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.

Current works of bike demand prediction mainly focus on cluster level and perform poorly on predicting demands of a single station. In the first task, we introduce a contextual based bike demand prediction model, which predicts bike demands for per station by combining spatio-temporal network and environment contexts synergistically. Furthermore, since people's movement information is an important factor, which influences the bike demands of each station. To have a better understanding of people's movements, we need to analyze the relationship between different places. In the second task, we propose an origin-destination model to learn place representations by using large scale movement data. Then based on the people's movement information, we incorporate the place embedding into our bike demand prediction model, which is built by using multi-source large scale datasets: New York Citi bike data, New York taxi trip records, and New York POI data. Finally, as deep learning methods have been successfully applied to many fields such as image recognition and natural language processing, it inspires us to incorporate the complex deep learning method into the bike demand prediction problem. So in this task, we propose a deep spatial-temporal (DST) model, which contains three major components: spatial dependencies, temporal dependencies, and external influence. Experiments on the NYC Citi Bike system show the effectiveness and efficiency of our model when compared with the state-of-the-art methods.

Context information is information that describes the user's context. The goal of the Adaptive Context Aware Services (ACAS) project is to enable applications to use context information in order to adapt their behaviour to the user and his environment without requiring the user to manually change/manage parameters. While the concept of linking context aware entities together to form a logical "context network" was introduced earlier in the project, some questions regarding context information discovery and the discovery of context aware entities were previously unanswered. The goal of this thesis was to design and evaluate such a context network allowing entities todiscover each other and exchange information regarding their services and context information. For this purpose, a "Context Entity Registrar" has been developed allowing entities to register, thus they can easily be found by other entities who can query this registrar. During the design of this proposed solution, a special focus has been given to the performance of the registrar, especially how it scales when answering a large number of requests, in order to validate the design's potential as a solution to context aware entity discovery. Measurements have shown that this proposed solution scales well, making it a key element of a context network. Discovery of other entities and of context information play an important role to determine the performances of a context aware implementation. This masters thesis addresses first the issue of the architecture of the context network and then some tests to measure the performances of the proposed solution. / Context information är information som beskriver användarens omgivning. Adaptive Context Aware Services (ACAS) projektet har som mål att möjliggöra applikationer att använda kontext information för att anpassa sitt beteende till användaren och dess miljö, utan att kräva att användaren ska sätta eller hantera alla parametrar manuellt. ACAS projektet har tidigare infört konceptet "context network" som förbinder context aware enheter. Det finns dock kvar några obesvarade frågor angående upptäckt av context information och av context aware enheter.Trots att sättet att länka ihop kontextmedvetna enheter för att forma ett logiskt 'kontext nätverk' introducerades tidigare i projektet, finns det kvar några obesvarade frågor angående upptäckt av kontext information och upptäckt av kontextmedvetna enheter. Examensarbetets mål är att utforma och utvärdera ett sådant kontextnätverk som ger enheterna möjlighet att upptäcka varandra och utbyta information om tjänster och context information. Därför utvecklades "Context Entity Registrar" så att enheterna kan registrera sig för att kunna bli upptäckta av andra enheter som kan göra förfrågningar till detta register. Under designen av denna föreslagna lösning har särskild fokus lagts på registrens prestanda, speciellt avseende skalbarhet med avseende på antalet förfrågningar for att validera designens potential som lösning för kontextmedveten upptäckt av enheter. Mätningar har visat att lösningen skalar bra vilket gör kontext registret till ett nyckelelement i ett kontextnätverk. Upptäkten av andra enheter och av kontextinformation har en viktig roll i att bestämma en kontextmedveten implementations prestanda. Detta examensarbete kommer först att behandla kontextnätverkets arkitektur och därefter några testerna för att mäta prestanda i den föreslagna lösningen.

