Associative Rule Mining in Large Datasets using Neural Network Algorithm and enhanced Apriori - Based algorithm
 |
International Journal of Computer Trends and Technology (IJCTT) |  |
| © 2018 by IJCTT Journal | ||
| Volume-59 Number-1 |
||
| Year of Publication : 2018 | ||
| Authors : Febin Issac, Yeshwant More | ||
| DOI : 10.14445/22312803/IJCTT-V59P106 |
Febin Issac, Yeshwant More "Associative Rule Mining in Large Datasets using Neural Network Algorithm and enhanced Apriori - Based algorithm". International Journal of Computer Trends and Technology (IJCTT) V59(1):30-34, May 2018. ISSN:2231-2803. www.ijcttjournal.org. Published by Seventh Sense Research Group.
Abstract
Unsupervised Learning is a type of machine learning algorithm, which is used to draw conclusions out of unlabeled input data. Association Rule Mining(ARM) is one of the key unsupervised data mining method, which is used to find interesting associations in large data sets. In this paper, we would explain about data clustering by K-means followed by Association Rule Mining using Apriori algorithm in each cluster to obtain meaningful associations and then comparing it with the Self Organizing Map(SOM) clustering method. SOM makes use of neural networks for generating frequent item sets and association rules from transaction data. We would compare the accuracy and performance of the above clustering algorithms based on parameters governed by each. When this comparison is applied on various kinds of datasets, the usage of the right algorithm for each type of dataset can be determined that increases the correctness of the machine learning solution.
Reference
[1] Andrew Moore: “K-means and Hierarchical Clustering – Tutorial Slides” http://www-2.cs.cmu.edu/~awm/tutorials/kmeans.html
[2] Wikipedia https://en.wikipedia.org/wiki/K-means_clustering
[3] Apriori Algorithm – Classical algorithm for data mining https://www.techleer.com/articles/155-apriori-algorithmclassical- algorithm-for-data-mining/
[4] Data Mining Using Neural Networks https://researchbank.rmit.edu.au/eserv/rmit:9493/Rahman.pdf
[5] Self-Organizing Map (SOM) http://users.ics.aalto.fi/jhollmen/dippa/node9.html
[6] Visualization Association Rules: Introduction to the R-extension Package arulesViz https://cran.rproject. org/web/packages/arulesViz/vignettes/arulesViz.pdf
[7] Market Basket Analysis/Association Rule Mining using R package – arules https://prdeepakbabu.wordpress.com/2010/11/13/market-basketanalysisassociation- rule-mining-using-r-package-arules/
[8] Market basket analysis with neural gas networks and selforganising maps https://link.springer.com/content/pdf/10.1057%2Fpalgrave.jt.574 0092.pdf
Keywords
Apriori Algorithm, Association Rule Mining, K Means, Kohonen’s Self Organizing Map, Machine Learning, Neural Network Algorithms, Self Organizing Map