Personalized Concept-Based Clustering of Search Engine Queries
IEEE Transactions on Knowledge and Data Engineering
| Hi-index | 0.00 |
Imagine a system that can push highly selective information right to our hands when and only when we need it. This requires a mind-reading machine, but unfortunately we don't have one --- yet. User profiling attempts to estimate what is most important to a user at a particular point in time and space. In this talk, I will start with simple raw data such as the users' queries and clicks on the web and places they have visited to estimate what they might be interested in. We further divide user interests into content-based and location-based. We discuss issues involving the transformation of raw activities to conceptual needs, identifying user groups for collaborative filtering and the roles of locations in personalized information delivery.