Combining classifiers for Chinese word segmentation

In this paper we report results of a supervised machine-learning approach to Chinese word segmentation. First, a maximum entropy tagger is trained on manually annotated data to automatically labels the characters with tags that indicate the position of character within a word. An error-driven transformation-based tagger is then trained to clean up the tagging inconsistencies of the first tagger. The tagged output is then converted into segmented text. The preliminary results show that this approach is competitive compared with other supervised machine-learning segmenters reported in previous studies.