Context Aware Recommender Systems

In the vision of ubiquitous computing, users are imagined as evolving in various, changing and not always foreseeable environments, in which platforms may arrive and disappear in an opportunistic manner. As a result, there is a need for User Interfaces (UI) to adapt to their context of use (<User, Platform, Environment>) while preserving usability. This capacity of UIs is called Plasticity. In a forward engineering approach, UIs are designed step by step starting from domainbased descriptions (user's tasks and concepts) t o code. It is now well understood that plasticity may impact UIs at any level of abstraction. This calls for keeping the UIs design rationale alive at runtime. As models are practiced since a long time in Human-Computer Interaction (HCI), this paper investigates to which extent Model-Driven Engineering (MDE) is relevant for plasticity. UIs are designed as a net of models that define different perspectives on a same UI (user's task, domain, concepts, widgets, etc.). The net is alive at run-time and transformed when the context of use changes. Transformations are performed with respect t o usability. This paper sketches the vision on a small running case study. It highlights a set of strengths and doubts that give rise to many perspectives.

This paper explores how user interface (UI) and user experience (UX) insights shape the success of information technology business applications. It highlights the close connection between UI and UX in driving user adoption, satisfaction, and engagement, as well as the diverse metrics and methods that provide a window into real user behavior. Drawing on real-world examples, the paper shows how thoughtful, evidence-based decisions can simplify workflows, cut down on mistakes, and boost overall performance-whether for tools used within a company or platforms intended for customers. It also looks at upcoming developments such as Integrated voice interfaces and augmented reality, emphasizing that as technology shifts, so must the ways we gather and apply user feedback. Throughout the discussion, the paper highlights the significance of consistent, ethical approaches to analyzing and refining UI/UX. Ultimately, it discusses that businesses that align a user-focused view with data-driven strategies gain a key advantage: delivering solutions that meet business goals while deeply resonating with the people who rely on them.

Received May 13, 2020 Revised June 21, 2020 Accepted June 24, 2020 Published July 15, 2020 Travel recommender systems have been developed to meet the needs of users in the field of tourism. This system has several versions depending on the characteristics of the country, users and filtering techniques used. The development of recommendation filtering system techniques is very rapid so that the recommendation system has high enough complexity, but it also must have high usability. This paper discusses how the travel recommender system architecture is built by examining data structures, processing procedures and interaction design. The goal is to obtain the best usability in implementing a travel recommendation system. The system is built using the example case of finding the right tourist spot in Yogyakarta, Indonesia. This system applies several filtering techniques such as knowledge-based filtering, content-based filtering, and collaborative filtering. The evaluation results show t...

Recommender systems are now a familiar part of the digital landscape helping us to choose which movies to watch and books to read. They guide us about where to stay and eat when we travel. They help us to keep in touch with friends and may even influence our choice of a mate. To do this recommender systems require data. Lots of data. In the early years this data came in the form of our online transactions and item ratings. More recently recommendations have been influenced by our social networks, the connections that link us, and the things that we share with others. Today there is a new form of data that has the potential to drive recommender systems of the future: user-generated reviews. Reviews are now a routine part of how we make decisions, large and small. Most of us wouldn't dream of booking a hotel without first checking out its reviews and companies like TripAdvisor and Yelp have build billion dollar enterprises on the opinions of millions of people. In this talk we will discuss the role of user generated reviews in a new generation of recommender systems and some of the ways that opinions can be leveraged to better understand users and generate new forms of recommendations. We will focus on how opinion mining techniques can be used to extract features and sentiment from unstructured review text and ways to use this information in recommendation ranking and explanation.

This paper is to present an overview of Collaborative Filtering (CF) recommender system and show the major CF challenges. In general, the recommendation systems are the best way to help users to overcome the information overload issue. The CF approach is one of the most widely used and most successful methods in the recommendation system, such as e-commerce. This paper introduced a brief description about recommender's approaches which are: content-Based, collaborative filtering and hybrid approach. Next, defined the main challenges which have clearly impact on the performance and accuracy of CF recommender system. The major finding of this paper is the CF main problems: Data sparsity, Cold-star, and Scalability. By presenting of these challenges the quality of recommendations can be improved by proposing new methods. The paper ends with conclusion summarizes the limitations of the existing methods and recommendations.

