Error-Driven Adaptive Language Modeling for Chinese Pinyin-to-Character Conversion

摘要

The performance of Chinese Pinyin-to-Character conversion is severely affected when the characteristics of the training and conversion data differ. As natural language is highly variable and uncertain, it is impossible to build a complete and general language model to suit all the tasks. The traditional adaptive MAP models mix the task independent data with task dependent data using a mixture coefficient but we never can predict what style of language users have and what new domain will appear. This paper presents a statistical error-driven adaptive language modeling approach to Chinese Pinyin input system. This model can be incrementally adapted when an error occurs during Pinyin-to-Character converting time. It significantly improves Pinyin-to-Character conversion rate.