Study on Multiple Classifiers for Chinese Word Sense Disambiguation

In this paper, a new method of multiple layer classifiers integration based on single classifier is proposed which called Auto Weight Adjust. In the most used classifiers, Maximum Entropy (ME) model has excellent performance, and Naïve Bayesian (NB) is preferred by researchers for it’s simple and useful. So in our experiments we chose ME and NB as single classifiers and use the ME classifier result and the NB classifier result to fuse the final result. We use People Daily News (PDN) datasets to test our model, according to experiments our algorithm leads to less error and better performance than other algorithms. It’s outside test accurate reach to 0.88798.