Academic Journals Database
Disseminating quality controlled scientific knowledge

Reduction of Negative and Positive Association Rule Mining and Maintain Superiority of Rule Using Modified Genetic Algorithm

ADD TO MY LIST
 
Author(s): Nikhil Jain,Vishal Sharma,Mahesh Malviya

Journal: International Journal of Advanced Computer Research
ISSN 2249-7277

Volume: 2;
Issue: 6;
Start page: 31;
Date: 2012;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: Association Rule Mining | Negative and Positive rules | Superiority | Genetic algorithm.

ABSTRACT
Association rule mining play important rule inmarket data analysis and also in medical diagnosisof correlated problem. For the generation ofassociation rule mining various technique are usedsuch as Apriori algorithm, FP-growth and treebased algorithm. Some algorithms are wonderperformance but generate negative association ruleand also suffered from Superiority measureproblem. In this paper we proposed a multi-objectiveassociation rule mining based on genetic algorithmand Euclidean distance formula. In this method wefind the near distance of rule set using Euclideandistance formula and generate two class higherclass and lower class .the validate of class check bydistance weight vector. Basically distance weightvector maintain a threshold value of rule itemsets.In whole process we used genetic algorithm foroptimization of rule set. Here we set population sizeis 1000 and selection process validate by distanceweight vector. Our proposed algorithm distanceweight optimization of association rule mining withgenetic algorithm compared with multi-objectiveassociation rule optimization using geneticalgorithm. Our proposed algorithm is better rule setgeneration instead of MORA method.
Affiliate Program      Why do you need a reservation system?