Please use this identifier to cite or link to this item: http://hdl.handle.net/2080/2728
Full metadata record
DC FieldValueLanguage
dc.contributor.authorBehera, Ranjan Kumar-
dc.contributor.authorSukla, Abhishek Sai-
dc.contributor.authorMahapatra, Sambit-
dc.contributor.authorRath, Santanu Kumar-
dc.contributor.authorSahoo, Bibhudatta-
dc.contributor.authorBhattacharya, Swapan-
dc.date.accessioned2017-07-11T10:08:58Z-
dc.date.available2017-07-11T10:08:58Z-
dc.date.issued2017-07-
dc.identifier.citation29th International Conference on Software Engineering and Knowledge Engineering (SEKE 2017), Wyndham Pittsburgh University Center, Pittsburgh, USA, 5 – 7 July 2017en_US
dc.identifier.other10.18293/SEKE2017-100-
dc.identifier.urihttp://hdl.handle.net/2080/2728-
dc.descriptionCopyright for the paper belongs to proceedings publisher.en_US
dc.description.abstractLink prediction is an important research direction in the field of Social Network Analysis. The significance of this research area is crucial especially in the fields of network evolution analysis and recommender system in online social networks as well as e-commerce sites. This paper aims at predicting the hidden links that are likely to occur in near future. The possibility of formation of links is based on the similarity score between pair of nodes that are not yet connected in the social network. The similarity score, which we call link prediction score has been evaluated in Map-Reduce programming model. The proposed similarity score is based on both the structural information around the nodes and the degree of influence for neighboring nodes. The proposed algorithm is scalable in nature and performs quite well for large scale complex networks having good number of nodes and edges based on large pool of data or often termed as big-data. The efficiency and effectiveness of the algorithms are extensively tested and compared against traditional link prediction algorithms using three real world social network datasets.en_US
dc.language.isoen_USen_US
dc.publisherKSI Research Incorporationen_US
dc.subjectLink Predictionen_US
dc.subjectPreferential Attachmenten_US
dc.subjectSim Ranken_US
dc.subjectJaccard Coefficienten_US
dc.subjectKartz Measureen_US
dc.titleMap-Reduce based Link Prediction for Large Scale Social Networken_US
dc.typeArticleen_US
Appears in Collections:Conference Papers

Files in This Item:
File Description SizeFormat 
2017_SEKE17_RKBehera.pdf222 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.