Activity Based Links as a Ranking Factor in Semantic Desktop Search

Existing desktop search systems, developed as a reaction to the rapidly increasing storage capacities of our hard disks, are an important step towards more efficient personal information management, yet they offer an incomplete solution. The industrial ones are capable of indexing a vast amount of file types, and the research prototypes can add very comprehensive metadata to increase recall. However, they all rely only on basic ranking techniques, as TFxIDF, which are comparable to the first generation of Web search engines. In this paper we propose to connect semantically related desktop items by exploiting activity information about accesses to local resources. We investigate and evaluate in detail the possibilities to translate this information into a desktop linkage structure, and we propose several algorithms that exploit these newly created links in order to efficiently rank desktop items. Finally, we show that the ranking results based on the user activity links surpass both TFxIDF ranking scores, as well as the rankings generated using the schema based approaches we devised in our previous work.