A Review of Rules Family of Algorithms

In recent years, there has been a growing amount of research on inductive learning. Out of this research a number of promising algorithms have surfaced. In the paper after a brief description of knowledge acquisition, induction and inductive learning; RULES family of inductive learning algorithms, their strengths as well as weaknesses are explained and discussed. The applications of inductive learning and particularly the applications of RULES family of algorithms are overviewed. Key wordsInduction, Inductive Learning, Knowledge Acquisition, Expert Systems 1. KNOWLEDGE ACQUISITION Knowledge-based expert systems consist of two main components: a knowledge base and an inference mechanism. Collecting knowledge to form the knowledge base is the main task in the process of building an expert system [1,2,3]. The process of acquiring knowledge through interaction with an expert consists of a prolonged series of intense, systematic interviews, usually extending over a long period [4]. Human experts are capable of using their knowledge in their daily work, but they usually cannot summarize and generalize their knowledge explicitly in a form which is sufficiently systematic, correct and complete for machine representation and application [1]. Expert systems require large amounts of knowledge to achieve high levels of performance, yet the acquisition of knowledge is slow and expensive [5]. The shortage of trained knowledge engineers to interview experts and capture their knowledge is another problem of knowledge acquisition [6]. The aforementioned problems are not just difficulties of the early days of the technology, but are still acknowledged today as paramount problems. Knowledge acquisition (and in particular machine learning) has become a major area of concern for expert systems research [5,7]. An alternative method of knowledge acquisition exists in which knowledge is learned, or induced, from examples. While it is very difficult for an expert to articulate his knowledge, it is relatively easy to document case studies of the expert's skills at work [5]. Instead of asking an expert to summarize and articulate his knowledge, the main idea of automatic induction is to have him provide a basic structure of his discipline. The knowledge itself will be induced from examples expressed in this structure. Recent developments have proved that this method of knowledge acquisition is entirely possible. Indeed, the main feature of the second generation expert systems is that the knowledge acquisition process is highly automated [8].