications, hotel recommendation and online recommendation, challenging several aspects of the traditional definition of context, including accessibility, relevance, acquisition, and modeling.The first part of the thesis investigates the problem of hotel recommendation which suffers from the continuous cold-start issue, limiting the performance of classical approaches for recommendation. Traveling is not a frequent activity and users tend to have multifaceted behaviors depending on their specific situation. Following an analysis of the user behavior in this domain, we propose novel recommendation approaches integrating partially observable context affecting users and we show how it contributes in improving the recommendation quality.The second part of the thesis addresses the problem of online adaptive recommendation in streaming environments where data is continuously generated. Users and items may depend on some unobservable context and can evolve in different ways and at different rates. We propose to perform online recommendation by actively detecting drifts and updating models accordingly in real-time. We design novel methods adapting to changes occurring in user preferences, item perceptions, and item descriptions, and show the importance of online adaptive recommendation to ensure a good performance over time

Travailler en collaboration n’est plus une question mais une réalité, la question qui se pose aujourd’hui concerne la mise en œuvre de la collaboration de façon à ce qu’elle soit la plus réussie possible. Cependant, une collaboration réussie n’est pas facile et est conditionnée par différents facteurs qui peuvent l’influencer. Il est donc nécessaire de considérer ces facteurs au sein du contexte de collaboration pour favoriser l’efficacité de collaboration. Parmi ces facteurs, le collaborateur est un facteur principal, qui est étroitement associé à l’efficacité et à la réussite des collaborations. Le choix des collaborateurs et/ou la recommandation de ces derniers en tenant compte du contexte de la collaboration peut grandement influencer la réussite de cette dernière. En même temps, grâce au développement des technologies de l’information, de nombreux outils numériques de collaboration sont mis à la disposition tels que les outils de mail et de chat en temps réel. Ces outils numériques peuvent eux-mêmes être intégrés dans un environnement de travail collaboratif basé sur le web. De tels environnements permettent aux utilisateurs de collaborer au-delà de la limite des distances géographiques. Ces derniers laissent ainsi des traces d’activités qu’il devient possible d’exploiter. Cette exploitation sera d’autant plus précise que le contexte sera décrit et donc les traces enregistrées riches en description. Il devient donc intéressant de développer les environnements de travail collaboratif basé sur le web en tenant d’une modélisation du contexte de la collaboration. L’exploitation des traces enregistrés pourra alors prendre la forme de recommandation contextuelle de collaborateurs pouvant renforcer la collaboration. Afin de générer des recommandations de collaborateurs dans des environnements de travail collaboratifs basés sur le web, cette thèse se concentre sur la génération des recommandations contextuelles de collaborateurs en définissant, modélisant et traitant le contexte de collaboration. Pour cela, nous proposons d’abord une définition du contexte de collaboration et choisissons de créer une ontologie du contexte de collaboration compte tenu des avantages de l’approche de modélisation en l’ontologie. Ensuite, une similarité sémantique basée sur l’ontologie est développée et appliquée dans trois algorithmes différents (i.e., PreF1, PoF1 et PoF2) afin de générer des recommandations contextuelles des collaborateurs. Par ailleurs, nous déployons l’ontologie de contexte de collaboration dans des environnements de travail collaboratif basés sur le web en considérant une architecture de système des systèmes d’informations du point de vue des environnements de travail collaboratif basés sur le web. À partir de cette architecture, un prototype correspondant d’environnement de travail collaboratif basé sur le web est alors construit. Enfin, un ensemble de données de collaborations scientifiques est utilisé pour tester et évaluer les performances des trois algorithmes de recommandation contextuelle des collaborateurs
Working collaboratively is no longer an issue but a reality, what matters today is how to implement collaboration so that it is as successful as possible. However, successful collaboration is not easy and is conditioned by different factors that can influence it. It is therefore necessary to take these impacting factors into account within the context of collaboration for promoting the effectiveness of collaboration. Among the impacting factors, collaborator is a main one, which is closely associated with the effectiveness and success of collaborations. The selection and/or recommendation of collaborators, taking into account the context of collaboration, can greatly influence the success of collaboration. Meanwhile, thanks to the development of information technology, many collaborative tools are available, such as e-mail and real-time chat tools. These tools can be integrated into a web-based collaborative work environment. Such environments allow users to collaborate beyond the limit of geographical distances. During collaboration, users can utilize multiple integrated tools, perform various activities, and thus leave traces of activities that can be exploited. This exploitation will be more precise when the context of collaboration is described. It is therefore worth developing web-based collaborative work environments with a model of the collaboration context. Processing the recorded traces can then lead to context-aware collaborator recommendations that can reinforce the collaboration. To generate collaborator recommendations in web-based Collaborative Working Environments, this thesis focuses on producing context-aware collaborator recommendations by defining, modeling, and processing the collaboration context. To achieve this, we first propose a definition of the collaboration context and choose to build a collaboration context ontology given the advantages of the ontology-based modeling approach. Next, an ontologybased semantic similarity is developed and applied in three different algorithms (i.e., PreF1, PoF1, and PoF2) to generate context-aware collaborator recommendations. Furthermore, we deploy the collaboration context ontology into web-based Collaborative Working Environments by considering an architecture of System of Information Systems from the viewpoint of web-based Collaborative Working Environments. Based on this architecture, a corresponding prototype of web-based Collaborative Working Environment is then constructed. Finally, a dataset of scientific collaborations is employed to test and evaluate the performances of the three context-aware collaborator recommendation algorithms

