A Novel Ranking Model for Papers, Keywords, and Author on Academic Community

• Main Authors: Tsai, Minhan, 蔡旻翰
• Other Authors: Yeh, Jiehshan
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/11699976599836293419

碩士 === 靜宜大學 === 資訊管理學系 === 100 === One of the research topics in academic community network analysis is to ranking authors, articles, or journals. Previous researches considered the importance of authors, articles, or journals mainly based on the statistical calculation. However, to the best of our knowledge, there is no study on the interaction among the importance of articles, keywords and authors. This study adopts the concept of the web data mining and combines HITS and PageRank algorithms to analyze the importance among articles, keywords, and authors on academic community network. This study first proposes a novel ranking model, PKARank, which defines the relations of the weights of articles, keywords, and authors on academic community. The proposed PKARank algorithm combines HITS and PageRank algorithms, and iteratively computes the weights. The experimental results on the simulating academic community network generated from SciVerse Scopus database also demonstrate that the proposed model provides the meaningful rankings among articles, keywords, and authors.