raditional recommender systems, such as those based on content-based and collaborative filtering, tend to use fairly simple user models. For example, user-based collaborative filtering generally models the user as a vector of item ratings. As additional observations are made about users' preferences, the user models are extended, and the full collection of user preferences is used to generate recommendations or make predictions. This approach, therefore, ignores the notion of "situated actions" (Suchman 1987), the fact that users interact with the system within a particular "context" and that preferences for items within one context may be different from those in another context. In many application domains, a context-independent representation may lose predictive power because potentially useful information from multiple contexts is aggregated. For example, when a user is buying books, the preferences the user expresses in one context, such as "books for my children," may be of no predictive value when the user seeks recommendations in a different context, such as "work-related books." The ideal contextaware recommendation system would, therefore, be able reliably to label each user action with an appropriate context and effectively tailor the system output to the user in that given context. The concept of "context" has been studied extensively in several areas of computing and other disciplines. For example, examine and compare some 150 different definitions of context from a number of different fields and conclude that the multifaceted nature of the concept makes it difficult to find a unifying definition: "Is context a frame for a given object? Is it the set of elements that have any influence on the object? Is it possible to define context a priori or just state Articles

This paper introduces the notions of mask and multiple format for setting up a progressive access to information in an Information System. Masks consist of more or less complete representations of the structure of information while multiple formats correspond to different representations available for a multimedia datum. We describe how these two notions are integrated into AROM, an object-based knowledge representation system which supports a UML-like formalism, by instantiating a model called PAM (Progressive Access Model). More generally, a progressive and adapted access to information can be given to users of any Information System by simply adapting the PAM to the underlying data schema and coupling it with the target data or knowledge base.

In this paper the architecture of a learner-centered e-Learning system, which aims to qualify and recommend learning material using semantic web technologies, is presented. In the proposed approach, the learner has a central role in the decision-making process of the distribution of learning material. The main idea relies on gathering appropriate reputation metadata from the learning material consumers. As a result of a proper material ranking, a better matching to future searches of learners having similar profile is anticipated. An experimental implementation of the presented architecture provided interesting results that better depict the applicability and usefulness of the proposals.

The production of Voluntary Geographic Information has been growing considerably and continues to be an active area of research. However, the lack of knowledge about the quality of information generated on a voluntary and participatory basis raises challenges and questions about its use. In the review carried out for the Colombian case, no studies related to the subject were identified; consequently, this study is presented on the evaluation of the quality of this type of information on the road network of Bogotá with respect to completeness, positional accuracy and thematic accuracy. This evaluation was carried out by means of a semi-automatic process that uses a mobile buffer and the centroid of the roads to make the corresponding comparisons between two data sources. The results found reveal that the method used allowed to compare up to 85.0% of the data, and that the OpenStreetMap mesh has a completeness of 85.4%, over the entire area of Bogotá. A positional accuracy of 3.98 m a...

ABSTRAK Pada era kecerdasan buatan modern, sistem dituntut mampu beradaptasi terhadap kompleksitas dan ketidakpastian lingkungan nyata. Penelitian ini memperkenalkan kerangka kerja inovatif yang mengombinasikan logika fuzzy dengan reinforcement learning untuk membentuk sistem cerdas hibrida yang responsif dan efisien. Pendekatan modular digunakan agar agen mampu belajar dan menyesuaikan aturan fuzzy berdasarkan umpan balik dari lingkungan. Eksperimen menunjukkan peningkatan akurasi dan kemampuan adaptif pada berbagai kondisi dinamis. Model ini memiliki potensi aplikasi dalam bidang robotika, kendaraan otonom, serta sistem rekomendasi berbasis konteks.

