Study On Knowledge Reduction Based On Rough Set And Its Application In Continuing Education

In modern times, society has entered the internet information age, the teacher training mode of the Primary and Secondary Teachers'Continuing Education has been changing from face-to-face mode into modern Distance Education mode. In this new mode,the students'learning content, learning methods,learning time and learning location all has brought about profound changes. It has become an important research topic how to evaluate comprehensively and accurately the quality of the teacher training of the Teachers'Continuing Education. At present, the teacher training institutions are researching the evaluating system of Teachers'Continuing Education. These evaluating systems give inspectively evaluating indexes, collect evaluating data, and then obtain decision rules by data minging. However, due to the processing data from network, these evaluating systems are almost faced with the problems of large amount data or incomplete data.Rough set theory proposed by Z.Pawlak in 1982 is just able to solve those problems. It is a mathematical tool for dealing with uncertain or incomplete knowledge. It is based on the indiscernibility relation that describes indistinguishable objects, and concepts are represented by lower and upper approximations. Not needing other information this theory can analyze and process the non-accurate non-integrity and incomplete data.This paper, based on the traditional rough set theory, takes the Teacher Continuing Education evaluating system as the actual background to reseach knowledge reduction in incomplete information system. Firstly, this paper introduces the rough sets theory and analyzes the limitation of the extended rough set models in existence, and then provides an improved rough set modle based on tolerance relation according to the shortcomings of the tolerance relation model.The example analysis proves that the new model is more fit for reality and has stronger classification ability and more flexiblity than the tolerance relation model. Secondly, this paper introduces the the traditional rough entropy based on the tolerance relation model and correlative knowledge reduction algorithm, analyzes the limitation of the traditional rough entropy, defines a new rough entropy based on the improved tolerance relation model, and then provides a new attribute reduction algorithm based on the rough entropy of the inproved tolerance relation. The example proves its validity. Tis paper introduces also multi-value discernable matrix and gives a value reduction algorithm based on multi-value discernable matrix,and then apply this value reduction algorithm to obtain decision rules. Finally, the paper designs and elaborates the overall structure and the function of Teacher Continuing Education evaluation system, and designs the data acquisition module and the student overall assessment module in detail, and then apply the knowledge reduction algorithm in this evaluation system. In this way, we have established the data mining model of the Teacher Continuing Education evaluation system based on the incomplete rough set.