Identification Of Complex Names Of Cambodian Institutions Based On Markov Logic Network

With the increasingly frequent exchanges and cooperation between China and Khmer,the work of Khmer's Natural Language Processing has become particularly important.Because of the great differences between different languages,the named entity recognition method of other languages can not be transplanted directly into Khmer.In order to improve the recognition accuracy of Khmer organization names,this paper constructed around the Khmer organization name recognition model,carried out research on key issues of extension organization name corpus,and has achieved the following results:(1)This paper proposes a method to recognition the Khmer organization name based on Tri-training.This method uses the improved Tri-training algorithm,conditional random fields,support vector machine and maximum entropy model for three different classifiers are combined into a classification system based on corpus,and then use a small amount of sample selection,on the basis of the optimization strategy to select the newly added samples,combined with the Khmer linguistic features.The results show that this method can realize the recognition of Khmer organization name by using a small amount of annotated corpus.(2)A name recognition method based on Markov logic network for the Khmer complex organization name is proposed.This method first uses conditional random field model of simple organization name recognition,and then combined with the linguistic features of Khmer,get the first-order logic rules into first-order logic rules to Markov logic network,and to recognition complex organizations name using LazySAT inference algorithm.The results show that the proposed method can achieve a better recognition effect of Khmer complex organization name.(3)The design and implementation of a prototype system for Khmer organization name recognition,which provides a strong support for the study of Khmer named entity recognition.