Papers by Keyword: Rule Extraction

Abstract: A model of interval-valued rough fuzzy set combining interval-valued fuzzy set and rough set is investigated in this paper. Firstly, considering the deficiency of general sorting method between any interval-valued fuzzy numbers, an improved sorting method and a pair of new approximation operators about minimum and maximum are presented. Based on the improved operators, a model of interval-valued rough fuzzy set is established. At last, by using the modified model of interval-valued rough fuzzy set, a method of knowledge discovery in interval-valued fuzzy information systems is investigated.

Abstract: A new rule extraction algorithm based on convex hull for strip hot-dip galvanizing process monitoring is proposed in this paper. It overcomes the black-box problem of support vector machine. The zinc coating weight is used as the investigated subject. The sample datasets are trained by support vector machine rule extraction method, and the quantitative relationship can be obtained in the form of knowledge rules among input variables (such as the parameters of raw materials and control parameters of production) and output ones (the quality parameters), with which the production control parameters can be set and updated easily.

Abstract: Rule extraction is a main goal for rough set theory. This paper mainly constructs a new algorithm (LBRM Algorithm) for rule extraction based on rough membership. The confidence principle is established based on rough membership. Thus, LBRM Algorithm is proposed by utilizing discretization and clearness strategies under the fuzzy environment, and is applied to both interval rules and general rules in fuzzy classification. LBRM Algorithm effectiveness is illustrated by a medical example. In particular, LBRM Algorithm integrates the confidence on both previous LBR Algorithm and fundamental rough membership, and has some improvements on rule confidence.

Abstract: Rough sets theory (RST) is a new data mining method that effectively deals with the problems with inexact, uncertain or vague knowledge in a complex information system. This paper investigates knowledge discovery methods from the textile industrial database, and then presents a RST-based intelligent control model (ICM) for spinning process. In order to analyze the yarn strength when the characteristics of fibers are given, a rule extraction method based on RST is researched. The logical rules extracted from the decision table indicate that the initial strength of fibers is a key factor influencing on the yarn strength. At the same time, the different values combination of the final reduced attributes also obviously influence on the yarn strength in different degree when the certain nominal yarn is being processed. Therefore, RST method can be taken into account for spinners to choose suitable fiber materials in order to ensure the quality and reduce cost.

Abstract: Due to absence of an integral mathematical model, quality control in spinning process has been hard problem for a long time. Rough sets theory (RST) is a methodology that effectively deals with the problems with inexact, uncertain or vague knowledge in a complex information system. Considering a mass of data from spinning process and inspection, as well as the variety of knowledge and experience from domain experts, an RST-based intelligent control model for spinning process is presented in this paper. In order to analyze the yarn strength when the characteristics of fibers are given, a rule extraction method based on RST is researched. The logical rules extracted from the decision table indicate that the initial strength of fibers is a key factor influencing on the yarn strength. At the same time, the different values combination of the final reduced attributes also obviously influence on the yarn strength in different degree when the certain nominal yarn is being processed. Therefore, RST method can be taken into account for spinners to choose suitable fiber materials in order to ensure the quality and reduce cost.

Abstract: The rate of false positives which caused by the variability of environment and user behavior limits the applications of intrusion detecting system in real world. Intrusion detection is an important technique in the defense-in-depth network security framework and a hot topic in computer security in recent years. To solve the intrusion detection question, we introduce the self-organizing map and artificial immunisation algorithm into intrusion detection. In this paper, we give an method of rule extraction based on self-organizing map and artificial immunisation algorithm and used in intrusion detection. After illustrating our model with a representative dataset and applying it to the real-world datasets MIT lpr system calls. The experimental result shown that We propose an idea of learning different representations for system call arguments. Results indicate that this information can be effectively used for detecting more attacks with reasonable space and time overhead. So our experiment is feasible and effective that using in intrusion detection.