Recommender system (RS) holds a significant place in the area of the tourism sector. The major factor of trip planning is selecting relevant Points of Interest (PoI) from tourism domain. The RS system supposed to collect information from user behaviors, personality, preferences and other contextual information. This work is mainly focused on user’s personality, preferences and analyzing user psychological traits. The work is intended to improve the user profile modeling, exposing relationship between user personality and PoI categories and find the solution in constraint satisfaction programming (CSP). It is proposed the architecture according to ambient intelligence perspective to allow the best possible tourist place to the end-user. The key development of this RS is representing the model in CSP and optimizing the problem. We implemented our system in Minizinc solver with domain restrictions represented by user preferences. The CSP allowed user preferences to guide the system toward finding the optimal solutions; RESUMO O sistema de recomendação (RS) detém um lugar significativo na área do sector do turismo. O principal fator do planeamento de viagens é selecionar pontos de interesse relevantes (PoI) do domínio do turismo. O sistema de recomendação (SR) deve recolher informações de comportamentos, personalidade, preferências e outras informações contextuais do utilizador. Este trabalho centra-se principalmente na personalidade, preferências do utilizador e na análise de traços fisiológicos do utilizador. O trabalho tem como objetivo melhorar a modelação do perfil do utilizador, expondo a relação entre a personalidade deste e as categorias dos POI, assim como encontrar uma solução com programação por restrições (CSP). Propõe-se a arquitetura de acordo com a perspetiva do ambiente inteligente para conseguir o melhor lugar turístico possível para o utilizador final. A principal contribuição deste SR é representar o modelo como CSP e tratá-lo como problema de otimização. Implementámos o nosso sistema com o solucionador em Minizinc com restrições de domínio representadas pelas preferências dos utilizadores. O CSP permitiu que as preferências dos utilizadores guiassem o sistema para encontrar as soluções ideais.

In general, recommender systems attempt to estimate user preference based on historical data. A context-aware recommender system attempts to generate better recommendations using contextual information. However, generating recommendations for specific contexts has been challenging because of the difficulties in using contextual information to enhance the capabilities of recommender systems. Several methods have been used to incorporate contextual information into traditional recommendation algorithms. These methods focus on incorporating contextual information to improve general recommendations for users rather than identifying the different context applicable to the user and providing recommendations geared towards those specific contexts. In this thesis, we explore different context-aware recommendation techniques and present our context-aware coupled matrix factorization methods that use matrix factorization for estimating user preference and features in a specific contextual condition. We develop two methods: the first method attaches user preference across multiple contextual conditions, making the assumption that user preference remains the same, but the suitability of items differs across different contextual conditions; i.e., an item might not be suitable for certain conditions. The second method assumes that item suitability remains the same across different contextual conditions but user preference changes. We perform a number of experiments on the last.fm dataset to evaluate our methods. We also compared our work to other context-aware recommendation approaches. Our results show that grouping ratings by context and jointly factorizing with common factors improves prediction accuracy.

Background. Code reviewing is a commonly used practice in software development. It refers to the process of reviewing new code changes, commonly before they aremerged with the code base. However, in order to perform the review, developers need to be assigned to that task. The problems with a manual assignment includes a time-consuming selection process; limited pool of known candidates; risk of high reuse of the same reviewers (high workload). Objectives. This thesis aims to attempt to address the above issues with a recommendation system. The idea is to receive feedback from experienced developers in order to expand upon identified reviewer factors; which can be used to determinethe suitability of developers as reviewers for a given change. Also, to develop and implement a solution that uses some of the most promising reviewer factors. The solution can later be deployed and validated through user and reviewer feedback in a real large-scale project. The developed recommendation system is named Carrot. Methods. An improvement case study was conducted at Ericsson. The identification of reviewer factors is found through literature review and semi-structured interviews. Validation of Carrot’s usability was conducted through static analysis,user feedback, and static validation. Results. The results show that Carrot can help identify adequate non-obvious reviewers and be of great assistance to new developers. There are mixed opinions on Carrot’s ability to assist with workload balancing and decrease of review lead time. The recommendations can be performed in a production environment in less than a quarter of a second. Conclusions. The implemented and validated approach indicates possible usefulness in performing recommendations, but could benefit significantly from further improvements. Many of the problems seen with the recommendations seem to be a result of corner-cases that are not handled by the calculations. The problems would benefit considerably from further analysis and testing.

The amount of information generated and maintained by information systems and their users leads to the increasingly important concern of information overload. Personalized systems have thus emerged to help provide more relevant information and services to the user. In particular, recommender systems appeared in the mid 1990's and have since then generated a growing interest in both industry and academia. Besides, context-aware systems have been developed to model, capture and interpret information about the user's situation, generally in dynamic and heterogeneous environments. Decision support systems like Business Intelligence (BI) platforms also face usability challenges as the amount of information available to knowledge workers grows. Remarkably, we observe that only a small part of personalization and recommendation techniques have been used in the context of data warehouses and analysis tools. Therefore, our work aims at exploring synergies of recommender systems and context-aware systems to develop personalization and recommendation scenarios suited in a BI environment. In response to this, we develop in our work an open and modular situation management platform using a graph-based situation model. Besides, dynamic aspects are crucial to deal with context data which is inherently time-dependent. We thus define two types of active components to enable dynamic maintenance of situation graphs, activation rules and operators. In response to events which can describe users' interactions, activation rules - defined using the event-condition-action framework - are evaluated thanks to queries on underlying graphs, to eventually trigger appropriate operators. These platform and framework allow us to develop and support various recommendation and personalization scenarios. Importantly, we design a re-usable personalized query expansion component, using semantics of multi-dimensional models and usage statistics from repositories of BI documents like reports or dashboards. This component is an important part of another experimentation we realized, Text-To-Query. This system dynamically generates multi-dimensional queries to illustrate a text and support the knowledge worker in the analysis or enrichment of documents she is manipulating. Besides, we also illustrate the integration and usage of our graph repository and situation management frameworks in an open and extensible federated search project, to provide background knowledge management and personalization.

Many countries depend heavily on tourism for their economic growth. The invention of the web has opened new opportunities for tourists to discover new places and live new adventures. However, the number of possible destinations has become huge and even an entire lifespan would not be enough to visit all of these places. Even for one city, there are a significant number of possible places to visit. Nowadays, searching online to find an interesting place to visit is harder than ever, not because there is a lack of information but rather due to the vast amount of information that can be found. Trip planning is a tedious task, especially when the tourist does not want to pick a preplanned itinerary from a traveling agency. That being said, even these preplanned itineraries need a lot of time and effort to be customized. Moreover, the set of itineraries that a tourist can select from is usually limited. In addition, there may be many places that tourists would enjoy visiting but that are not included in the itineraries. Thus, static planners do not always choose the right place at the right time. This is why the planning process should take into consideration many factors in order to give the tourist the best possible suggestions. In this Thesis, we propose an algorithm called the Balanced Orienteering Problem to design trips for tourists. This algorithm, combined with a context-aware recommender system for tourism suggestions, create the infrastructure of the mobile application for the augmented reality tourism guide that we developed. We cover the background knowledge of tour planning problems and tourism recommender systems and describe the existing techniques. Furthermore, a comparison between the existing systems and our algorithm is completed to illustrate that our proposed algorithm yields better results. We also discuss the workflow of our system implementation and how our mobile application is designed. Lastly, we address suggestions for future works and end with a conclusion.

