A sort Approach for Anaphora Resolution of Chinese Personal Pronoun Based on Machine Learning Method

Anaphora occurs throughout discourse or dialogue. Their high frequencies make anaphora resolution one key problem in discourse processing which attract attention of increasing researchers. In the paper, according to features of Chinese personal pronoun we present an approach which is based on corpus. It adopts maximum entropy method. Then we discovered characteristic of pronoun anaphora referring to inhuman is different form that referring to people. So the paper put forward a new approach about processing the Chinese personal pronoun anaphora. The approach divided the anaphora resolution system into two parts, the PARS subsystem and the IARS subsystem. They will separately process the personal pronoun anaphora referring to people and that referring to inhuman. The paper described the design and the realization of the system, tests the new system in the Chinese Tree Bank and evaluates the arithmetic in the round. The experiment demonstrates that the method achieves the desired result.