Today customization of services and applications is one of the major challenges in facilitating ease of use. More and more people are interested in context aware services. In this work, I will study context awareness: What contributions can it make? What technical issues are raised? I will concentrate on the semantic problem and show how new technologies such as a web ontology language can facilitate creating context aware services. An application was implemented using these principles as a proof of concept and to enable some evaluation of this approach. This application has shown that combining semantic processing with SIP call processing is feasible and measurements have demonstrated a highly scalable context aware application for at least simple CPL scripts. / I dagens läge är en av de stora utmaningarna att skräddarsy applikationer och tjänster efter kundens begär för att höja användarvänligheten. Fler och fler kunder intresserar sig för tjänster som är kontextuppmärksamma. Jag kommer i mitt arbete att studera detta, mina frågeställningar är huvudsakligen, hur är detta användbart? Vad slags tekniska problem uppstår kring applikationen av detta koncept? Jag kommer att koncentrera mig på semantiska problem och visa hur nya teknologier så som webbontologispråk kan underlätta då man skapar tjänster som är kontextuppmärksamma. Jag kommer även att skapa en applikation där jag använder dessa principer för att visa konceptet och för att underlätta för en utvärdering av detta tillvägagångssätt. Denna applikation samt mätdata visar att det är möjligt att kombinera semantisk bearbetning och SIP anropsbearbetning för att skapa oerhört robusta applikationer, åtminstonde för enkla CPL script, som är hållbara i ett stort omfång av systemstorlekar.

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.

No description available.

Em NGN (Next Generation Networks), os usuários podem se conectar em diferentes tecnologias de rede e desejam, além de uma comunicação transparente, novos serviços personalizados. Neste sentido, este trabalho explora informações de contexto em NGN. O principal objetivo é capturar as informações de contexto envolvidas no momento da utilização da rede sem fio, na escolha do novo ponto de acesso e no procedimento do handover. Este contexto capturado é aplicado para serviços cientes de contexto. A proposta é validada por meio de dois cenários, o primeiro é um mashup que exibe as redes sem fio disponíveis de um determinado local e o segundo um protótipo gerenciador de conexões para atender as preferências do usuário. Adicionalmente, são realizadas duas avaliações do impacto do uso de contexto em handovers. Para tanto, foi implantado um testbed NGN com o protocolo Mobile IP / In the NGN (Next Generation Networks) users can connect their Internet devices to different network technologies. In addition to a seamless communication, users desire new and personalized services. In this sense, this thesis exploits the context-aware information in NGN environments. The main propose is to capture the context information about the time involved by the use of wireless link for selecting the target access point and on the handover procedure. This context information is applied to the context-aware services. The propose is validated through two scenarios: the first is a mashup that shows the availability of wireless networks in a particular location, and the second is a prototype of a connection manager to attend user´s preferences. In addition, two evaluations are accomplished on the impact by the use of context information over handovers. Thus, an NGN testbed is deployed with Mobile IP protocol

Em NGN (Next Generation Networks), os usuários podem se conectar em diferentes tecnologias de rede e desejam, além de uma comunicação transparente, novos serviços personalizados. Neste sentido, este trabalho explora informações de contexto em NGN. O principal objetivo é capturar as informações de contexto envolvidas no momento da utilização da rede sem fio, na escolha do novo ponto de acesso e no procedimento do handover. Este contexto capturado é aplicado para serviços cientes de contexto. A proposta é validada por meio de dois cenários, o primeiro é um mashup que exibe as redes sem fio disponíveis de um determinado local e o segundo um protótipo gerenciador de conexões para atender as preferências do usuário. Adicionalmente, são realizadas duas avaliações do impacto do uso de contexto em handovers. Para tanto, foi implantado um testbed NGN com o protocolo Mobile IP / In the NGN (Next Generation Networks) users can connect their Internet devices to different network technologies. In addition to a seamless communication, users desire new and personalized services. In this sense, this thesis exploits the context-aware information in NGN environments. The main propose is to capture the context information about the time involved by the use of wireless link for selecting the target access point and on the handover procedure. This context information is applied to the context-aware services. The propose is validated through two scenarios: the first is a mashup that shows the availability of wireless networks in a particular location, and the second is a prototype of a connection manager to attend user´s preferences. In addition, two evaluations are accomplished on the impact by the use of context information over handovers. Thus, an NGN testbed is deployed with Mobile IP protocol

