AntSeg: an Ant Approach to Disambiguation of ChineseWord Segmentation

Abstract

The ambiguity of Chinese word segmentation is one of the pivotal factors that affect the segmentation precision. Disambiguation is such a hard problem for Chinese information processing (CIP) that it attracts many researchers. In order to present a new solution to disambiguation as well as to the related researches in CIP, this paper applies an ant colony optimization(ACO) approach, called AntSeg, to the elimination of the segmentation ambiguity. AntSeg converts the traditional segmentation problem into a graph search problem. This paper implements a simple ACO algorithm and conducts a lot of experiments, comparing with the state-of-the-art system. The successful primary results suggest that AntSeg could be a new effective computing framework for solving the Chinese segmentation problem and other natural language processing problems