The Research Of KA's Methods Based On Rough Set And It's Application

Information technology is used widely these years.The amount of information acquired by people from every domain increases rapidly.Facing so abundant information resource,but people acquire the knowledge difficulty.Rough set theory is a sort of availability math tool to deal with fuzzy incertitude knowledge.It can analyze the data and found the connotative knowledge without any known knowledge.In this article the rough set is used in knowledge acquire.Data discretization,attribute reduction,rules extract are discussed greatly which are used to herbalist doctor diagnose system.The works in this article as follows:1) Due to the commonly discretization method can't assure the swatches differentiate relationship in old information system,and the effect of discretization is not good.So in this article a new data discretization method based on rough set is brought forward.The cuts core is the jumping-off point of this algorithm and cuts core frequency in the discernibility matrix is used as heuristic information in this algorithm. Then the most important cut is selected,that is added to the cuts gather every time. The presented example validates this algorithm that has a good discretization effect.2) A heuristic algorithm of attribute reduction is brought forward.The condition entropy is used as heuristic information for reducing the searching room.And a process of reducing again is used in the improved algorithm in order to eliminate redundancy attributes.The example validates the improved algorithm with good reduction effect without increasing computing time.At the same time,due to commonly attribute reduction algorithms can't settle the incompatibility decision making tables.An attribute reduction algorithm on incompatibility decision making tables is presented.The discernibility matrix is the foundation of the algorithm to deal with the incompatibility decision making tables;the attribute frequency is used as heuristic information in this algorithm,reducing the computing time availably.3) A concept of attribute value is presented and a new method of value reduction based on discernibility matrix is presented.The important of attribute value is used as heuristic information that reduces the computing time without visiting the tables many times.At the same time,the method makes attribute value reduction with more reason that makes rules with higher degree of confidence.4) According to above theory research,the algorithms presented in this article are used to the actual herbalist doctor diagnose system.Which can help doctors make quickly and accurate decision-making(diagnose).