Information Finding from Biological Papers

Yoshihiro Ohta [1] (
Yasunori Yamamoto [2] (
Ikuo Uchiyama [1] (
Toshihisa Takagi [1] (

[1] Human Genome Center
Institute of Medical Science, University of Tokyo
Shiroganedai, Minato-ku, Tokyo 108, Japan
[2] Graduate School of Information Science and Engineering,
Tokyo Institute of Technology Oookayama, Meguro-ku, Tokyo 152, Japan


We have developed computer technologies for a system that extracts domain specific knowledge from human written biological papers. This system consists of two components, Information Retrieval (IR) and Information Extraction (IE). We propose a query modification method using automatically constructed thesaurus for IR and a statistical keyword prediction method for IE. Although by a purely statistical model with no heuristics, the experimental result has shown the good performance.