The Disposal of Incomplete Classification Data in Teaching Evaluation System

In view of the complexity, ambiguity and multi-factor of teaching evaluation, it is an important topic that classification is applied to the domain. In the field of classification, incomplete data is one of thorny issues. There have been many methods to deal with the problem in many studies. In the paper, the MVCM is introduced particularly. It fills these observations with incomplete data by constructing fuzzy members. The member of fuzzy set with maximum probability is exactly the replacement. Experiments are conducted on students´ achievement data set to demonstrate the solution.