本系统试图把解释学习这一机器学习技术应用于汉语处理,其目的是为了解决汉语句子分析中的效用性问题.通过对250个例句的运行。得出了具有120条学习规则的规则集.这样,对大部分句子可绕过常规的分析器,直接用学习规则集进行处理,速度提高约1/3,大大提高了分析效率. This system attempts to apply interpretation learning to machine learning in Chinese processing, and its purpose is to solve the problem of utility in Chinese sentence analysis.Through the operation of 250 examples, We get a rule set with 120 learning rules, so that most of the sentences can be bypassed by conventional analyzers and processed directly by learning rule sets, increasing the speed by about 1/3, greatly improving the analysis efficiency.