The evolving structure of the modern workplace-driven by hybrid work models and remote collaboration-has necessitated a redefinition of User Experience (UX) frameworks in digital enterprise ecosystems. In this context, Artificial Intelligence (AI) has emerged as a pivotal enabler for enhancing UX by facilitating intelligent, adaptive, and personalized interactions across distributed digital environments. This study presents a comprehensive systematic literature review examining how AI-driven tools such as chatbots, recommendation engines, emotion-aware systems, and context-aware automation contribute to UX optimization in digital workplaces. Drawing on 87 peer-reviewed articles published between 2010 and 2024, and employing the PRISMA 2020 methodology, the review synthesizes empirical and theoretical insights across key themes, including AI-powered remote support, personalized interfaces, intelligent user guidance, and emotional intelligence integration in hybrid systems. The findings reveal that AI enhances UX at multiple levels: (1) by automating routine support functions to reduce user friction and improve response accuracy; (2) through adaptive personalization based on user behavior, roles, and preferences; (3) by enabling emotional intelligence features that detect and respond to user moods, stress, and disengagement; and (4) through real-time contextual adaptations that adjust interfaces based on environmental cues. AI systems integrated into platforms such as Microsoft Teams, Zoom, Slack, Salesforce, and Google Workspace were found to improve usability, satisfaction, and task efficiency while supporting digital wellbeing. Additionally, trust and transparency emerged as critical UX factors in AI adoption, emphasizing the importance of explainable AI and user autonomy in interface design. This review contributes to the evolving discourse on human-centered AI by framing UX not just as a functional outcome but as a multi-dimensional construct shaped by affective, cognitive, and behavioral interactions across AI-augmented platforms. By analyzing the convergence of AI technologies and UX principles in enterprise settings, the study provides a structured framework for designing adaptive, inclusive, and ethically aligned digital work environments. The synthesis also identifies gaps in longitudinal evaluations, emotional diversity modeling, and cross-cultural personalization strategies, offering directions for future empirical and design-focused research in AI-powered UX. Ultimately, this review underscores the transformative impact of AI in redefining the contours of user interaction, engagement, and satisfaction within the digital workplace paradigm.

In this study, an attempt was made using machine learning algorithm with the user data store in the mobile cloud framework to solve the problem of data over-collection. This was achieved by designing a model using the security risk level of the applications and the corresponding class level of the users on the smartphone that will help in preventing smartphone apps from accessing and collecting users' private data while still within the permission scope. Users can store information in the cloud environment where the huge numbers of users are involved. We develop a mobile agent simulator to generate data, and determine the security risk level of the apps on users' data with the class level of the data. The permission model was designed to determine whether the app is granted permission to access user's data or not. The data was trained with the use of Neural Network. The evaluation metrics used were accuracy and comparison. For accuracy, the algorithm was compared with the existing algorithm. The data analysis showed that there was restriction for apps accessing the users' data. The model if deployed on the smartphone will prevent apps from over collect users' data even while still within the permission scope. This study proved that neural network with mobile cloud computing can be applied to prevent data over-collection in smart devices.

AI-driven personalization represents a transformative force in customer engagement, utilizing advanced algorithms to deliver tailored experiences at individual levels. This article explores the architectural foundations, core algorithms, implementation challenges, evaluation frameworks, and industry-specific applications that power modern personalization systems. From collaborative filtering and deep learning networks to real-time processing engines and privacy-preserving techniques, the technological ecosystem supporting personalization continues to evolve rapidly. The discussion addresses how organizations overcome critical challenges including cold-start problems, data sparsity, and filter bubbles while measuring success through both technical and business metrics. By examining applications across e-commerce, media, finance, healthcare, education, and retail sectors, the content illuminates how domain-specific adaptations create value through dynamic pricing, adaptive interfaces, customized recommendations, and seamless omnichannel experiences.

Middleware support for pervasive context-aware systems relieves context-aware applications from dealing with the complexity of context-specific operations such as context acquisition, aggregation, reasoning and distribution. The middleware decouples applications from the underlying heterogeneous context sensors, and offers advantages such as rapid development of context-aware applications and efficient usage of the context sensors. Context sensors have inherent limitations with respect to the quality of the context information they produce. Without breaking the decoupling, the middleware needs to explicitly model and quantify this Quality of Context in order to allow application to adapt their behavior based on the Quality of Context, for efficiency reasons and to enable Quality-of-Context-aware privacy policies. In this paper we identify and define five Quality-of-Context indicators for context-aware middleware, and discuss different alternatives for their quantification. These Quality-of-Context indicators are: precision, freshness, spatial resolution, temporal resolution and probability of correctness.

