文章基本信息

标题：Comparative Analysis of Clustering & Enhancing Classification Using Bio- Inspired Approaches
本地全文：下载
作者：Navpreet Rupal ; Poonam Kataria
期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2014
卷号：5
期号：5
页码：6453-6457
出版社：TechScience Publications
摘要：Data Mining is a technique of discovering hidden patterns & relationship in data with the help of various tools & techniques to make a valid prediction . Clustering is defined as process of partitioning a set of objects or data into a set of meaningful sub-classes called as clusters. It helps the users to understand the structure in a data set. Classification groups the data under different classes. Bio- inspired approaches are various evolutionary algorithms inspired from nature and solves hard and complex computing problems. In this work , we first form the clusters of the dataset of a bank with the help of h-means clustering. This work is also based on comparative study of GA, PSO & BFO based Data clustering methods. To compare the results we use different performance parameters for classification such as precision, cohesion, recall and variance. The results prove that BFO yields better outputs as compared to GA and PSO. So this work shows that BFO results as a better optimization technique
关键词：Data Mining; Clustering; Bio-inspired approaches;GA; BFO; PSO