Abayomi Moradeyo Otebolaku

Recognizing High-Level Contexts from Smartphone Built-In Sensors for Mobile Media Content Recommendation

Context Recognition is an important element for developing context aware mobile applications. However, context is mostly available as low-level sensor data that are in form not suitable for mobile applications. In this paper, we present a process that uses classifiers for recognizing high-level contexts from low-level sensor data. The process demonstrates accurate recognition of user activity contexts, using smart-phone built-in sensors. We describe and illustrate our context recognition model and then demonstrate its application in a context aware mobile multimedia recommendation system.

On Modeling Adaptation in Context-Aware Mobile Grid Systems

Mobile and smart devices have the potential to form part of Grid infrastructures. These devices can serve two important functions in Grid environment either as service consumer or as valuable service providers. However, there are a number of challenging issues in using a mobile device as service consumer in Grid systems when this fluctuating mobile environment is taken into consideration. Intermittent connectivity, bandwidth fluctuations, high response time between service request and delivery, device heterogeneity and weak security among others. Among these, this paper focuses on modeling a context aware solution to address the problem of high response time in this environment which is caused by some execution context variations. A model of an adaptation mechanism is proposed that monitors execution context of the Grid clients, it evaluates the monitored contexts and decides on what reconfiguration decision to take. The overall impact is to reduce the time between when a service request is made and when the service is delivered to mobile clients.

Supporting Context-Aware Cloud-Based Media Recommendations for Smartphones

The unprecedented advancements in broadband and mobile networks, the proliferation and the incredible appeal of smart devices such as smartphones, and the recent emergence of cloud computing are poised to drive the next generation of ubiquitous media delivery and consumption. As more media services become available, mobile users will waste invaluable time, seeking relevant media items. Therefore, to deliver relevant media services, with rich experience to mobile users, media service providers must consider the services that match users contextual consumption choices. This paper proposes context-aware recommendation techniques to support the delivery of contextually relevant cloud-based media items to mobile users. The recommendation service works with a contextual user profile service, which relates user preferences to contexts in which such preferences are expressed, relying on a context recognition service, which identifies the users dynamic contextual situation from smartphone built-in sensors. Experimental evaluations, using real world user and online movie data, established that the context-aware recommendation techniques are promising.

Context Representation for Context-Aware Mobile Multimedia Content Recommendation

Very few of the current solutions for content recommendation take into consideration the context of usage when analyzing the preferences of the user and issuing recommendations. Nonetheless, context can be extremely useful to help identify appropriate content for the specific situation or activity the user is in, while consuming the content. In this paper, we present a solution to allow content-based recommendation systems to take full potential of contextual data, by defining a standards-based representation model which accounts for possible relationships among low-level contexts. The MPEG-7 and MPEG-21 standards are used for content description and low-level context representation. OWL/RDF ontologies are used to capture contextual concepts and, together with SWRL to establish relationships and perform reasoning to derive high-level concepts the way humans do. This knowledge is then used to drive the recommendation and content adaptation processes. As a side achievement, an extension to the MPEG-21 specification was developed to accommodate the description of user activities, which we believe have a great impact on the type of content to be recommended.

CAAM: A Context Aware Adaptation Model for Mobile Grid Service Infrastructure

Mobile and smart devices can play important roles in service provisioning and consumption in a grid-based service infrastructure. However, these devices and their unstable network are plagued with challenging issues that have made their integration with grid infrastructure a serious burden. Intermittent connectivity, bandwidth fluctuation, low memory, among others, cause high responsiveness in service delivery when these devices are used for service access. Adaptation capabilities can be very useful at addressing these issues. Therefore, this paper presents a utility function based context aware adaptation model (CAAM) that can monitor the service clientpsilas context, and uses a utility function based algorithm to take self reconfiguration decision in order to adapt the interaction between service consumers and service providers. Results from simulation experiments carried out on the model proved that adaptation is very useful at addressing problem of high responsiveness thereby improving on the quality of service.