The growing demand for seamless and personalized customer experiences has transformed how businesses approach self-service and promotional strategies. This research explores implementing customized recommendation systems to enhance customer engagement, satisfaction, and loyalty across various industries. By leveraging advanced algorithms and customer data, these systems enable businesses to offer tailored solutions that meet individual preferences, streamline self-service interactions, and improve promotional effectiveness. Through surveys, experiments, and case studies, the study highlights the positive impact of personalized recommendations on customer behavior, including increased engagement rates, improved retention, and higher conversion rates. The findings underscore the potential of such systems to enhance effortless customer experiences and drive business growth by fostering deeper connections with consumers. This paper provides actionable insights for businesses aiming to adopt or optimize recommendation systems to stay competitive in an increasingly customer-centric marketplace.

Context-aware systems are an emerging genre of computer systems that help add some forms of intelligence to our surroundings. The ATRACO project uses the ambient ecology metaphor to conceptualize a space populated by connected devices and services that are interrelated with each other, the environment and the people, supporting the users' everyday activities in a meaningful way. Everyday appliances, devices, and context aware artifacts are part of the ATRACO ambient ecologies. In this paper we present the connected home platform adopted by ATRACO and its evolution to provide network adaptation and context-aware services. A flexible and distributed context-aware service model is introduced using the OSGi and UPnP frameworks. UPnP is used to converge the existing network infrastructure comprising of heterogeneous technologies and protocols at the IP level. Furthermore, we introduce a contextaware service model and provide paradigms of context aware services that build upon perceptual and context aware components of the platform.

Este trabajo de fin de grado consiste en el desarrollo de un sistema en el lenguaje Python que realiza inferencia lógica, implementando una redefinición del dialecto de reglas de producción del estándar Rule Interchange Format, un lenguaje para la unificación de documentos de reglas en diferentes lenguajes. En adición, el motor de inferencia desarrollado usa ontologías escritas en un subconjunto de OWL 2, e implementa una versión del algoritmo Rete para la realización de matching en reglas. Este trabajo se engloba en las tecnologías de la web semántica, y pretende ser una alternativa cómoda y fácil de usar que no requiera de traductores entre lenguajes de reglas. También se ha realizado un estudio de rendimiento del motor, en el que se observa el peso que tiene el pobre rendimiento de Python sobre el motor y la influencia del tamaño de la red Rete sobre el tiempo de ejecución.

Tourist recommendation system is a system that provides information and recommendations for tourists. This system will help users to reduce the search process and help to make choices. Currently, the recommendation system uses the internet because it is cheap, requires a short time, can be accessed from anywhere and always update. The tourism recommendation system works by using the restrictions and preferences of tourists, then it will recommend tourism travel routes. Recommendation systems can be grouped into four types which are content-based, collaborative-filtering, knowledge-based, and hybrid systems that combine two or more methods. Usability is a product condition that can be used specifically by users to achieve effectiveness, efficiency, and satisfaction in the context of use. Usability test is conducted to see the level of comfort and ease of use of a system recommendation. In general, there are two types of measurements, which are quantitative and qualitative measurement...

Location-aware messages left by people can make visible some aspects of their everyday experiences at a location. To understand the contextual factors surrounding how users produce and consume location-aware multimedia messaging (LMM), we use an experience-centered framework that makes explicit the different aspects of an experience. Using this framework, we conducted an exploratory, diary study aimed at eliciting implications for the study and design of LMM systems. In an earlier pilot study, we found that subjects did not have enough time to fully capture their everyday experiences using an LMM prototype, which led us to conduct a longer study using a multimodal diary method. The diary study data (verified for reliability using a categorization task) provided a closer look at the different aspects (spatiotemporal, social, affective, and cognitive) of people's experience. From the data, we derive three main findings (predominant LMM domains and tasks, capturing experience vs. experience of capture, context-dependent personalization) to inform the study and design of future LMM systems.

We describe Streamwatchr, a real-time system for analyzing the music listening behavior of people around the world. Streamwatchr collects music-related tweets, extracts artists and songs, and visualizes the results in three ways: (i) currently trending songs and artists, (ii) newly discovered songs, and (iii) popularity statistics per country and worldwide for both songs and artists.

