Research on Methods of Semantic Disambiguation about Natural Language Processing
Guohuan Lou, Hao Zhang, Wang Honghui · 2009
Natural language processing is one of the most important applications in artificial intelligence (AI), while semantic disambiguation is one of branches and difficulties in natural language processing. This paper introduces three semantic disambiguation models, Bayesian model, hidden Markov model, and maximum entropy model. These three models are used to test and compare with. The results show that the correct rate of disambiguation used by Bayesian model is the best one, the other two are also well. Every model has its own advantages.