Case-based reasoning (CBR), as one of the problem solving paradigms in the field of Artificial Intelligence (AI), is an approach to the re-use of experience to solve problem. The aim of this research was to identify and evaluate existing and new approaches to elicit and formalise knowledge for context-aware systems as well as systems that are able to perform explanation-aware computing. The research was centred on systems that employ the specific AI approach of CBR. The research identified positive and negative effects of knowledge formalisation as well as synergies of knowledge formalisation for context-awareness and explanation-aware computing. The research focused on a set of specific knowledge sources such as sensors, human experts, online sources such as web communities and social media as well as a combination of these sources. A set of knowledge formalisation approaches was evaluated during the implementation of six prototype systems, representing a series of product- and work-flow recommender systems. Example domains for the systems developed include CBR-based recommendation in audio mastering, gold ore refinement and travel medicine. Test data gathered from real-world use of the prototypes formed the basis for a quantitative and qualitative analysis to establish the performance and quality of the knowledge formalisation approaches used within the prototypes development. The outcome of this research work consists of new approaches to knowledge elicitation and formalisation for expert work-flow recommender systems, new approaches to context- and explanatory-knowledge formalisation in combination with software engineering techniques, new approaches to knowledge extraction and formalisation from web sources and contributions to the further development of the myCBR 3 software, an open source software for the rapid prototyping of CBR systems.

L'accès aux informations pertinentes, adaptées aux besoins et au profil de l'utilisateur est un enjeu majeur dans le cadre actuel caractérisé par une prolifération massive des ressources d'information hétérogènes. Le développement d'appareils mobiles équipés de plusieurs fonctionnalités telles que la caméra, le WIFI, la géo-localisation et bien plus d'autres permettent aux systèmes mobiles de recommandation actuels d'être hautement contextualisés et pouvant fournir à l'utilisateur des informations pertinentes au bon moment quand il en a le plus besoin, sans attendre qu'il établisse une interaction avec son appareil. C'est dans ce cadre que s'insère notre travail de thèse. En effet, nous proposons une approche de recommandation contextuelle et proactive dans un environnement mobile qui permet de recommander des informations pertinentes à l'utilisateur sans attendre à ce que ce dernier initie une interaction. Un système proactif peut prendre la forme d'un guide touristique personnalisé qui se base sur la localisation et les préférences de l'utilisateur pour suggérer à ce dernier des endroits intéressants sans qu'il fournisse, sa préférence ou une requête explicite. Cela permettra de réduire les efforts, le temps et l'interaction de l'utilisateur avec son appareil mobile et de présenter les informations pertinentes au bon moment et au bon endroit. Cette approche prend aussi en considération les situations où la recommandation pourrait déranger l'utilisateur. Il s'agit d'équilibrer le processus de recommandation contre les interruptions intrusives. En effet, il existe différents facteurs et situations qui rendent l'utilisateur moins ouvert aux recommandations. Comme nous travaillons dans le contexte des appareils mobiles, nous considérons que les applications mobiles telles que la caméra, le clavier, l'agenda, etc., sont de bons représentants de l'interaction de l'utilisateur avec son appareil puisqu'ils représentent en quelque sorte la plupart des activités qu'un utilisateur pourrait entreprendre avec son appareil mobile au quotidien, comme envoyer des messages, converser, tweeter, naviguer ou prendre des photos
Just-In-Time recommender systems involve all systems able to provide recommendations tailored to the preferences and needs of users in order to help them access useful and interesting resources within a large data space. The user does not need to formulate a query, this latter is implicit and corresponds to the resources that match the user's interests at the right time. Our work falls within this framework and focuses on developing a proactive context-aware recommendation approach for mobile devices that covers many domains. It aims at recommending relevant items that match users' personal interests at the right time without waiting for the users to initiate any interaction. Indeed, the development of mobile devices equipped with persistent data connections, geolocation, cameras and wireless capabilities allows current context-aware recommender systems (CARS) to be highly contextualized and proactive. We also take into consideration to which degree the recommendation might disturb the user. It is about balancing the process of recommendation against intrusive interruptions. As a matter of fact, there are different factors and situations that make the user less open to recommendations. As we are working within the context of mobile devices, we consider that mobile applications functionalities such as the camera, the keyboard, the agenda, etc., are good representatives of the user's interaction with his device since they somehow stand for most of the activities that a user could use in a mobile device in a daily basis such as texting messages, chatting, tweeting, browsing or taking selfies and pictures

Recommender systems support online customers by suggesting products and services of likely interest to them. Research in recommender systems is now starting to recognise the importance of multiple selection criteria and the role of customer context in improving the recommendation output. This thesis investigates the inclusion of criteria and context information in the recommendation process. Firstly, a novel technique for multi-criteria recommendation is proposed. It assumes that some selection criteria for an item (product or a service) will dominate the overall rating, and that these dominant criteria will be different for different users. Following this assumption, users are clustered based on their criteria preferences, creating a “preference lattice”. The recommendation output for a user is then based on ratings by other users from the same or nearby clusters. Secondly, a context similarity metric for context aware recommendation is presented. This metric can help improve the prediction accuracy in two ways. On the one hand, the metric can guide the aggregation of the feedback from similar context to improve the prediction accuracy. This aggregation is important because the recommendation generation based on prior feedback by similar customers reduces the quantum of feedback used, resulting in a reduction in recommendation quality. On the other hand, the value returned by the context similarity metric can also be used to indicate the importance of the context information in the prediction process for a context aware recommendation.The validation of the two proposed techniques and their applications are conducted in the service domain because the relatively high degree of user involvement attracts users to provide detailed feedback from multiple perspectives, such as from criteria and context perspectives. In particular, hotel services and web services areas are selected due to their different levels of maturity in terms of users’ feedback. For each area, this thesis proposes a different recommendation approach by combining the proposed techniques with a traditional recommendation approach. The thesis concludes with experiments conducted on the datasets from the two aforementioned areas to evaluate the proposed techniques, and to demonstrate the process and the effectiveness of the techniques-based recommendation approaches.