Travel recommender systems have been developed to meet the needs of users in the field of tourism. This system has several versions depending on the characteristics of the country, users and filtering techniques used. The development of recommendation filtering system techniques is very rapid so that the recommendation system has high enough complexity, but it also must have high usability. This paper discusses how the travel recommender system architecture is built by examining data structures, processing procedures and interaction design. The goal is to obtain the best usability in implementing a travel recommendation system. The system is built using the example case of finding the right tourist spot in Yogyakarta, Indonesia. This system applies several filtering techniques such as knowledge-based filtering, content-based filtering, and collaborative filtering. The evaluation results show that the system architecture optimized gets a usability level acceptable.

We describe an architecture for generating context-aware recommendations along with detailed textual explanations to support the user in the decision-making process. CARE (Context-Aware Recommender with Explanation) incorporates a hierarchical structure, in which independent modules embodying different aspects of the context cooperate together to generate recommendations for the user with accompanying rationales. We follow the Interaction Design principles to develop personas, goals and user scenarios, based on which a prototype system is developed. We present here two examples of its performance when processing movieratings data set with contextual information. We argue that our architecture is extensible in that more modules can be added as needed, and the approach can be applied to other domains as well.

The vertiginous penetration and advance of mobile devices in all levels of our society leads towards new technological challenges. This paper is focused on the application of mobile devices in the learning environment, since the current mobile learning development is finding several problems and lacks, such as use of the new features (e.g. location and motion sensors) and the desigh without having including the existing e-learning services. This paper introduces a framework created as a need to interconnect the mobile learning environment with context aware systems and existing e-learning resources. The paper describes the requirements that a framework like this must support, to finally give an overview of the author's solution.

Recommender systems (RSs) often focus on learning users' long-term preferences, while the sequential pattern of behavior is ignored. On the other hand, sequential RSs try to predict the next action by exploring relations between items in a user's last interactions but do not consider the general preference. Recently, the performance of RSs has increased by unifying these two types of paradigms. However, existing methods still have two limitations. First, the user's behavior uncertainty impedes precise learning of preferences. Second, being unable to understand the semantics of items makes the effect of the same item considered in the same way. These limitations jointly prevent RS from learning multifaceted preferences to capture the actual intentions of users. Existing methods have not properly addressed these problems since they ignore context-aware interactions between the user and item in terms of the links between the user and item attributes and sequential user actions over time. To address these challenges, this paper proposes a novel model, called the Dynamic Intention-Aware Recommendation with attention-based context-aware item attributes modeling (DIARec), which is capable of determining users' preferences based on their goal intention, taking into account the influence of various item features on user decision-making in their current context. Specifically, to model users' dynamic intentions, we introduce a dynamic intent-aware module to represent the hierarchical relations between items and their attributes in a given session. Experiments on benchmark datasets indicate that the proposed model DIARec outperforms other stateof-the-art methods.

One of the most currently used knowledge representation forms are ontologies; which offer dissimilar advantages for modeling, generation, distribution and use of knowledge produced and accumulated in organizations. Given these advantages for knowledge management, ontologies are widely used to manage large volumes of environmental information that arise from this process, primarily the historical storage of environmental indicators. It is for this reason has implemented a computer system for the management of environmental knowledge based on ontology OntoEnvironmental; which model the environmental indicators that rules the environmental management business process. To facilitate access to modeled knowledge in the ontology was implemented a computer system built on a software architecture based on ontology, which integrates the technologies needed to manage the semantic knowledge. This architecture allows through the inference engine constituted by Jena and OWL API frameworks, along ...

Context-aware systems are an emerging genre of computer systems that help add some forms of intelligence to our surroundings. The ATRACO project uses the ambient ecology metaphor to conceptualize a space populated by connected devices and services that are interrelated with each other, the environment and the people, supporting the users' everyday activities in a meaningful way. Everyday appliances, devices, and context aware artifacts are part of the ATRACO ambient ecologies. In this paper we present the connected home platform adopted by ATRACO and its evolution to provide network adaptation and context-aware services. A flexible and distributed context-aware service model is introduced using the OSGi and UPnP frameworks. UPnP is used to converge the existing network infrastructure comprising of heterogeneous technologies and protocols at the IP level. Furthermore, we introduce a contextaware service model and provide paradigms of context aware services that build upon perceptual and context aware components of the platform.

