Study On Incremental Data Mining Technology Based On Rough Set Theory

Rough set theory is a mathematic tool to study information and knowledge which is imprecise, uncertain and incomplete information. Since the Rough Set theory was put forward, plentiful achievements have been made both in theory research and application. It has been applied to many fields successfully, such as machine learning, knowledge acquisition, decision analysis, knowledge in databases, expert systems and pattern recognition etc. Now, Rough Set theory is becoming a new hotspot of artificial intelligence domain. More and more scholar focuse on the incremental data mining technology based on rough set theory.At first, this paper summarizes the basic concepts and characters of Rough Set theory in detail, both and its development and academic fruit.Secondly, the paper analyzes and sums up the thinking and flow of classical rough set data reduction arithmetic. The existing arithmetic of attribute reduction and value reduction is analyzed, especially the reduction arithmetic based on the discernibility matrix which is also summed up integrally. On this bedrock, there are two betterment thinking are advanced.In succession, the technology of incremental data mining based on Rough Set theory is worked over. The new think of incremental attribute reduction just as the question solving the incremental data mining is brought forward, the question recurs to distributing and combination data mining arithmetic.According to in-depth study, the arithmetic of incremental data mining based on dynamic database is comprehended. Then the paper discusses the Rough Set theory based arithmetic of incremental attribute reduction, analyzing the exiting arithmetic of incremental data mining of dynamic database combined with the data mining arithmetic being based upon discernibility matrix. Via the deep compare and analysis, the corresponding arithmetic is optimized and come true by program. It validates that the improved arithmetic compared with the inhere arithmetic has distinctness amelioration both in time complexity and performing efficiency.