In the last decade, with the extensive use of mobile electronic and wireless communication devices, there is a growing need for context aware applications and many pervasive computing applications have become integral parts of our daily lives. Context aware recommender systems are one of the popular ones in this area. Such systems surround the users and integrate with the environment
hence, they are aware of the users'
context and use that information to deliver personalized recommendations about everyday tasks. In this manner, predicting user&rsquo
s next activity preferences with high accuracy improves the personalized service quality of context aware recommender systems and naturally provides user satisfaction. Predicting activities of people is useful and the studies on this issue in ubiquitous environment are considerably insufficient. Thus, this thesis proposes an activity prediction model to forecast a user&rsquo
s next activity preference using past preferences of the user in certain contexts and current contexts of user in ubiquitous environment. The proposed model presents a new approach for activity prediction by taking advantage of ontology. A prototype application is implemented to demonstrate the applicability of this proposed model and the obtained outputs of a sample case on this application revealed that the proposed model can reasonably predict the next activities of the users.

L'immense quantité d'information générée et gérée au quotidien par les systèmes d'information et leurs utilisateurs conduit inéluctablement ?a la problématique de surcharge d'information. Dans ce contexte, les systèmes de recommandation traditionnels fournissent des informations pertinentes aux utilisateurs. Néanmoins, avec la propagation récente des dispositifs mobiles (Smartphones et tablettes), nous constatons une migration progressive des utilisateurs vers la manipulation d'environnements pérvasifs. Le problème avec les approches traditionnelles de recommandation est qu'elles n'utilisent pas toute l'information disponible pour produire des recommandations. Davantage d'informations contextuelles pourraient être utilisées dans le processus de recommandation pour aboutir à des recommandations plus précises. Les systèmes de recommandations sensibles au contexte (CARS) combinent les caractéristiques des systèmes sensibles au contexte et des systèmes de recommandation an de fournir des informations personnalisées aux utilisateurs dans des environnements ubiquitaires. Dans cette perspective ou tout ce qui concerne l'utilisateur est dynamique, les contenus qu'il manipule et son environnement, deux questions principales doivent être adressées : i) Comment prendre en compte la dynamicité des contenus de l'utilisateur ? et ii ) Comment éviter d'être intrusif en particulier dans des situations critiques ?. En réponse ?a ces questions, nous avons développé un système de recommandation dynamique et sensible au risque appelé DRARS (Dynamic Risk-Aware Recommender System), qui modélise la recommandation sensible au contexte comme un problème de bandit. Ce système combine une technique de filtrage basée sur le contenu et un algorithme de bandit contextuel. Nous avons montré que DRARS améliore la stratégie de l'algorithme UCB (Upper Con dence Bound), le meilleur algorithme actuellement disponible, en calculant la valeur d'exploration la plus optimale pour maintenir un compromis entre exploration et exploitation basé sur le niveau de risque de la situation courante de l'utilisateur. Nous avons mené des expériences dans un contexte industriel avec des données réelles et des utilisateurs réels et nous avons montré que la prise en compte du niveau de risque de la situation de l'utilisateur augmentait significativement la performance du système de recommandation.

Strömmande musiktjänster erbjuder ett stort utbud av musik. Förutom att användare kan skapa egna listor från detta utbud, erbjuder tjänsterna ofta personliga rekommendationer. Även om rekommendationerna passar användaren väl, passar de inte alltid för situationen de befinner sig i. I denna studie presenteras en artefakt i form av en kontextmedveten musikapplikation som använder Spotify för rekommendation och uppspelning av musik. En kontextmedveten musikapplikation är en applikation som tar användarens kontext i beaktning vid musikrekommendation. I denna studie refererar kontext till den situation som en potentiell användare befinner sig i, exempelvis "ute och springer i parken klockan tre". Vi presenterar en enkätundersökning om vilka kontextuella faktorer användare tycker är viktiga, och frågor kring lyssnarbeteende. Artefakten testas i en användarstudie och resultaten analyseras och diskuteras i relation till tidigare forskning. Vi ser att användare har en positiv inställning till att kontextuella faktorer påverkar vilken musik de lyssnar på, och att det finns en positiv inställning till kontextmedvetna musikapplikationer. Vidare ser vi att aktivitet är den mest relevanta kontextuella faktorn för användare.
Music streaming services offer a large quantity of music. Apart from users being able to create their own playlists, these services also offer personal music recommendations. Even though these recommendations meet the users preferences, they don't always fit the users current situation. In this study, we present an artifact in the shape of a context-aware music recommender application, that uses Spotify services to recommend and handle playback of music. A context-aware music application is an application that takes a users current context in consideration when recommending music. In this study, context-aware refers to the situation a given user might find themselves in, e.g. "jogging in the park at 3pm". We present a questionnaire about which contextual factors users think are important, and questions about listening preferences. The artifact is tested in a user study, and the results are analysed and discussed in relation to previous studies. We found that users have a positive attitude towards contextual factors influencing which music they listen to, and that there is a positive attitude towards contextual music recommenders. Furthermore we found that activity is the most relevant contextuall factor to users.

Recommender systems are helpful tools employed abundantly in online applications to help users find what they want. This thesis re-purposes a collaborative filtering recommender built for incorporating social media (hash)tags to be used as a context-aware recommender, using time of day and activity as contextual factors. The recommender uses a matrix factorization approach for implicit feedback, in a music streaming setting. Contextual data is collected from users' mobile phones while they are listening to music. It is shown in an offline test that this approach improves recall when compared to a recommender that does not account for the context the user was in. Future work should explore the qualities of this model further, as well as investigate how this model's recommendations can be surfaced in an application.