Tailored services or information for mobile users became very crucial; due to the enormous amount of information and services that are available for mobile users during their various activities. Recommendation system can be merged with context awareness to provide tailored services, information, moreover learning resources for users, based on their current context and profile. Context Aware Recommendation Systems (CARSs) have various usages also many applications and services depend on it, CARSs nowadays become widely used in various domains. This paper presents a survey of CARSs overview, frameworks, techniques, algorithms, and applications. Moreover, the paper discusses the issues of CARSs in general and investigates the proposed solutions.

Context-aware mobile computing aims at designing applications that automatically adapt their behaviors to the available location information and the available nearby sensors and devices. This is done in order to fulfill tasks in a way that suits users' current context best. To achieve this, context representation and manipulation are important issues, so as to establish formal context models. In this paper, basic elements of context-aware systems are described with an emphasis on location information representations. Space models for location-based applications are presented. Considered realistic applications concern intelligent vehicles and pervasive road networks.

While content-based recommendation has been applied successfully in many different domains, it has not seen the same level of attention as collaborative filtering techniques have. In recent years, competitions like the Netflix Prize, CAMRA, and the Yahoo! Music KDD Cup 2011 have spurred on advances in collaborative filtering and how to utilize ratings and usage data. However, there are many domains where content and metadata play a key role, either in addition to or instead of ratings and implicit usage data. For some domains, such as movies, the relationship between content and usage data has seen thorough investigation already, but for many other domains, such as books, news, scientific articles, and Web pages we do not know if and how these data sources should be combined to provide the best recommendation performance. The CBRecSys workshop series aims to address this by providing a dedicated venue for papers dedicated to all aspects of content-based recommendation. The first edition in Silicon Valley in 2014, and the second one in Vienna were a big success. For the third edition, CBRecSys 2016, we once again issued a call for papers asking for submissions of novel research papers addressing recommendation in domains where textual content is abundant (e.g., books, news, scientific articles, jobs, educational resources, Web pages, etc.) as well as dedicated comparisons of contentbased techniques with collaborative filtering in different domains. Other relevant topics included opinion mining for text/book recommendation, semantic recommendation, content-based recommendation to alleviate cold-start problems, deep learning for content representation, as well as serendipity, diversity and cross-domain recommendation. Each submission was rewiewed by three members of the program committee consisting of experts in the field of recommender systems and information retrieval. We selected 9 papers from the 14 submissions for presentation at the workshop. We are also happy to have Prof. Barry Smyth of the University College Dublin and Prof. Bamshad Mobasher of the DePaul Univesity as keynote speakers.

Personalization has emerged as the bedrock technology for digital businesses that seek to deliver exceptional customer experiences, increase interactions, and lift NPS. Right at the center of personalization are the recommender systems, powerful and complex algorithms that, given several inputs, can forecast user habits and recommend what is most likely to be of interest to the user. In this paper, the author draws the contour of using recommender systems for personalization, considering the stages of candidate generation and ranking, the mechanisms of explicit and implicit feedback, and the major difficulties of employing these systems. By methodical examination, this paper discusses the progress of recommender systems and their influence on the field of personals.

The increasing proliferation of mobile devices has raised the expectations for user-customized and environment-aware services. However, mobile context-aware systems inherently feature characteristics of distribution and heterogeneity which pose great challenges to their developers. This paper focuses on context distribution in mobile and ubiquitous computing environments. We propose a peer-to-peer based context distribution approach and evaluate it against the derived requirements.

Context-aware recommender systems (CARS) go beyond traditional recommender systems, that only consider users' profiles, by adapting their recommendations also to users' contextual situations. Several contextual recommendation algorithms have been developed by incorporating context into recommendation algorithms in different ways. The most effective approaches try to model deviations in ratings among contexts, but ignore the correlations that may exist among these contexts. In this paper, we highlight the importance of contextual correlations and propose a correlationbased context-aware matrix factorization algorithm. Through detailed experimental evaluation we demonstrate that adopting contextual correlations leads to improved performance.

Context-aware recommender systems try to adapt to users' preferences across different contexts and have been proven to provide better predictive performance in a number of domains. Emotion is one of the most popular contextual variables, but few researchers have explored how emotions take effect in recommendationsespecially the usage of the emotional variables other than the effectiveness alone. In this paper, we explore the role of emotions in context-aware recommendation algorithms. More specifically, we evaluate two types of popular context-aware recommendation algorithms-context-aware splitting approaches and differential context modeling. We examine predictive performance, and also explore the usage of emotions to discover how emotional features interact with those context-aware recommendation algorithms in the recommendation process.

