目前,机器学习研究方兴未艾,同时,对一些基本概念、基本问题的提法尚各执一词。本文首先概述了我们关于机器学习有关问题的基本认识,在此基础上,分析了符号机器学习研究中比较重要的两个领域(归纳学习和基于解释的学习)的研究现状,最后提出了我们的一些想法。关于机器学习的一些看法何谓机器学习?我们认为,机器学习从内在行为看,是从未知到知的过程,是知识增加的过程。从外在表现看,是系统的某些适应性改变,使得系统能完成原来不能完成的任务或把原来能完成的任务做得更好。这是一个问题的两个方面,各有侧重,归纳学习就侧重于概念获取即知识增加的一面,而基于 At present, the research on machine learning is still in the ascendant. At the same time, the formulation of some basic concepts and basic problems still has its own meaning. This paper first outlines our basic understanding of machine learning related problems. Based on this, we analyze the research status of the two most important areas in the study of symbol machine learning (inductive learning and interpretation-based learning), and finally put forward our some thoughts. Some views on machine learning What is machine learning? We think that machine learning from the internal behavior of the unknown process is to know, is the process of increasing knowledge. From the external performance point of view, it is the system of some adaptive changes, making the system can not do the original task can be completed or the original task can be done better. This is one of the two aspects of a problem, each focusing on inductive learning focuses on the concept of gain that knowledge is added, and based on