A Chinese Pin Yin-input method for association of input word using inductive learning

摘要

We introduce an efficient Chinese PinYin input method of intelligent association using Inductive Learning. As for the Chinese PinYin input system, a large number of keystrokes and the selection from candidate words limit the speed increasing. Association of input words is one of the most important methods for Chinese character input to decrease keystrokes and to improve the input speed. Our proposed method enables us to promptly input the Chinese words into a computer; that is, not necessary to spell complete character of PinYin only by choosing the words from candidates. Moreover, the Inductive Learning approach gives system two characteristics. One is the dynamically adaptation to current situation, and the other is the capability of acquisition rules even if no initial rules. Furthermore, we make use of the relationship between Chinese words and characters in context to generate a rule. When the correct association rate is 40%, the decreasing rate of keystrokes is 39% in the experiment for the performance evaluation. It shows that this method is effective for Chinese input.