We describe the process and challenges of integration of movie data from Movie Lens, Netflix and RecSys Challenge 2014 with IMDB and DBPedia. Thanks of this integration we can enhance information by semantic data and improve prediction of customer preferences and recommendation. These data were collected in different situation by different methodologies. We want to use these data to be able to extend and further enhance our machine learning approaches developed for individual datasets to other datasets.

Most of mobile tourism recommender systems take into account mainly the users preferences and location when providing personalized suggestions about what to see in the area (either reactively or proactively). MyMap is a mobile recommender system able to provide personalized presentation of objects of interest in the Tourism Domain that exploits two main knowledge sources: the user profile, that is formalized in order to express situational statements and the current context. In our evaluation experiments we discovered that MyMap suggestions missed of common sense knowledge. For instance: if it is raining recommendations about open-air places should be avoided even if the user strongly prefers these kind of places and has not expressed any preference about alternatives. In this paper, we will illustrate how this functionality has been added to the system and we present the first results of the application of common sense knowledge to the recommendation engine of MyMap.

Large and various amounts of context data related to a user's environment are available from different domains including mobile devices, smarthomes, wearable sensors, and social networking services. These context domains are interconnected and the context data from them can be shared thanks to mobile, pervasive, convergent, and ubiquitous technologies. We can provide user-centric context-aware services by aggregating and associating the diverse types of context data distributed over multiple domains around a user. However, state-of-theart research efforts have been devoted to managing context only in a single domain. In this paper, for user-centric context management, we propose 1) a flexible and extensible technologyneutral information model that can represent generic concepts of context; 2) a hierarchical context management architecture that can understand and manage complex interactions among multiple contextual entities.

Large and various amounts of context data related to a user's environment are available from different domains including mobile devices, smarthomes, wearable sensors, and social networking services. These context domains are interconnected and the context data from them can be shared thanks to mobile, pervasive, convergent, and ubiquitous technologies. We can provide user-centric context-aware services by aggregating and associating the diverse types of context data distributed over multiple domains around a user. However, state-of-theart research efforts have been devoted to managing context only in a single domain. In this paper, for user-centric context management, we propose 1) a flexible and extensible technologyneutral information model that can represent generic concepts of context; 2) a hierarchical context management architecture that can understand and manage complex interactions among multiple contextual entities.

Mobile information appliances are increasingly used in numerous different situations and locations, setting new requirements to their interaction methods. When the user's situation, place or activity changes, the functionality of the device should adapt to these changes. In this work we propose a layered real-time architecture for this kind of context-aware adaptation based on redundant collections of low-level sensors. Two kinds of sensors are distinguished: physical and logical sensors, which give cues from environment parameters and host information. A prototype board that consists of eight sensors was built for experimentation. The contexts are derived from cues using real-time recognition software, which was constructed after experiments with Kohonen's Self-Organizing Maps and its variants. A personal digital assistant (PDA) and a mobile phone were used with the prototype to demonstrate situational awareness. On the PDA font size and backlight were changed depending on the demonstrated contexts while in mobile phone the active user profile was changed. The experiments have shown that it is feasible to recognize contexts using sensors and that context information can be used to create new interaction metaphors.

The consideration of personalization politics in the context of any web application modelling method obliges to the revision of its different modelling activities, which must be adapted to take into account the information regarding the user (usually gathered in a user model) to define aspects such as navigation or presentation. Additionally, they must provide a set of techniques to populate such user model. Finally, and because of the rapid pace at which personalization politics usually change, the modelling process should provide support not only for static personalization rules (known at design time) but also for the definition or change of these rules once the application has been deployed. This article presents, in the context of the Object Oriented Hypermedia Method (OO-H), a personalization framework that fulfils these requirements, and is organized around four main concepts: (1) a set of design activities that capture the personalization requirements known at design time, (2) a mechanism for the specification of personalization rules, defined by means of an XML template, that decouples the definition of the personalization model from the remaining models, (3) an execution architecture that supports the change at execution time of these rules and (4) an extensible repository that includes a set of register mechanisms for the user activity in the system. The possibility of extension of this repository facilitates its adaptation to the particular characteristics of any particular application.

