Knowledge View Based on Rough Set and Similarity

View maintenance is a hot topic in database. But unlike database view, the knowledge view of knowledge management system has its particularity. The knowledge view is a cache of knowledge item that can be access quickly. Based on rough set, this paper proposes an approach of constructing knowledge view. The approach divides the knowledge set into three region using rough set theory, then reduces the three region by calculating interest similarity of every knowledge item, and finally the knowledge view will be constructed by filtering the knowledge item of merged region that consists of the three reduced region. The experiment results show that the approach can construct knowledge view efficiently. Introduction With the development of information technology and the growing popularity of Internet, much more knowledge-intensive corporations have raised concern about Knowledge Management (KM) to increase their competitive ability, since KM is regarded as the formal management of knowledge for facilitating creation, accessing, and reuse of knowledge, typically using advanced technology. Nowadays, all KM System (KMS) have a large amount of knowledge and are updated continually. This makes user have adequate knowledge to use, but in the same time causes trouble to user. In general, each user has constant interest in a period of time. To solve a problem, the user retrieves many knowledge items from KMS and navigates some of them which he thinks is available to him. When he enters KMS again for the same problem, to avoid the troublesome retrieval, he wishes that KMS will automatically provide the same knowledge to him. On the other hand, if the user viewed a knowledge item that is transferred from remote database, when he want to view the same knowledge item again, he wish to view it directly instead of time-consuming transferring from remote database again. These demands can be satisfied with the concept of view. Traditionally, a view is defined as a function from a set of base tables to a derived table and the function is recomputed every time the view is referenced. On the other hand, a view is like a cache of data that can be accessed quickly. So views are useful in applications such as data warehousing, replication servers, data recording system, data visualization and mobile systems [1-3]. But, because the object of KMS is different from these applications, the approach of view management in database can not been used in KMS directly. So, developing an approach of view management in KMS is necessary. [4] proposes an approach of creating user view in KMS based on user navigation map. It used navigation map and no-loop map to generate judge rule which is used in creating the user view. This approach can create user view that is combined with user interest, but it is limited to the knowledge structure defined in [4]. Other researchers [5,6,7] propose many approaches of view maintenance, but these approaches are not suitable for KMS. So, based on rough set and similarity, this paper proposes a knowledge view constructing approach. This approach can construct knowledge view without the limitation of particular knowledge structure. 6th International Conference on Machinery, Materials, Environment, Biotechnology and Computer (MMEBC 2016) © 2016. The authors Published by Atlantis Press 1983 Organization of the Text In order to comprehensive the approach proposed in this paper, relative conceptions are introduced as follows. Rough Set Theory. Rough set theory[8,9] combines the classification and knowledge together, and it can portray approximately the uncertain or imprecise knowledge based on the known knowledge in database. Here are a few related basic concepts. (1) Suppose S= (U, A, V, f) is a information system, where U is a non-empty finite set called universe including all elements, A is a non-empty finite set with attribute for each element, V is the range of A (V=