Back to Search
Start Over
The Effects of Ant Colony Optimization on Graph Anonymization.
- Source :
-
GSTF Journal on Computing . Aug2016, Vol. 5 Issue 1, p91-100. 10p. - Publication Year :
- 2016
-
Abstract
- The growing need to address privacy concerns when social network data is released for mining purposes has recently led to considerable interest in various techniques for graph anonymization. These techniques and definitions, although robust are sometimes difficult to achieve for large social net-works. In this paper, we look at applying ant colony opti-mization (ACO) to two known versions of social network anonymization, namely k-label sequence anonymity, known to be NPhard for k = 3. We also apply it to the more recent work of [23] and Label Bag Anonymization. Ants of the artificial colony are able to generate successively shorter tours by using information accumulated in the form of pheromone trails deposited by the edge colonies ant. Computer simu-lations have indicated that ACO are capable of generating good solutions for known harder graph problems. The contributions of this paper are two fold: we look to apply ACO to k-label sequence anonymity and k=label bag based anonymization, and attempt to show the power of ap-plying ACO techniques to social network privacy attempts. Furthermore, we look to build a new novel foundation of study, that although at its preliminary stages, can lead it ground breaking results down the road. [ABSTRACT FROM AUTHOR]
- Subjects :
- *GRAPH theory
*DATA mining
*ANT algorithms
Subjects
Details
- Language :
- English
- ISSN :
- 20102283
- Volume :
- 5
- Issue :
- 1
- Database :
- Academic Search Index
- Journal :
- GSTF Journal on Computing
- Publication Type :
- Academic Journal
- Accession number :
- 117624291
- Full Text :
- https://doi.org/10.5176/2251-3043_4.4.358