Building an Expert Finding System by Using Google Distance Approach

• Main Authors: Yu-Li Lin, 林侑立
• Other Authors: Kai-Hsiang Yang
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/rq95dp

碩士 === 國立臺北教育大學 === 數學暨資訊教育學系(含數學教育碩士班) === 103 === Expert finding have always gotten much attention at information retrieval for a long time. Many researches provide many better achievement than before, but it still have room for improvement. Query Expansion is a method to add the feedback results, and can promote the search precision effectively. In generally, this method had often involved the term frequency, and will generate the terms problem about noise in the process. Google platform have possessed a large number of web contents, each correlation between two terms is hidden in there. We can assess the semantic distance by analyzing the whole web content. Therefore, this research provides a NGD model to solve the problem about expert finding by sorting the expansion terms by Google. The experiment result shows that it do not really enhance the precision by using Google information, but when the expansion terms selected more, the precision will increase. On the other hands, the model also can solve the terms problem about noise.