Context-Aware Personalization Using Neighborhood-Based Context Similarity

With the overwhelming volume of online multimedia content and increasing ubiquity of Internet-enabled mobile devices, pervasive use of the Web for content sharing and consumption has become our everyday routines. Consequently, people seeking online access to content of interest are becoming more and more frustrated. Thus, deciding which content to consume among the deluge of available alternatives becomes increasingly difficult. Context-aware personalization, having the capability to predict user’s contextual preferences, has been proposed as an effective solution. However, some existing personalized systems, especially those based on collaborative filtering, rely on rating information explicitly obtained from users in consumption contexts. Therefore, these systems suffer from the so-called cold-start problem that occurs as a result of personalization systems’ lack of adequate knowledge of either a new user’s preferences or of a new item rating information. This happens because these new items and users have not received or provided adequate rating information respectively. In this paper, we present an analysis and design of a context-aware personalized system capable of minimizing new user cold-start problem in a mobile multimedia consumption scenario. The article emphasizes the importance of similarity between contexts of consumption based on the traditional k-nearest neighbor algorithm using Pearson Correlation model. Experimental validation, with respect to quality of personalized recommendations and user satisfaction in both contextual and non-contextual scenarios, shows that the proposed system can mitigate the effect of user-based cold-start problem.

Context-Aware Media Recommendations

Media content recommendations for a mobile user based on his changing contextual preferences, otherwise called context-aware media recommendations, constitute a very important challenge. Context-aware media recommendation systems take context information such as user preferences, activities, time, location, device, and network capabilities as inputs for media recommendations, whereas the traditional recommendation systems use only user preferences in the form of ratings to deliver media recommendations. This paper presents a generic high-level architecture of context-aware recommendations, discussing its key techniques and solutions, which are based on context acquisition, recognition, and representations, using MPEG-21 and ontology model, and a contextual user profiling process, as well as MPEG-7 for media description model and media presentation adaptation.

Context-Aware User Profiling and Multimedia Content Classification for Smart Devices

Current solutions for delivering adapted multimedia content in mobile environments take into account only a limited set of contextual information, and can be regarded as passive solutions. We propose a new solution that anticipates users needs based on the contexts of use and preferences to deliver media content to mobile users. This paper describes the profiling approach of the proposed solution, and a context-aware content-based recommendation for smart devices. Recommendations are issued based on user history, driven by real-time contextual conditions.

A Context-Aware Framework for Media Recommendation on Smartphones

The incredible appeals of smartphones and the unprecedented progress in the development of mobile and wireless networks in recent years have enabled ubiquitous availability of myriad media contents. Consequently, it has become problematic for mobile users to find relevant media items. However, context awareness has been proposed as a means to help mobile users find relevant media items anywhere and at any time. The contribution of this paper is the presentation of a context-aware media recommendation framework for smart devices (CAMR). CAMR supports the integration of context sensing, recognition, and inference, using classification algorithms, an ontology-based context model and user preferences to provide contextually relevant media items to smart device users. This paper describes CAMR and its components, and demonstrates its use to develop a context-aware mobile movie recommendation on Android smart devices. Experimental evaluations of the framework, via an experimental context-aware mobile recommendation application, confirm that the framework is effective, and that its power consumption is within acceptable range.

Towards context classification and reasoning in IoT

Internet of Things (IoT) is the future of ubiquitous and personalized intelligent service delivery. It consists of interconnected, addressable and communicating everyday objects. To realize the full potentials of this new generation of ubiquitous systems, IoTs ‘smart’ objects should be supported with intelligent platforms for data acquisition, pre-processing, classification, modeling, reasoning and inference including distribution. However, some current IoT systems lack these capabilities: they provide mainly the functionality for raw sensor data acquisition. In this paper, we propose a framework towards deriving high-level context information from streams of raw IoT sensor data, using artificial neural network (ANN) as context recognition model. Before building the model, raw sensor data were pre-processed using weighted average low-pass filtering and a sliding window algorithm. From the resulting windows, statistical features were extracted to train ANN models. Analysis and evaluation of the proposed system show that it achieved between 87.3% and 98.1% accuracies.