Research On Uncertainty Measurement And Knowledge Acquiring In Information System

The uncertainty measurement and knowledge acquiring in information system is an issue in the study of information science field, and many researchers home and overseas have pay more attention to it. Furthermore, scientific achievements of it have been widely applied in data mining, decision analyses, pattern recognition, etc.In the paper, under the frame of Rough Set Theory, the uncertainty measurement and knowledge acquiring in information system is further studied based on the view of knowledge induced by partition. And the following results are obtained.In the relation measurement between knowledge in information system, a new measurement—knowledge inclusion is proposed, and it can be used to measure the certainty of rules gained by attribute set. Further, it has been proved that there is a strict complement relationship between knowledge inclusion degree and the conditional information entropy. Moreover, a reduc algorithm for a decision table based on knowledge inclusion was designed, and it is proved effectively. The results given in this paper extent the inclusion theory, make for analyzing the relations between knowledge successfully, and provide a new measurement for acquiring knowledge in information system.In the research of topology structure of knowledge in information system, the concept of knowledge distance was given and the properties of knowledge distance are studied. Following, we have proved that the all knowledge in information system constructs a metric space. The knowledge distance can be used to measure rough entropy of knowledge. Further more, the properties of new rough entropy, which are same to the classic ones, are proved. These results will be very helpful for analyzing the relationships among knowledge and the roughness of knowledge thoroughly in information system.In knowledge acquiring in incomplete system, the advantages of the maximal consistent block technique for acquiring knowledge are analyzed, and the properties of maximal consistent blocks, the number of maximal consistent blocks of objects relates to the completeness of object and the inclusion relations between maximal consistent are relate to the inclusion relation between attribute set, are given. In the chapter 5, a hierarchical algorithm for constructing the maximal consistent blocks in incomplete system is proposed, which will be helpful for acquiring knowledge efficiently in incomplete information system.We have done some study on the uncertainty measurement and knowledge acquiring in information system, but the research about uncertainty measurement and knowledge acquiring is in a booming stage and there are many problems worth studying on it. Our work is just a beginning, and related work needs to be further developed.