Research On Method Of Knowledge Acquisition Based On Rough Set

Since rough set theory was proposed in the early 80s of last century, it has developed very rapidly. As an important mathematical tool of dealing with ambiguity and uncertainty knowledge, researchers have paid more and more attention to it. It has been widely used in data mining, machine learning, pattern recognition and other fields. Attribute reduction and rules acquisition are two most important aspect, Based on these ideas,this paper systematically studied attribute discretization and rule extraction for information system and achieved some results. The major results as follow:(1) Proposed a kind of distinguish matrices that can take full advantage of the characteristics of decision-making information system. It can effectively measure the importance of discrete breakpoints, and constructed a discrete algorithm with importance of discrete breakpoints as heuristic information. Theoretical analysis and experimental results show that the algorithm is effective.(2) Improved a rule acquisition algorithm. The algorithm can handle instances with inconsistent decision tables, extract inconsistent rules. The algorithm use the first order vector's distinguish degree as heuristic information, can gradually reduce the required storage space and time complexity of the algorithm is low, therefore it is suitable for handling large data sets. Experimental results show the effectiveness of the algorithm.These research results will enrich the rough set theory and provide effective knowledge acquisition techniques for large-scale complex data.