Extraction of Chronological Statistics Using Domain Specific Knowledge.
Tatsukuni Inoue, Takashi Yamamoto, Makoto Toriyabe, Erina Shimizu, Hiroya Susuki, Hiroaki Saito · NTCIR · 2008
This paper reports a system constructed for our participation as a group of “keio01” of Keio University in the T2N (text to number) task at the NTCIR7 MuST (Multimodal Summarization for Trend Information) task. The constructed system uses newspaper article corpora, task description and domain specific knowledge, and the system outputs chronological statistics. The statistics are ternary data which are pairs of a statistic name and pairs of date and value. They are visualized as drawing charts. The system was evaluated by comparison with manually extracted data, and it achieved 0.785 F-score.