EFFICIENT LINK PREDICTION METHOD IN DARK NETWORKS ANALYSIS

A.W. MAHESAR, Z. BHATTI, A. WAQAS, M. Y. KOONDHAR, M. M. RIND, S. NIZAMANI

Abstract


The prediction of future links is very important problem in the field of complex network analysis. Due to the intricate interplay between nodes in dark networks the links prediction problem become very hard. As the dark networks evolve slowly with few nodes and gradually formed as well organized networks due to scale-free topology and therefore need efficient method of links prediction. In this paper, we apply two useful metrics of links prediction namely Jaccard’s coefficient and Adamic/Aadr on the 9/11 dark network dataset. The results of both these metrics has been compared and it has been observed that Adamic/Aadr metric is more efficient method of link prediction in case of dark networks as compared to Jaccard’s coefficient. The performance of both these methods is evaluated by using standard evaluation metrics which is precision.

Full Text:

PDF

Refbacks

  • There are currently no refbacks.


Copyright (c) 2016 Sindh University Research Journal - SURJ (Science Series)

 Copyright © University of Sindh, Jamshoro. 2017 All Rights Reserved.
Printing and Publication by: Sindh University Press.