An Efficient Ranking Based Clustering Algorithm
Dipak R. Pardhi1, Akhilesh A. Waoo2
1Dipak R.Pardhi, M.Tech Student, C.S.E .Dept. Bansal Institute of Science & Technology, Bhopal, India.
2Akhilesh A.Waoo, Asst. Prof. M.Tech.(CSE), Dept. Bansal Institute of Science & Technology, Bhopal, India.
Manuscript received on October 06, 2011. | Revised Manuscript received on October 12, 2011. | Manuscript published on October 30, 2011 . | PP: 35-40 | Volume-1 Issue-1, October 2011 . | Retrieval Number: A0105101111/2011©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: There are several databases, which contain large information about research publications in various fields, for examples, DBLP in computer science and PubMed in medical science. US Census data set which contains information with 68 categorical attributes, which is very complex to get the information. Zoo data set which having information with 17 attributes, Plant Cell Signalling data set which describes the interactions of the nodes within the plant signalling network by considering 43 different attributes. Each such database forms an immense size of information network connecting in very complex ways. In this work, we are proposing an approach for “information network mining” on such a database. We consider DBLP as an example. The database contains information about research papers, authors, conferences and journals. It also includes the date, year and the place of publication of particular journals and conferences. Various users have very specific personalized search criteria for profiling such patterns and verifying the interest, we are proposed an algorithm RBC_A, so that- (1) In-depth information about research, such as the clustering of conferences due to their sharing of many common authors can be categorized; (2) The reputation of a conference can be evaluated; finally (3) Time relevant information can be inferred. The above have been addressed in the design and development of this work.
Keywords: Information Network, Data Mining, Profiling, Ranking, Clustering, Classifications, Associations, User interface.