Fuzzy Rough Set Model And Its Data Mining Methods

The amount of data is growing exponentially in the modern information society.People feel at a loss what to do when the volume of data grows too fast,therefore,resulting in a huge waste of resources.With the rise and the rapid popularization of knowledge processing technology,the means of ''knowledge acquisition'' is more and more rich and it also makes people find hope in the mastery and application of knowledge.In order to solve the the problem of information loss in the discretization of continuous attributes,fuzzy rough set theory is proposed and successfully applied to deal with the fuzzy,uncertain and inaccurate information.First of all,it not only introduces the significance of the selected topic,but also expounds the research background and research status of fuzzy rough set theory.Secondly,an improved fuzzy rough set model is established on the fuzzy rough set model.A domain is expanded into two domains,and the membership function of fuzzy rough set is redefined.Therefore,the rough approximation precision of the new model is improved effectively and the application range is enlarged.Theoretical analysis and practical calculation all prove the validity and practicability of the new model.At last,based on the attribute dependency degree,a heuristic attribute reduction algorithm is proposed for two domains fuzzy rough set model.The algorithm optimizes the calculation process of the relative normal domain using the stepwise calculation,thus reducing the time complexity of the algorithm.