A method to monitor a BBS using feature extraction of text data

Abstract

Electronic bulletin board (BBS) has a problem which is action to obstruct communications (the authors use "obstruction on bulletin board"). The authors introduced system that aids an operator to find such obstruction based on feature extraction of words which causes harmful psychological effects on readers. We propose a new evaluation method using "Ruination Figure" which is calculated by feature extraction. This index is displayed in two ways of stock chart for BBS operator to see trend of the index and find harmful obstruction in the BBS. © Springer-Verlag Berlin Heidelberg 2005.