Personalized search in digital libraries via spreading activation model

With the tremendous development of information technology, the volume of data in digital libraries is increasing enormously, and the magnitude is putting users at risk of information overload. Personalized search aims at solving the problem by tailoring search results for individual demands. We propose a novel approach to provide personalized search by learning users' interests from historic behaviors and re-ranking search results by a Spreading Activation SA model. In our approach, users' interests are categorized by the level of recency to form user profiles, which in turn serve as the input of the SA model. Then, SA runs on the domain ontology incorporated with a newly defined relationship borrowIntent derived from the assumption of collaborative filtering. We further demonstrate the effectiveness of the proposed methodology through experiments on real data from a university library. The presented approach can also be applied in other contexts such as electronic commerce.