A Menu-Based Content Search System Based on Relationships between Mobile User Context and Information Needs

Due to the popularization of smart phones, many people search contents in various situations (e.g., walking or sitting). It is known that user´s search objective is affected by the user´s situation such as time and location. Meanwhile, when users are walking, they cannot pay enough attentions to operate their devices. To assist contents search for mobile users, in our previous work, we proposed a contents search system that presents a few options that enable user to get desired contents easily according to their situations. However, in this system, we manually constructed a tree structured ontology which is used for the system to infer user´s situation and search objective based on sensory data. Thus, it was difficult to handle a large number of situations and could not guarantee the accuracy of the constructed ontology. To solve these problems, in this paper, we extend our system to automatically learn the relationships between user context and information needs by using a Bayesian network. Through user experiments, we confirmed that our system could learn the relationships and infer users´ situation effectively.