Word Segmentation as Graph Partition
This addresses word segmentation for Chinese language processing, presenting a novel graph-based method.
The authors tackled Chinese word segmentation by modeling sentences as undirected graphs of characters and using spectral graph partition algorithms, achieving inspiring results on electronic health records and benchmark datasets.
We propose a new approach to the Chinese word segmentation problem that considers the sentence as an undirected graph, whose nodes are the characters. One can use various techniques to compute the edge weights that measure the connection strength between characters. Spectral graph partition algorithms are used to group the characters and achieve word segmentation. We follow the graph partition approach and design several unsupervised algorithms, and we show their inspiring segmentation results on two corpora: (1) electronic health records in Chinese, and (2) benchmark data from the Second International Chinese Word Segmentation Bakeoff.