Recommender systems are an advanced form of software applications, more specifically decision-support systems, that efficiently assist the users in finding items of their interest. Recommender systems have been applied to many domains from music to e-commerce, movies to software services delivery and tourism to news by exploiting available information to predict and provide recommendations to end user. The suggestions generated by recommender systems tend to narrow down the list of items which a user may overlook due to the huge variety of similar items or users’ lack of experience in the particular domain of interest. While the performance of traditional recommender systems, which rely on relatively simpler information such as content and users’ filters, is widely accepted, their predictive capability perfomrs poorly when local context of the user and situated actions have significant role in the final decision. Therefore, acceptance and incorporation of context of the user as a significant feature and development of recommender systems utilising the premise becomes an active area of research requiring further investigation of the underlying algorithms and methodology. This thesis focuses on categorisation of contextual and non-contextual features within the domain of context-aware recommender system and their respective evaluation. Further, application of the Multilayer Perceptron Model (MLP) for generating predictions and ratings from the contextual and non-contextual features for contextual recommendations is presented with support from relevant literature and empirical evaluation. An evaluation of specifically employing artificial neural networks (ANNs) in the proposed methodology is also presented. The work emphasizes on both algorithms and methodology with three points of consideration: contextual features and ratings of particular items/movies are exploited in several representations to improve the accuracy of recommendation process using artificial neural networks (ANNs), context features are combined with user-features to further improve the accuracy of a context-aware recommender system and lastly, a combination of the item/movie features are investigated within the recommendation process. The proposed approach is evaluated on the LDOS-CoMoDa dataset and the results are compared with state-of-the-art approaches from relevant published literature.

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.

Pervasive computing systems aim to improve human-computer interaction, using users situation variables that define context. The boom of Internet makes growing availables items to choose, giving cost in made decision process. Affective Computing has in its goals to identify user s affective/emotional state in a computing interaction, in order to respond to it automatically. Recommendation systems help made decision selecting and suggesting items in scenarios where there are huge information volume, using, traditionally, users prefferences data. This process could be enhanced using context information (as physical, environmental or social), rising the Context-Aware Recommendation Systems. Due to emotions importance in our lives, that could be treated with Affective Computing, this work uses affective context as context variable, in recommendation process, proposing the Affective-Recommender a recommendation system that uses user s affective state to select and to suggest items. The system s model has four components: (i) detector, that identifies affective-state, using the multidimesional Pleasure, Arousal and Dominance model, and Self-Assessment Maniking instrument, that asks user to inform how he/she feels; (ii) recommender, that selects and suggests items, using a collaborative-filtering based approache, in which user s prefference to an item is his/her affective reaction to it as the affective state detected after access; (iii) application, which interacts with user, shows probable most interesting items defined by recommender, and requests affect identification when it is necessarly; and (iv) data base, that stores available items and users prefferences. As a use case, Affective-Recommender is used in a e-learning scenario, due to personalization obtained with recommendation and emotion importances in learning process. The system was implemented over Moodle LMS. To exposes its operation, a use scenario was organized, simulating recommendation process. In order to check system applicability, with students opinion about to inform how he/she feels and to receive suggestions, it was applied in three UFSM graduation courses classes, and then it were analyzed data access and the answers to a sent questionnaire. As results, it was perceived that students were able to inform how they feel, and that occured changes in their affecive state, based on accessed item, although they don t see improvements with the recommendation, due to small data available to process and showr time of application.
Sistemas de Computação Pervasiva buscam melhorar a interação humano-computador através do uso de variáveis da situação do usuário que definem o contexto. A explosão da Internet e das tecnologias de informação e comunicação torna crescente a quantidade de itens disponíveis para a escolha, impondo custo para o usuário no processo de tomada de decisão. A Computação Afetiva tem entre seus objetivos identificar o estado emocional/afetivo do usuário durante uma interação computacional, para automaticamente responder a ele. Já Sistemas de Recomendação auxiliam a tomada de decisão, selecionando e sugerindo itens em situações onde há grandes volumes de informação, tradicionalmente, utilizando as preferências dos usuários para a seleção e sugestão. Esse processo pode ser melhorado com o uso do contexto (físico, ambiental, social), surgindo os Sistemas de Recomendação Sensíveis ao Contexto. Tendo em vista a importância das emoções em nossas vidas, e a possibilidade de tratamento delas com a Computação Afetiva, este trabalho utiliza o contexto afetivo do usuário como variável da situação, durante o processo de recomendação, propondo o Affective-Recommender um sistema de recomendação que faz uso do estado afetivo do usuário para selecionar e sugerir itens. O sistema foi modelado a partir de quatro componentes: (i) detector, que identifica o estado afetivo, utilizando o modelo multidimensional Pleasure, Arousal e Dominance e o instrumento Self-Assessment Manikin, solicitando que o usuário informe como se sente; (ii) recomendador, que escolhe e sugere itens, utilizando uma abordagem baseada em filtragem colaborativa, em que a preferência de um usuário para um item é vista como sua reação estado afetivo detectado após o contato ao item; (iii) aplicação, que interage com o usuário, exibe os itens de provável maior interesse definidos pelo recomendador, e solicita que o estado seja identificado, sempre que necessário; e (iv) base de dados, que armazena os itens disponíveis para serem sugeridos e as preferências de cada usuário. Como um caso de uso e prova de conceito, o Affective-Recommender é empregado em um cenário de e-learning, devido à importância da personalização, obtida com a recomendação, e das emoções no processo de aprendizagem. O sistema foi implementado utilizando-se como base o AVEA Moodle. Para expor o funcionamento, estruturou-se um cenário de uso, simulando-se o processo de recomendação. Para verificar a aplicabilidade real do sistema, ele foi empregado em três turmas de cursos de graduação da UFSM, sendo analisados dados de acesso e aplicado um questionário para identificar as impressões do alunos quanto a informar como se sentem e receber recomendações. Como resultados, percebeu-se que os alunos conseguiram informar seus estados afetivos, e que houve uma mudança em neste estado com base no item acessado, embora não tenham vislumbrado melhorias com as recomendações, em virtude da pequena quantidade de dados disponível para processamento e do curto tempo de aplicação.

