Graphs are widely used in large scale social network analysis. Graph mining increasingly important in modelling complicated structures such as circuits, images, web, biological networks and social networks. The major problems occur in this graph mining are computational efficiency (CE) and frequent sub graph mining (FSM). Computational Efficiency describes the extent to which the time, effort or efficiency which use computing technology in information processing. Frequent Subgraph Mining is the mechanism of candidate generation without duplicates. FSM faces the problem on counting the instances of the patterns in the dataset and counting of instances for graphs. The main objective of this project is to address CE and FSM problems. The paper cited in the reference proposes an algorithm called Mirage algorithm to solve queries using sub graph mining. The proposed work focuses on enhancing An Iterative MapReduce based Frequent Subgraph Mining Algorithm (MIRAGE) to consider optimum computational efficiency. The test data to be considered for this mining algorithm can be from any domains such as medical, text and social data’s (twitter).The major contributions are: an iterative MapReduce based frequent subgraph mining algorithm called MIRAGE used to address the frequent subgraph mining problem. Computational Efficiency will be increased through MIRAGE algorithm over Matrix Vector Multiplication. Performance of the MIRAGE will be demonstrated through different synthetic as well as real world datasets. The main aim is to improvise the existing algorithm to enhance Computational Efficiency.

Ms. Babbitha.M,Dr. Angelina Geetha,Mr. Mohammed Jaffer.A.R."Social Mining to Improve the Computational Efficiency Using MapReduce". International Journal of Computer Engineering In Research Trends (IJCERT) ,ISSN:2349-7084 ,Vol.2, Issue 05,pp.288-292, May - 2015, URL :https://ijcert.org/ems/ijcert_papers/V2I53.pdf,

[1] Mansurul A Bhuiyan and Mohammad Al
Hasan, ―MIRAGE: An Iterative MapReduce based
Frequent Subgraph Mining Algorithm‖, ACM
Computing Research Repository, arXiv: 1307.5894,
Volume 1, 2013.
[2] Yi-Chen Lo, Hung-CheLai, Cheng-Te Li and
Shou-De Lin,‖ Mining and Generating Large Scaled
Social Networks via MapReduce‖, Springer-Verlag
Advances in Social Networks Analysis and Mining, pp -
1449–1469, 2013.
[3] SabaSehrish, Grant Mackey, Pengju Shang, Jun
Wang and John Bent,‖Supporting HPC Analytics
Applications with Access Patterns Using Data
Restructuringand Data-Centric Scheduling
TechniquesinMapReduce‖ IEEE Transactions on Parallel
and Distributed Systems, Volume 24, 2013.