The basic problem in using Pinyin (a phonetic system) to enter Chinese characters into a computer lies in the fact that one Pinyin code leads to multiple character choice. The user has to search through many candidates of characters to find the desired one. This is obviously very time consuming.
The purpose of this study is to develop a program that uses the context to increase the matching rate from the Pinyin code to the target words. Three strategies were implemented in the experimental program: (1) using syntax to predict the most probable word classes and to sort the homophones list in an optimal order; (2) creating a large internal dictionary to accommodate longer inputting contexts; (3) tracking frequency of word usage in run time to improve the target prediction further. Of the 579 words used in the test samples, the experimental program successfully converted 82.38% of Pinyin codes into Chinese words without user intervention. This is 9.15% higher than the commercial software program used for comparison.