Comparative Performance Of Using PCA With K-Means And Fuzzy C Means Clustering For Customer Segmentation
Journal: International Journal of Scientific & Technology Research (Vol.4, No. 10)Publication Date: 2015-10-15
Authors : Fahmida Afrin; Al-Amin; Mehnaz Tabassum;
Page : 70-74
Keywords : Index Terms Data Mining; Clustering; K-means; Principal component analysis; Fuzzy C means; Customer segmentation; Crisp Set;
Abstract
Abstract Data mining is the process of analyzing data and discovering useful information. Sometimes it is called knowledge Discovery. Clustering refers to groups whereas data are grouped in such a way that the data in one cluster are similar data in different clusters are dissimilar. Many data mining technologies are developed for customer segmentation. PCA is working as a preprocessor of Fuzzy C means and K- means for reducing the high dimensional and noisy data. There are many clustering method apply on customer segmentation. In this paper the performance of Fuzzy C means and K-means after implementing Principal Component Analysis is analyzed. We analyze the performance on a standard dataset for these algorithms. The results indicate that PCA based fuzzy clustering produces better results than PCA based K-means and is a more stable method for customer segmentation.
Other Latest Articles
- Internal Electric Field In The Space Charge Layer Of A Solar Cell Based On Silicon In The Presence Of Excitons
- Query Optimization Techniques - Tips For Writing Efficient And Faster SQL Queries
- Impact Of Different Time Planting In Soybeans And Neem Seed Extract Application To Insect Population On Rice Field
- The Influence Of Policy Implementation From The Change Of Institutional Status Toward Quality Of Patient Service In Hospital
- Crimes In New York And Seoul A Comparison
Last modified: 2015-11-13 18:33:27