EFFICIENT COMPUTATION OF SIMRANK FOR STATIC AND DYNAMIC DATASETS USING MAPREDUCE FRAMEWORK

Soujanya Duvvi, Venkata Ramana Kondapalli

Abstract


The growth of data dynamically over the internet and the need to store, access information efficiently brings up new challenges of finding related documents, similar nodes, domain & inter-domain similarities etc. Though SimRank is applicable to wide range of areas, we use this similarity ranking to find similarity between neighbours in a contextual way and evaluate in a numerical way. Here we use Jaccard Similarly for calculating similarity by using LSH and various other methods. We further optimize the Jaccard Algorithm by using Token Optimization join method. The obtained result is further evaluated with a combination of four other parameters and from the result obtained the similarity values of nodes that are greater than the optimal threshold value φ are retrieved from the huge graph.

Full Text:

PDF2

Refbacks

  • There are currently no refbacks.


© 2017 International Journal of Global Research in Computer Science (JGRCS)
Copyright Agreement & Authorship Responsibility