Building a Scalable Eservice Recommender System

  IJCTT-book-cover
 
International Journal of Computer Trends and Technology (IJCTT)          
 
© 2017 by IJCTT Journal
Volume-46 Number-1
Year of Publication : 2017
Authors : S.J. Savitha, D. Betteena Sheryl Fernando, K. Saranya
DOI :  10.14445/22312803/IJCTT-V46P102

MLA

S.J. Savitha, D. Betteena Sheryl Fernando, K. Saranya "Building a Scalable Eservice Recommender System". International Journal of Computer Trends and Technology (IJCTT) V46(1):5-9, April 2017. ISSN:2231-2803. www.ijcttjournal.org. Published by Seventh Sense Research Group.

Abstract -
E-commerce Recommender system is proposed to solve Big Data problem due to huge amount of data, prevailing in many of the service recommender systems in the market. And to build scalable, efficient and precise service comparison and recommender system is highly needed. This system enables the shoppers to deeply analyses on what product to choose in various services. This system recommends the user to purchase the product and grab the data from various web services, loads to hadoop file system and clustered and classified the product using mapreduce framework. This recommender system will recommend the product based on the Case Based Collaborative Filtering (CBCF). CBCF is to filter the product information from huge amount of data for product comparison. Model based method is used to predict the item. Pearson Correlation Coefficient is used to measure the similarity value of the items. This proposed system avoids the scalability problem of existing recommender system. It reduces the overall time required by the user to analyses the services on the e-commerce environment and the users can effectively retrieve and identify the suitable product from the e-commerce system.

References
[1] S. Papadimitriou and J. Sun, “Disco: Distributed Co-Clustering with Map-Reduce”, 8th Conference in Data Mining, pp. 512-521, June 2008.
[2] D. Luo, C. Ding, and H. Huang, “Parallelization with Multiplicative Algorithms for Big Data Mining”,12th International conference on Data Mining, pp. 489-498, Nov 2012.
[3] C. Ranger, R. Raghu Raman, A. Penmetsa, G. Bradski, and C. Kozyrakis, “Evaluating Mapreduce for Multi-Core and Multiprocessor Systems”, 13th Conference on Computer Architecture, pp. 13-24, June 2007.
[4] C. Ranger, R. Raghu Raman, A. Penmetsa, G. Bradski, and C. Kozyrakis, “Evaluating Mapreduce for Multi-Core and Multiprocessor Systems”, 13th Conference on Computer Architecture, pp. 13-24, June 2007.
[5] S. Aral and D. Walker, “Identifying Influential and Susceptible Members of Social Networks”, IEEE Transaction on e-commerce recommender systems, vol.no.1, pp.337-341, Dec.2012.
[6] R. Ahmed and G. Karypis, “Algorithms for Mining the Evolution of Conserved Relational States in Dynamic Networks”, 15th conference on Knowledge and Information Systems, pp. 603-630, Aug 2013.
[7] P. Resnick and H. R. Varian, “Recommender systems,” IEEE Transaction on Communications of the ACM, vol. no. 3, pp. 56–58, June 2013.
[8] M. J. Pazzani and D. Billsus, “Content-based recommendation Systems,” IEEE Transactions in The adaptive web, vol. no.5, pp. 325–341, Dec 2012.
[9] T. Zhang and V. S. Iyengar, “Recommender systems are using linear classifiers”, The Journal of Machine Learning Research, vol. 2, pp. 313–334, May 2002.
[10] X. Wu and S. Zhang, “Synthesizing High-Frequency Rules from Different Data Sources”, IEEE Transaction on Knowledge and Data Engineering vol. No. 2, pp. 353-367, Mar/Apr 2003.

Keywords
Hadoop,Mapreduce,Fuzzy-KmeansClustering ,Naïv Base Classification, Recommendation System, Case based Collaborative Filtering, Similarity Measure.