Los sistemas de alta variabilidad son sistemas que representan cientos de configuraciones distintas. En un contexto particular, estas configuraciones pueden ser desplegadas en distintos entornos de despliegue lo cual es una decisión crítica para el correcto funcionamiento de la misma. Por ejemplo, determinar en qué dispositivos móviles se va a ejecutar correctamente nuestra app, es una tarea difícil de resolver. En este artículo presentamos RESDEC, un prototipo de herramienta para asistir al ingeniero de software en la toma de decisiones para el despliegue. Concretamente RESDEC provee algoritmos de recomendación para tres escenarios de despliegue distintos.

Pervasive computing is nowadays becoming a reality, exploiting the capabilities offered by both computing infrastructure and communication facilities. The pervasive computing environment encompasses a multitude of diverse devices, operating systems, protocols, and standards. It includes mobile devices such as cellular phones, smart phones, PDAs, and handheld computers for information access, smart cards, and smart labels for identification and authentication, smart sensors, and actuators that perceive the surroundings and react accordingly. Voice technologies such as automatic speech recognition (ASR), text to speech (TTS) and VoiceXML enable the construction of convenient user interfaces and Web services are a key mechanism for interoperability. Wireless wide area networking allows long distance communication through cellular radio while wireless local and personal area networking and standards such as the Wi-Fi, Bluetooth and IrDA allow short distance communication through radio w...

The Semantic Web is built on top of Knowledge Organization Systems (KOS) (vocabularies, ontologies, concept schemes) that provide a structured, interoperable and distributed access to Linked Data on the Web. The maintenance of these KOS over time has produced a number of KOS version chains: subsequent unique version identifiers to unique states of a KOS. However, the release of new KOS versions pose challenges to both KOS publishers and users. For publishers, updating a KOS is a knowledge intensive task that requires a lot of manual effort, often implying deep deliberation on the set of changes to introduce. For users that link their datasets to these KOS, a new version compromises the validity of their links, often creating ramifications. In this paper we describe a method to automatically detect which parts of a Web KOS are likely to change in a next version, using supervised learning on past versions in the KOS version chain. We use a set of ontology change features to model and predict change in arbitrary Web KOS. We apply our method on 139 varied datasets systematically retrieved from the Semantic Web, obtaining robust results at correctly predicting change. To illustrate the accuracy, genericity and domain independence of the method, we study the relationship between its effectiveness and several characterizations of the evaluated datasets, finding that predictors like the number of versions in a chain and their release frequency have a fundamental impact in predictability of change in Web KOS. Consequently, we argue for adopting a release early, release often philosophy in Web KOS development cycles.

Abstract. This paper describes some of the complexity that needs be captured in any location model of an intimate environment. We take the position that a location model for such an environment must be part of a wider framework for handling context that includes ...

Access to relevant and accurate information is at the heart of tourism, more so in this era of the Internet information overload has become a prevalent phenomenon and as such a serious issue for those seeking for appropriate information. Furthermore, various researches have been carried out on how to make information on tourism website more effective. Intelligent tourism management system tries to bridge the gap by noting what a tourist perceives as relevant, in terms of content pertaining to tourism products in tourism websites. This study focuses mainly on content because it is seen as the key factor associated with an effective website. Hence, the aim of this research entails the design and implementation of an intelligent platform that will assist tourists in gaining access to information on tourist locations in Nigeria. In view of the forgoing, the system was implemented using Rational Unified Process as the adopted software development process, whereas MySQL, HTML and PHP were...

We propose an improved learning model for non-negative matrix factorization in the context-aware recommendation. We extend the collective non-negative matrix factorization through hybrid regularization method by combining multiplicative update rules with Barzilai-Borwein optimization. This provides new improved way of learning factorized matrices. We combine ratings, content features, and contextual information in three different 2-dimensional matrices. We study the performance of the proposed method on recommending top-N items. The method was empirically tested on 4 datasets, including movies, music, and mobile apps, showing an improvement in comparison with other state-of-the-art for top-N recommendations, and time convergence to the stationary point for larger datasets.