Finding Related Articles for a Content Stream Using Iterative Merge-Split Clusters
作者： Kostas Tsioutsiouliklis , Sainath Vellal
DOI:
关键词:
摘要: Software generates an article signature for each in a plurality of articles. The software initializes clustering algorithm with initial clusters that are non- overlapping. A centroid is generated cluster from the signatures articles cluster. performs succession alternating merges and splits using to create non-overlapping coherent clusters. identifies related specific by mapping at least one comparing cluster, similarity measure. displays proximity other content stream.
lens.org 参考文章(6)
Jiyong Jang, Scaling Software Security Analysis to Millions of Malicious Programs and Billions of Lines of Code Carnegie Mellon University. ,(2013) , 10.1184/R1/6721400.V1
Edwin Peter Dawson Pednault, Ramesh Natarajan, Method for constructing segmentation-based predictive models ,(2002)
Joshua M. Koran, User similarity groups for on-line marketing ,(2014)
Madhusudan Chinthakunta, Larry Paul Heck, Daniel Dole, Lisa Stifelman, Julian James Odell, Facilitating conversations with automated location mapping ,(2015)
Srinivas Vadrevu, Alex J. Smola, Byron E. Dom, Choon Hui Teo, Suju Rajan, Kunal Punera, Utilizing offline clusters for realtime clustering of search results ,(2011)
Sainath Vellal, Naveen Belkale, System and method for contextual mail recommendations ,(2013)