论文部分内容阅读
本文提出以实例空间中状态划分概率的大小作为启发式信息，以提供的正反实例集为依据，基于二叉树分类方法的示例式归纳学习算法ＣＡＰ２．它输出的分类规则是谓词演算表达式．该算法可根据用户对精度的要求控制分类深度，得到不同精度的规则，并能处理连续数据、噪音数据和利用用户提供的背景知识，既适用于同时给定概念的正、反例集的情况，也适用于只给正例集的情况．本文还介绍了ＣＡＰ２算法的应用情况，并和著名的ＩＤ３算法进行了比较．ＣＡＰ２已嵌入到一个自动知识获取系统．
This paper proposes an example inductive learning algorithm CAP2 based on binary tree classification based on the probability distribution of state partitioning in instance space as heuristic information and on the basis of the provided positive and negative instance sets. The classification rules it outputs are predicate calculus expressions. The algorithm can control the depth of classification according to user’s requirements of precision, get different precision rules, and can process continuous data, noise data and user-provided background knowledge, which is applicable to both positive and negative examples of given concepts, It also applies to positive cases only. This article also introduces the application of CAP2 algorithm and compares it with the famous ID3 algorithm. CAP2 has been embedded into an automated knowledge acquisition system.