Information Extraction Systems of BUPT_PRIS at TREC2014 Temporal Summarization Track.

摘要

This paper describes the information extraction systems of BUPT_PRIS at Temporal Summarization Track, Which includes data obtaining and preprocessing, index building and query expansion, sentences scoring module. This year only keep one task: sequential update summarization, the task: value tracking is cancelled. For the sequential update summarization we focus attention on queries expansion and sentence scoring. There are three methods of query expansion introduced in this report: WordNets, Word representation, spatial analysis method. We also show the evaluation results for our team and the comparison with the best and median evaluations.