Building a semantic representation for personal information

A typical collection of personal information contains many documents and mentions many concepts (e.g., person names, events, etc.). In this environment, associative browsing between these concepts and documents can be useful as a complement for search. Previous approaches in the area of semantic desktops aimed at addressing this task. However, they were not practical because they require tedious manual annotation by the user. In this work, we suggest a methodology and a prototype system for building a semantic representation of personal information based on click feedback from the user. We employed a feature-based model of associations between the concepts and documents. Our initial evaluation shows that the suggested semantic representation can play an important role in the known-item finding task and that the system can learn to predict such associations with a small amount of click data.