中文摘要：

为了提高基于规则的分类法中挖掘规则的效率,提出了将基因表达式编程用于挖掘规则的分类方法.针对规则分类问题,设计出了一种新形式的染色体终端符号,引入规则的正确率作为适应度函数度量;将适应度由高到低排序,建立备选规则集;通过使用基因表达式编程挖掘Monk与Acute Inflammations中的规则,利用挖掘出的规则对数据集进行分类.实验结果表明了基于基因表达式编程的挖掘规则分类算法的准确率会高于传统分类算法.

英文摘要：

Rule-based classification is a commonly used method of data classification.To increase the Mining rubes efficiency of this method when mining rules,using GEP to mine out rules is proposed.Firstly a new pattern of chromosome terminal symbol is designed and the ratio of using rules correctly is taken as the measurement of fitness function,aiming at the problem of rules classification.Then the fitness is sorted in descending order and an alternative set of rules are established.By using GEP to mine out classification rules in the Monk and Acute Inflammations data sets and making use of these rules to classify data sets,the outcome of experiments on these two data sets can be seen that the new method has higher accuracy than those traditional ones when mining data sets.