O processo de digitalização da TV em diversos países do mundo tem contribuído para o aumento do volume de programas de TV, o que gera uma sobrecarga de informação. Consequentemente, o usuário está enfrentando dificuldade para encontrar os programas de TV favoritos dentre as várias opções disponíveis. Diante deste cenário, os sistemas de recomendação destacam-se como uma possível solução. Tais sistemas são capazes de filtrar itens relevantes de acordo com as preferências do usuário ou de um grupo de usuários que possuem perfis similares. Entretanto, em diversas recomendações o interesse do usuário pode depender do seu contexto. Assim, torna-se importante estender as abordagens tradicionais de recomendação personalizada por meio da exploração do contexto do usuário, o que poderá melhorar a qualidade das recomendações. Para isso, este trabalho descreve uma infraestrutura de software de suporte ao desenvolvimento e execução de sistemas de recomendação sensíveis ao contexto para TV Digital Interativa - intitulada de PersonalTVware. A solução proposta fornece componentes que implementam técnicas avançadas para recomendação de conteúdo e processamento de contexto. Com isso, os desenvolvedores de sistemas de recomendação concentram esforços na lógica de apresentação de seus sistemas, deixando questões de baixo nível para o PersonalTVware gerenciar. As modelagens de usuário, e do contexto, essenciais para o desenvolvimento do PersonalTVware, são representadas por padrões de metadados flexíveis usados na TV Digital Interativa (MPEG-7 e TV-Anytime), e suas devidas extensões. A arquitetura do PersonalTVware é composta por dois subsistemas: dispositivo do usuário e provedor de serviços. A tarefa de predição de preferências contextuais é baseada em métodos de aprendizagem de máquina, e a filtragem de informação sensível ao contexto tem como base a técnica de filtragem baseada em conteúdo. O conceito de perfil contextual também é apresentado e discutido. Para demonstrar e validar as funcionalidades do PersonalTVware em um cenário de uso, foi desenvolvido um sistema de recomendação sensível ao contexto como estudo de caso.
The process of digitalization of TV in several countries around the world has, contributed to increase the volume of TV programs offered and it leads, to information overload problem. Consequently, the user facing the difficulty to find their favorite TV programs in view of various available options. Within this scenario, the recommender systems stand out as a possible solution. These systems are capable of filtering relevant items according to the user preferences or the group of users who have similar profiles. However, the most of the recommender systems for Interactive Digital TV has rarely take into consideration the users contextual information in carrying out the recommendation. However, in many recommendations the user interest may depend on the context. Thus, it becomes important to extend the traditional approaches to personalized recommendation of TV programs by exploiting the context of user, which may improve the quality of the recommendations. Therefore, this work presents a software infrastructure in an Interactive Digital TV environment to support context-aware personalized recommendation of TV programs entitled PersonalTVware. The proposed solution provides components which implement advanced techniques to recommendation of content and context management. Thus, developers of recommender systems can concentrate efforts on the presentation logic of their systems, leaving low-level questions for the PersonalTVware managing. The modeling of user and context, essential for the development of PersonalTVware, are represented by granular metadata standards used in the Interactive Digital TV field (MPEG-7 and TV-Anytime), and its extensions required. The PersonalTVware architecture is composed by two subsystems: the users device and the service provider. The task of inferring contextual preferences is based on machine learning methods, and context-aware information filtering is based on content-based filtering technique. The concept of contextual user profile is presented and discussed. To demonstrate the functionalities in a usage scenario a context-aware recommender system was developed as a case study applying the PersonalTVware.

