统计与神经网络相结合的词义消歧模型

介绍了一种基于BP神经网络和统计方法相结合的有导词义消歧模型，阐述了BP神经网络原理，通过对使用这种混合人工智能的消歧模型的可能性和优越性进行了讨论，最后通过试验发现实际和预测结果的误差并不随着试验迭代次数而递减，而是实际误差随着次数的增加在零的附近呈现波动状态，即使用很少的迭代次数也可以得到比较好的结果。

The Model of Word Sense Disambiguation Combining Statistics and BP Neural Networks

This paper firstly introduced a Chinese WSD model which was combining the BP neural networks and statistics method,and then discussed the feasibility and advantage of this WSD model.At the last,it was found that the error between the actual and predictive gather was fluctuant through the experiment,namely,the experiment error hadn't the notable trend to zero by increasing iterative times,and it could get the better results from the less iterative times.Accordingly,the BP neural network model had the good application foreground in WSD.