Scientific Quality Towards Specific Topics

The studies of citations are comprehensively carried out with the increasing electronically citation data on the Web. Most of the metrics observe scientific quality in a global view instead of in multiple fine-grained views. In this paper, we suggest to apply Topic Model and adaptive PageRank algorithm to assess the relative importance of scientific objects including articles, authors, conferences and journals. The scientific quality is measured by an aggregation PageRank metric towards some topics. This metric considers the impact of a paper both in global view and local view. The experiments on ACL Anthology bibliographic corpus show our method is a useful measure to observe scientific quality on multi-views. KeywordsPageRank; Topic Model; Citation Analysis

[1]  Dragomir R. Radev,et al.  The ACL Anthology Reference Corpus: A Reference Dataset for Bibliographic Research in Computational Linguistics , 2008, LREC.

[2]  Taher H. Haveliwala.  Topic-Sensitive PageRank: A Context-Sensitive Ranking Algorithm for Web Search , 2003, IEEE Trans. Knowl. Data Eng..

[3]  Sergey Brin,et al.  The Anatomy of a Large-Scale Hypertextual Web Search Engine , 1998, Comput. Networks.

[4]  Sergei Maslov,et al.  Finding scientific gems with Google's PageRank algorithm , 2006, J. Informetrics.

[5]  Carl T. Bergstrom,et al.  The Eigenfactor™ Metrics , 2008, The Journal of Neuroscience.

[6]  Mark Steyvers,et al.  Finding scientific topics , 2004, Proceedings of the National Academy of Sciences of the United States of America.

[7]  Michael H. MacRoberts,et al.  Problems of citation analysis , 1992, Scientometrics.

[8]  Johan Bollen,et al.  Journal status , 2006, Scientometrics.