Discovering User Behavioral Features to Enhance Information Search on Big Data

Due to the emerging Big Data paradigm, driven by the increasing availability of intelligent services easily accessible by a large number of users (e.g. social networks), traditional data management techniques are inadequate in many real life scenarios. In particular, the availability of huge amounts of data pertaining to user social interactions, user preferences and opinions, calls for advanced analysis strategies in order to understand potentially interesting social dynamics. Furthermore, heterogeneity and high speed of user generated data require suitable data storage and management tools to be designed from scratch. This pa- per presents a framework tailored for analyzing user interactions with intelligent systems while seeking for some domain specific information (e.g., choosing a good restaurant in a visited area). The framework enhances user quest for information by exploiting previous knowledge about their social environment, the extent of influence the users are potentially subject to and the influence they may exert on other users. User influence spreads across the network are dynamically computed as well to improve user search strategy by providing specific suggestions, represented as tailored faceted features. Such features are the result of data exchange activity (called data posting) that enriches information sources with additional background in- formation and knowledge derived from experiences and behavioral properties of domain experts and users. The approach is tested in an important application scenario such as tourist recommendation but it can be profitably exploited in several other contexts, e.g., viral marketing and food education.