PROJECT TITLE :

Efficient and Exact Local Search for Random Walk Based Top-K Proximity Query in Large Graphs

ABSTRACT:

High- proximity question in massive graphs may be a fundamental problem with a wide selection of applications. Varied random walk primarily based measures have been proposed to measure the proximity between totally different nodes. Though these measures are effective, efficiently computing them on giant graphs may be a challenging task. During this paper, we tend to develop an economical and exact native search method, FLoS (Fast Native Search), for high- proximity question in large graphs. FLoS guarantees the exactness of the solution. Moreover, it can be applied to a selection of commonly used proximity measures. FLoS relies on the no native optimum property of proximity measures. We have a tendency to show that several measures don't have any native optimum. Utilizing this property, we tend to introduce several operations to control transition chances and develop tight lower and higher bounds on the proximity values. The lower and upper bounds monotonically converge to the exact proximity value when more nodes are visited. We have a tendency to more extend FLoS to measures having local optimum by utilizing relationship among different measures. We perform comprehensive experiments on real and artificial large graphs to evaluate the potency and effectiveness of the proposed methodology.