Research on Knowledge Acquisition and Knowledge Extraction of Forest Management

Abstract

The abundance of knowledge seriously influence the ability of expert decision support system deal with problems, it makes the problem that how to find the reliable knowledge on the Internet. A method that use web crawler crawl forest management web and extract expert knowledge in them is put forward in this paper, to solve the problem. This method limit the web crawler, and it is adopted to grasp the knowledge related to forest management and to extract the expert knowledge therein, enriching knowledge base of expert decision support system. Crawling web priority is calculated by reliability value, and Pearson correlation is used to make sure that the collected knowledge is related to forest management. The knowledge of forest management is extracted and repetitive knowledge is removed on the basis of collecting knowledge. Included angle cosine and Euclidean distance are used in this method to avoid acquit duplicate content. Experimental results show that this method has good accuracy.