La diversité des contenus recommandation et la variation des contextes des utilisateurs rendent la prédiction en temps réel des préférences des utilisateurs de plus en plus difficile mettre en place. Toutefois, la plupart des approches existantes n'utilisent que le temps et l'emplacement actuels séparément et ignorent d'autres informations contextuelles sur lesquelles dépendent incontestablement les préférences des utilisateurs (par exemple, la météo, l'occasion). En outre, ils ne parviennent pas considérer conjointement ces informations contextuelles avec les interactions sociales entre les utilisateurs. D'autre part, la résolution de problèmes classiques de recommandation (par exemple, aucun programme de télévision vu par un nouvel utilisateur connu sous le nom du problème de démarrage froid et pas assez d'items co-évalués par d'autres utilisateurs ayant des préférences similaires, connu sous le nom du problème de manque de donnes) est d'importance significative puisque sont attaqués par plusieurs travaux. Dans notre travail de thèse, nous proposons un modèle probabiliste qui permet exploiter conjointement les informations contextuelles actuelles et l'influence sociale afin d'améliorer la recommandation des items. En particulier, le modèle probabiliste vise prédire la pertinence de contenu pour un utilisateur en fonction de son contexte actuel et de son influence sociale. Nous avons considérer plusieurs éléments du contexte actuel des utilisateurs tels que l'occasion, le jour de la semaine, la localisation et la météo. Nous avons utilisé la technique de lissage Laplace afin d'éviter les fortes probabilités. D'autre part, nous supposons que l'information provenant des relations sociales a une influence potentielle sur les préférences des utilisateurs. Ainsi, nous supposons que l'influence sociale dépend non seulement des évaluations des amis mais aussi de la similarité sociale entre les utilisateurs. Les similarités sociales utilisateur-ami peuvent être établies en fonction des interactions sociales entre les utilisateurs et leurs amis (par exemple les recommandations, les tags, les commentaires). Nous proposons alors de prendre en compte l'influence sociale en fonction de la mesure de similarité utilisateur-ami afin d'estimer les préférences des utilisateurs. Nous avons mené une série d'expérimentations en utilisant un ensemble de donnes réelles issues de la plateforme de TV sociale Pinhole. Cet ensemble de donnes inclut les historiques d'accès des utilisateurs-vidéos et les réseaux sociaux des téléspectateurs. En outre, nous collectons des informations contextuelles pour chaque historique d'accès utilisateur-vidéo saisi par le système de formulaire plat. Le système de la plateforme capture et enregistre les dernières informations contextuelles auxquelles le spectateur est confronté en regardant une telle vidéo.Dans notre évaluation, nous adoptons le filtrage collaboratif axé sur le temps, le profil dépendant du temps et la factorisation de la matrice axe sur le réseau social comme tant des modèles de référence. L'évaluation a port sur deux tâches de recommandation. La première consiste sélectionner une liste trie de vidéos. La seconde est la tâche de prédiction de la cote vidéo. Nous avons évalué l'impact de chaque élément du contexte de visualisation dans la performance de prédiction. Nous testons ainsi la capacité de notre modèle résoudre le problème de manque de données et le problème de recommandation de démarrage froid du téléspectateur. Les résultats expérimentaux démontrent que notre modèle surpasse les approches de l'état de l'art fondes sur le facteur temps et sur les réseaux sociaux. Dans les tests des problèmes de manque de donnes et de démarrage froid, notre modèle renvoie des prédictions cohérentes différentes valeurs de manque de données
Due to the diversity of alternative contents to choose and the change of users' preferences, real-time prediction of users' preferences in certain users' circumstances becomes increasingly hard for recommender systems. However, most existing context-aware approaches use only current time and location separately, and ignore other contextual information on which users' preferences may undoubtedly depend (e.g. weather, occasion). Furthermore, they fail to jointly consider these contextual information with social interactions between users. On the other hand, solving classic recommender problems (e.g. no seen items by a new user known as cold start problem, and no enough co-rated items with other users with similar preference as sparsity problem) is of significance importance since it is drawn by several works. In our thesis work, we propose a context-based approach that leverages jointly current contextual information and social influence in order to improve items recommendation. In particular, we propose a probabilistic model that aims to predict the relevance of items in respect with the user's current context. We considered several current context elements such as time, location, occasion, week day, location and weather. In order to avoid strong probabilities which leads to sparsity problem, we used Laplace smoothing technique. On the other hand, we argue that information from social relationships has potential influence on users' preferences. Thus, we assume that social influence depends not only on friends' ratings but also on social similarity between users. We proposed a social-based model that estimates the relevance of an item in respect with the social influence around the user on the relevance of this item. The user-friend social similarity information may be established based on social interactions between users and their friends (e.g. recommendations, tags, comments). Therefore, we argue that social similarity could be integrated using a similarity measure. Social influence is then jointly integrated based on user-friend similarity measure in order to estimate users' preferences. We conducted a comprehensive effectiveness evaluation on real dataset crawled from Pinhole social TV platform. This dataset includes viewer-video accessing history and viewers' friendship networks. In addition, we collected contextual information for each viewer-video accessing history captured by the plat form system. The platform system captures and records the last contextual information to which the viewer is faced while watching such a video. In our evaluation, we adopt Time-aware Collaborative Filtering, Time-Dependent Profile and Social Network-aware Matrix Factorization as baseline models. The evaluation focused on two recommendation tasks. The first one is the video list recommendation task and the second one is video rating prediction task. We evaluated the impact of each viewing context element in prediction performance. We tested the ability of our model to solve data sparsity and viewer cold start recommendation problems. The experimental results highlighted the effectiveness of our model compared to the considered baselines. Experimental results demonstrate that our approach outperforms time-aware and social network-based approaches. In the sparsity and cold start tests, our approach returns consistently accurate predictions at different values of data sparsity

abstract: Recommender systems are a type of information filtering system that suggests items that may be of interest to a user. Most information retrieval systems have an overwhelmingly large number of entries. Most users would experience information overload if they were forced to explore the full set of results. The goal of recommender systems is to overcome this limitation by predicting how users will value certain items and returning the items that should be of the highest interest to the user. Most recommender systems collect explicit user feedback, such as a rating, and attempt to optimize their model to this rating value. However, there is potential for a system to collect implicit user feedback, such as user purchases and clicks, to learn user preferences. Additionally with implicit user feedback, it is possible for the system to remember the context of user feedback in terms of which other items a user was considering when making their decisions. When considering implicit user feedback, only a subset of all evaluation techniques can be used. Currently, sufficient evaluation techniques for evaluating implicit user feedback do not exist. In this thesis, I introduce a new model for recommendation that borrows the idea of opportunity cost from economics. There are two variations of the model, one considering context and one that does not. Additionally, I propose a new evaluation measure that works specifically for the case of implicit user feedback.
Dissertation/Thesis
M.S. Computer Science 2012

碩士
國立臺灣海洋大學
運輸科學系
104
In the online-to-offline (O2O) and omni-channel retailing environment, consumers can switch different channels very easily. In response to the changing of consumer shopping behavior toward M-Commerce, companies strive to develop novel customer interactive IoT systems to engage consumers. IoT-based interactive digital signage has rich multimedia effects, making it a successful marketing tool for many retail firms; they can use the digital signage to provide wayfinding and location information and personal message and advertising to people who stay in front of the digital signage. Digital signage has become an important Business 4.0 smart retailing tool for most department stores and large shopping centers. However, most digital signage systems today lack interactive functions to the customer. And few digital signage systems available in the recent years are also insufficient in many aspects, such as, their recommendation model is deficiency in engaging anonymous passengers/visitors or customers without purchasing record and also not considering people will switch the preferences when the condition change. Few research works concerning the interactive digital signage can be found. Thus, this proposal delves into this new important research area in IoT and smart retailing. This proposal plan to collect historical data set of market transaction collected from mobile device users. Then, based on these data, various research methods from data mining, recommendation system development and experiment are applied to develop an analytical model for context-Aware recommender system for Interactive Digital Signage. This model is applied to a Recommendation analyze to target anonymous viewer and improve system prediction accuracy, in order to enhance the benefits of deploying DS in omni-channel retailing environment.

碩士
國立彰化師範大學
資訊管理學系所
104
With the development of Internet, more and more travelers search for travel information on the Internet. The travelers could use those information to help them plan itinerary. Because of the universal mobile internet, we could get a variety of information immediately, and it helps us make decision better. There are many recommendation systems using context-aware and user-user similarity model, but these systems did not have an organizational structure. In this study, we propose a travel recommendation system, providing recommendation based on travel topic. The system introduces ontology technology to discover the relationship between each tourist attractions, and build a travel topic model. It also aim to find potential travel topic from user’s travel histories. The system is context-aware which can perceive the context around the user. It will provide different recommendation results when the context changes. The season and distance context information is considered during the recommendation process. The evaluation result shows that our system can provide high satisfaction. The system is sufficient to meet user’s requirement, and it is capable to provide the tourist attraction that user is interested.