Submitted by Kamila Costa (kamilavasconceloscosta@gmail.com) on 2015-06-23T17:26:53Z No. of bitstreams: 1 Dissertação-Wesley de S Viana.pdf: 3248107 bytes, checksum: 052d64e91b1af49c56df079ee11e65af (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2015-07-07T18:08:42Z (GMT) No. of bitstreams: 1 Dissertação-Wesley de S Viana.pdf: 3248107 bytes, checksum: 052d64e91b1af49c56df079ee11e65af (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2015-07-07T18:23:03Z (GMT) No. of bitstreams: 1 Dissertação-Wesley de S Viana.pdf: 3248107 bytes, checksum: 052d64e91b1af49c56df079ee11e65af (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2015-07-07T18:39:44Z (GMT) No. of bitstreams: 1 Dissertação-Wesley de S Viana.pdf: 3248107 bytes, checksum: 052d64e91b1af49c56df079ee11e65af (MD5) / Made available in DSpace on 2015-07-07T18:39:44Z (GMT). No. of bitstreams: 1 Dissertação-Wesley de S Viana.pdf: 3248107 bytes, checksum: 052d64e91b1af49c56df079ee11e65af (MD5) Previous issue date: 2015-03-30 / CNPq - Conselho Nacional de Desenvolvimento Científico e Tecnológico / The distance education is a technology-mediated education mode in which the teacher becomes an intermediary of knowledge and the student is primarily responsible for the pursuit of knowledge and learning. In recent years, there has been a rapid growth in the use of this form of education in public and private institutions, boosting the development of new technologies and technological tools in order to support this growth. In this context, the Learning Management System (LMS) have become very used to support for distance education. However, these environments present some limitations, requiring the construction of technological tools to circumvent such restrictions. In the midst of this information it has been found that some teachers have difficulties in tracking students in designated locations for supervised internship, both physically and by E-learning support environment. Therefore, this paper proposes an approach involving mobile technologies and location context to enable the preparation and realization of tasks in specific locations. This allows the teacher to create a task and assign it to a specific location and make sure that the student was at the scene, because the task will be available only in the location set. In addition, we developed a Multi-agent System (SMA) to notify about deadlines and tasks available in the environment through SMS and email. The approach was validated in two disciplines of distance courses and internship in a pilot course conducted at the Federal University of Amazonas (UFAM) that allowed the initial validation with real users in various scenarios of applicability of the approach. / A Educação a Distância (EaD) é uma modalidade de educação mediada por tecnologia no qual o professor torna-se um intermediário do saber e o aluno é o principal responsável pela busca do conhecimento e aprendizado. Nos últimos anos, houve um crescimento acelerado na utilização dessa forma de educação em instituições públicas e privadas, impulsionando o desenvolvimento de novas tecnologias e ferramentas tecnológicas a fim de apoiar esse crescimento. Nesse contexto, os Ambientes Virtuais de Aprendizagem (AVAs) tornaram-se muito utilizados para suporte à EaD. No entanto, esses ambientes apresentam algumas limitações, sendo necessária a construção de ferramentas tecnológicas para contornar tais limitações. Em meio a essas informações foi constatado que alguns professores possuem dificuldades em acompanhar os alunos em locais designados para a realização de estágio supervisionado, tanto fisicamente quanto pelo ambiente de suporte à EaD. Portanto, o presente trabalho propõe uma abordagem envolvendo tecnologias móveis e contexto por localização para possibilitar a elaboração e a realização de tarefas em locais específicos. Isso permite ao professor criar uma tarefa e designá-la a um local especifico e se certificar-se que o aluno foi ao local, pois a tarefa será disponibilizada apenas no local definido. Além disso, foi desenvolvido um Sistema Multiagente (SMA) para notificar sobre prazos e tarefas disponíveis no ambiente por meio de SMS e e-mail. A abordagem foi validada em duas disciplinas de estágio em cursos a distância e em um curso piloto conduzido na Universidade Federal do Amazonas (UFAM) que permitiu a validação inicial com usuários reais em vários cenários de aplicabilidade da abordagem.

Master of Philosophy (Architecture) / Context-aware computing is a mobile computing paradigm in which applications can discover, use, and take advantage of contextual information, such as the location, tasks and preferences of the user, in order to adapt their behaviour in response to changing operating environments and user requirements. A problem that arises is the inability to respond to contextual information that cannot be classified into any known context. Many context-aware applications require all discovered contextual information to exactly match a type of context, otherwise the application will not react responsively. The ability to learn and recall contexts based on the contextual information discovered has not been very well addressed by previous context-aware applications and research. The aim of this thesis is to develop a component middleware technology for mobile computing devices for the discovery and capture of contextual information, using the situated reasoning concept of constructive memory. The research contribution of this thesis lies in developing a modified architecture for context-aware systems, using a constructive memory model as a way to learn and recall contexts from previous experiences and application interactions. Using a constructive memory model, previous experiences can be induced to construct potential contexts, given a small amount of learning and interaction. The learning process is able to map the many variations of contextual information currently discovered by the user with a predicted type of context based on what the application has stored and seen previously. It only requires a small amount of contextual information to predict a context, something common context-aware systems lack, as they require all information before a type of context is assigned. Additionally, some mechanism to reason about the contextual information being discovered from past application interactions will be beneficial to induce contexts for future experiences.