R sk nn- knn search on road networks by incorporating social influence

DATA ALCOTT SYSTEMS www.finalsemprojects.com 09600095046
www.ns2projects.com dataalcott@gmail.com
RSKNN: KNN SEARCH ON ROAD NETWORKS BY INCORPORATING
SOCIAL INFLUENCE

ABSTRACT
Although kNN search on a road network Gr, i.e., finding k nearest objects to a
query user q on Gr, has been extensively studied, existing works neglected the fact
that the q’s social information can play an important role in this kNN query. Many
real-world applications, such as location-based social networking services, require
such a query. A new problem is proposed: kNN search on road networks by
incorporating social influence (RSkNN). Specifically, the state-of-the-art
Independent Cascade (IC) model in social network is applied to define social
influence. One critical challenge of the problem is to speed up the computation of
the social influence over large road and social networks. To address this challenge,
three efficient index-based search algorithms is proposed, i.e., road network-based
(RN-based), social network-based (SN-based) and hybrid indexing algorithms. In
the RN-based algorithm, a filtering-and-verification framework is employed for
tackling the hard problem of computing social influence. In the SN-based
algorithm, embed social cuts into the index, so that speed up the query. In the
hybrid algorithm, an index is proposed, summarizing the road and social networks,
based on which we can obtain query answers efficiently. Finally, real road and
social network data is used to empirically verify the efficiency and efficacy of
solutions.

INTRODUCTION
With the ever-growing popularity of mobile devices (e.g., smartphones), location-
based service (LBS) systems have been widely deployed and accepted by mobile
users. The k-nearest neighbor (kNN) search on road networks is a fundamental
problem in LBS. Given a query location and a set of static objects (e.g., restaurant)
on the road network, the kNN search problem finds k nearest objects to the query
location. Alone with the popular usage of LBS, the past few years have witnessed a
massive boom in location-based social networking services like Foursquare, Yelp,
Loopt, Geomium and Facebook Places. In all these services, social network users
are often associated with some locations. Such location information, bridging the
gap between the physical world and the virtual world of social networks, presents
new opportunities for the kNN search on road networks.

EXISTING SYSTEM
Geo-Social Query Processing
 Yang et al. study the Socio-Spatial Group Query (SSGQ): given a query
location q, the SSGQ returns a group of users, such that (a) each user in the
group is socially connected with at least a number of other members, and (b)
the sum of distances of all members in the group to q is minimized. The
SSGQ queries are NP-hard, and thus the authors present approximation
algorithms.
 Liu et al. propose a circle-of-friend query to find minimal-diameter social
groups, in which a combined distance based on spatial and social distances is
proposed.
 Existing system formulated a framework for geo-social query processing
that builds queries based upon atomic operations. Some complex queries can
be answered by combining these atomic operations.
 Li et al study spatial-aware interest group queries in location-based social
networks and give efficient processing algorithms.
 Sun et al. propose k-nearest neighbor temporal aggregate queries that have
emerging applications in location-based social networks. They process
queries based on a tree index, TAR-tree, that adds temporal information to
R-tree.
Influence Maximization in Social Networks

 Kempe et al. propose a widely accepted discrete influence propagation
model, Independent Cascade (IC) model. The work prove the influence
maximization problem under IC model is NP-hard and give a (1 − 1/e)-
approximation algorithm. To avoid the NP-hard problem, authors use
shortest paths to estimate the IC model. However, this estimation leads to a
large error from the true value, if shortest paths are used to estimate social
influence.

PROPOSED SYSTEM
 Proposed a road network-based (RN-based) indexing algorithm.
 In the RN-based indexing algorithm, a balanced tree index IRN is used,
based on which a best-first search can be conducted to obtain nearest objects
to q.
 For each returned object or, give a filtering-and-verification framework to
compute the social influence SI(or) efficiently.
 During the filtering, tight lower and upper bounds of SI(or) is developed so
as to validate or prune.
 For a candidate or after the filtering, a sampling algorithm is designed to
verify or.
Advantages
 Proposed social network based (SN-based) and hybrid indexing algorithms
 RSkNN query is over road network
 RSkNN query finds a group of objects instead of individual users and
 RSkNN query considers social influence instead of social distance.

HARDWARE REQUIREMENTS
Processor : Any Processor above 500 MHz.
Ram : 128Mb.
Hard Disk : 10 Gb.
Compact Disk : 650 Mb.
Input device : Standard Keyboard and Mouse.
Output device : VGA and High Resolution Monitor.
SOFTWARE SPECIFICATION
Operating System : Windows Family.
Techniques : JDK 1.5 or higher
Database : MySQL 5.0

R sk nn- knn search on road networks by incorporating social influence

Recommended

Recommended

More Related Content

Similar to R sk nn- knn search on road networks by incorporating social influence

Similar to R sk nn- knn search on road networks by incorporating social influence (20)

Recently uploaded

Recently uploaded (20)

R sk nn- knn search on road networks by incorporating social influence