In the past few years, the general work habits of people have changed dramatically, raising concerns about their well-being. Numerous health-related problems have been observed from their health records such as obesity, diabetes or heart diseases, and unhealthy eating is one of its factors. But these problems can be prevented if people start eating healthy food. The population, in general, is also realizing that healthy eating is important for their well-being. However, they usually resist because they assume that healthy food is not tasty and they do not want to comprise their taste preferences. Moreover, they have various other considerations that become barriers for them while selecting a healthy recipe. These are:(1) their complex, restrained needs (i.e., allergies and nutritional goals), (2) their strict lifestyle or dietary preferences (i.e., their desire to eat only vegan or vegetarian food), (3) lack of knowledge about how to choose healthy recipes while exploiting their taste preferences, (4) choosing recipes that maximize the use of available ingredients in their kitchen. Numerous researchers have been working in this field and developed various applications and systems to suggest healthy recipes. Apart from unhealthy eating, household food wastage has become a public problem, and some of the causes, which trigger it are users’ taste preferences (i.e., disliking of the food), and not cooking food before ingredients expiry dates. Thus, we propose a personalized recipes recommender system as a proof of concept called CAPRECIPES, which is based on context-awareness. It tackles the aforementioned barriers and improves the users’ experiences by providing the recommendations of personalized recipes with minimal efforts while exploiting their dynamically changing contexts. CAPRECIPES also helps in the reduction of food wastage as it first shows the recipes, which contain the ingredients that are expiring soon and matches with users’ taste preferences. It also considers that recipes do not violate users’ health restrictions and nutritional goals, and use the maximum number of available ingredients in users’ kitchen. The proposed system gathers users’ taste preferences by exploiting two third-party social media applications (i.e., Facebook and YouTube) and collaborative-based filtering algorithm. This thesis also explores various natural language processing techniques such as text analysis and parts of speech tagging to identify the recipes’ preferences and to find the most relevant match for each recipe or ingredient having different names.
Graduate

碩士
國立彰化師範大學
資訊管理學系數位內容科技與管理碩士班
104
Currently under development and the growing popularity of mobile devices and the Location Based Service is more popularity issue. If imported a context-aware recommendation system on mobile device which capable of accordance with the user's current location, not only can quickly and easily get your store information, you can quickly find what you want to goods. The context-aware recommendation system is introduces ontology technologies to analyze the relations between the user and the store and the information about the store. The context-aware recommendation system can provide tailored, personalized results for the user in real-time. Finally the evaluation results show the context-aware recommendation system can provide high user satisfaction, and it is capable to solve the problems that users when they are looking for appropriate store to shopping.

碩士
國立臺灣大學
工程科學及海洋工程學研究所
100
With the increasing popularity of mobile devices and mobile networks, people can get a soaring amount of information, anywhere, anytime. How to solve the problem of the current information overload and provide personalized recommendation services is an important research topic. This thesis exploits the check-ins of Facebook Open Graph to design a mobile restaurant recommender system, which is based on collaborative filtering. The system summarizes the group preferences from individual users check-in in order to provide group recommendation services. Furthermore, the system considers social graph and contextual information to enhance the recommendation quality. These contextual information includes location, distance, age, sex index, time of day, weekday, month, number of companion and type of companion. In this thesis, we also proposed a method to evaluate the impact of location and distance context. Our experimental data is collected from the 69 volunteers in Facebook, which includes the 8264 check-ins. These check-ins are contributed by 3928 users in 2691 different restaurants from 2010/8/15 to 2012/4/30. The experimental results reveal that the accuracy of our system can be increased by approximately 38% while suggest restaurants within the area of 3-5 km radius, compared to popularity-based recommendation. It means that the proposed system can provide better recommendations than popularity-based recommendations, if the user asks for a restaurant suggestion in a larger area.

Currently recommender systems are incorporating context and social information of the user, producing context aware recommender systems. In the future, they will use implicit, local and personal information of the user from the Internet of Things; where anyone and anything will be connected at anytime and anywhere. A Context- Aware Recommendation System has been introduced in this thesis. The fact that the future is for Internet of Things, and the multiple recommendation leads to my system design, in which multi-type rather than one type of recommendations will be recommended to the user. In this paper, a design of a context aware recommender system that recommends different types of items under the Internet of Things paradigm is proposed. A major part of this design is the context aware management system. In this system, we have used a neural network that will do the reasoning of the context to determine whether to push a recommendation or not and what type of items to recommend. The neural network inputs are derived virtually from the Internet of Things, and its outputs are scores for three types of recommendations, they are: songs, movies and none. These scores have been used to decide whether to push a recommendation or not, and what type of recommendations to. The results of 1000 random contexts were tested. For an average of 98.80% of them, our trained neural network generated correct recommendation types in the correct times and contexts.

A learner in an informal learning environment, such as a museum, encounters various challenges. After initial assessment, a set of methods were proposed that may enhance a learner’s experience in a museum using computer aided technologies. The most important insight was the need to support the museum visitor in three phases of activity: prior to the visit, during the visit, and after the visit. We hypothesized that software tools that could help connect these three phases would be helpful and valuable supports for the visitor. To test and evaluate our hypothesis, a system called “The Museum Explorer” was built and instantiated using the collection in the Museum of Antiquities located at the University of Saskatchewan. An evaluation of the Museum Explorer was conducted. Results show that the Museum Explorer was largely successful in achieving our goals. The Museum Explorer is an integrated solution for visitors in museums across the pre-visit, visit, and post-visit phases. The Museum Explorer was designed to provide a means to connect and transfer user experience across the major phases of a museum visit. For each phase of a visitor’s experience, a set of tools was built that provides intelligent and interactive communication features. To assist visitors selecting artefacts to visit, a recommender system allows users to select a set of constraints. To better manage interactivity, features and functions were offered based on context. A study was conducted with volunteer museum visitors. Results from the study show that the Museum Explorer is a useful support. Analysis of the usage data captured by the Museum Explorer has revealed some interesting facts about users’ preferences in the domain that can be used by future researchers.