UMASS Approaches to Detection and Tracking at TDT2

The following work describes our solutions to the detection and tracking problems defined by the Topic Detection and Trackin g (TDT2) research initiative. We discuss the implementation and re- sults of the approaches which were recently tested on the TDT2 eval- uation corpus. Our solutions to these problems extend text- based ranked retrieval techniques previously used for document clustering and filtering tasks. We present the effects of different on-l ine hi- erarchic clusterings on the detection task. In addition we c ompare adaptive and static approaches for building linear text cla ssifiers for the tracking task.