Research On The Evolution Laws Of Knowledge Category Structure In Network Information World

With the development of Web2.0, some new ideas and technologies are occurring which could meet the growing demands for personalized and social needs, making the users in network who were knowledge "recipients" become knowledge creators, and thus changing the ways to share and creation of knowledge. In network information world, users ignoring the limitation of time and space interact, explore, communicate, share, exchange and cooperate with others with different knowledge background and cultural environment. This way containing mass collaboration and contents created by users gathers the users’wisdom from bottom to top and emerges some new knowledge structures based on kinds of platform such as knowledge category in Wikipedia, network tagging, folksonomy and so on. Different from the traditional knowledge category, knowledge category in Wikipedia is one kind of knowledge category which is spontaneous, self-organized, and collaboratively edited by many users. This makes knowledge category in Wikipedia become one of hot research field.Currently the research on knowledge category in Wikipedia can be divided into three aspects:research on category itself, research on category optimization and research on evolution of category. The existing studies generally preprocessed category as tree, then analyzed it in a traditional way. These ways destroyed the relations between categories which cannot reflect the real evolution of knowledge category in Wikipedia; other researches, retaining the integrity of collaborative editing, primarily focused on macro level of category and the indexes they used are relatively simple. The research on the micro level of category is relatively rare, even worse on identification of border of category in Wikipedia.Category structure in Wikipedia is one kind of knowledge organization that is typically collaboratively edited by users. Thus, the research on category can help us know about the shortage of knowledge category systems in network information world and clarify the status of domain knowledge and hot areas of it, then promote the innovation of knowledge. So in this paper, we take category in Wikipedia as an example, based on the analysis and definition of Wikipedia category, and propose a new measurement model by using self-avoiding random walk and similarity. Based on the proposed model, we can get the diversity entropy of node in category structure. According to the diversity entropy we got, we can identify the border and find the evolution laws of knowledge category structure in Wikipedia.In this paper, our study deepen the theoretical knowledge on Wikipedia knowledge building process and the proposed model enriches the theory of complex network dynamics and provides a new view to study the relations between connection and dynamics. What’s more, research on the border identification of category in Wikipedia can help us know about the status of domain knowledge and then promote the innovation of knowledge.