Context relevance assessment and exploitation in mobile recommender systems

Abstract

In order to generate relevant recommendations, a context-aware recommender system (CARS) not only makes use of user preferences, but also exploits information about the specific contextual situation in which the recommended item will be consumed. For instance, when recommending a holiday destination, a CARS could take into account whether the trip will happen in summer or winter. It is unclear, however, which contextual factors are important and to which degree they influence user ratings. A large amount of data and complex context-aware predictive models must be exploited to understand these relationships. In this paper, we take a new approach for assessing and modeling the relationship between contextual factors and item ratings. Rather than using the traditional approach to data collection, where recommendations are rated with respect to real situations as participants go about their lives as normal, we simulate contextual situations to more easily capture data regarding how the context influences user ratings. To this end, we have designed a methodology whereby users are asked to judge whether a contextual factor (e.g., season) influences the rating given a certain contextual condition (e.g., season is summer). Based on the analyses of these data, we built a contextaware mobile recommender system that utilizes the contextual factors shown to be important. In a subsequent user evaluation, this system was preferred to a similar variant that did not exploit contextual information.keywords Context Á Collaborative filtering Á Recommender system Á Mobile applications Á User study 1 Introduction

Impact of the Context Relevancy on Ratings Prediction in a Movie-Recommender System

“…We proposed the Freeman-Halton statistical test with power analysis for the detection of the relevant contextual information when other tests fail to reach the conclusion. We also proposed incorporating context in ratings prediction by adding parameters describing contextuser interaction, based on the method in [7,26].…”

“…Social context was also used in [16]. Weather, mood and temperature, among others, were used in the tourist domain [26].…”

“…The authors in [15] used the χ 2 test for the detection of the relevant context; however, this test could be inappropriate for small databases, i.e., for new systems and the cold-start problem, as we will explain later in the article. In [26] a contextrelevance assessment was conducted to determine the influence of some contextual conditions on the users' ratings in the tourist domain, by asking users to imagine a given contextual condition and evaluate the influence of that condition. However, as they state, such an approach is problematic, since the users rate differently in real and …”

“…where λ is the constant that controls the regularization, and To incorporate contextual information in the matrix factorization, we used an algorithm described in [7,26], and a slightly modified one. As an extension to matrix factorization the authors added parameters that model the interaction between context and the items:…”

ANT: Agent Stigmergy-Based IoT-Network for Enhanced Tourist Mobility

“…Then they prepare a personalized itinerary [8-11, 20, 21] and in some cases they receive user feedback, which provides the possibility of recommendation [9][10][11]21]. The systems presented in [9,10,22,23] have mobility support and are also location-aware, assisting the user with data of the closest POIs. Cellular broadband connections have enabled the development of mobile augmented reality-based (MAR) POI recommendation systems.…”