An Intelligent Tool for Selecting Abstracts in MEDLINE
Miyako Tanaka, Hiroshi Matsuno, Sanae Nakazono, Satoru Miyano · Proceedings Genome Informatics Workshop/Genome informatics · 1999
Without intensive reading of abstracts by experts, it is almost impossible to decide if a given MEDLINE record is a target article or not. However, there may be a way to reduce the hard task of experts in selecting right articles from MEDLINE. For this purpose, we have developed an intelligent tool for selecting MEDLINE abstracts whose mechanism is based on the iterative method proposed in the paper [3]. This paper reports that, with this intelligent system, 90% of target abstracts can be selected while leaving half amount of abstracts unread with the assistance of the machine learning system BONSAI [1].