Topic Identification in Chinese Discourse Based on Centering Model

In this article we are concerned with identifying topics of utterances in texts, which are discourse elements reflecting the links between a sentence and its context. The information carried by the topics can be used to contribute to a number of natural language processing applications, such as information retrieval, text categorization and discourse segmentation etc. However, the phenomenon of zero anaphora frequently occurs in Chinese texts, topics in utterances are frequently omitted from expressions, due to their prominence in discourse. For solving the problem of zero anaphora, we employ the key elements of the centering model of local discourse coherence to extract structures of discourse segments and then identify the topics of utterances in discourse.