A Study Of Comparing Conceptual And Performance of K-Means and Fuzzy C-Means Algorithms (Clustering Method of Data Mining) of Consumer Segmentation

Yunita Yunita (STMIK LIKMI)
Sukrina Herman (STMIK LIKMI)
Ahsani Takwim (STMIK LIKMI)
Septian Rheno Widianto (STMIK LIKMI)

Jurnal Riset Informatika
Vol. 2 No. 2 (2020): March 2020 Edition

Consumers, especially potential customers, are an important asset in a company that should be maintained properly. The tight competition requires companies to focus on the customer's needs. Consumer segmentation is one of the processes carried out in the marketing strategy. Consumer or consumer segmentation data mining plays a very important role in supporting the grouping process results. Based on mapping studies on data mining in support of consumer segmentation, two algorithms are often used: K-means clustering and Fuzzy C-means clustering. The attributes used for mining in customer segmentation processes are customer data, products, demographics, consumer behaviour, transactions, RFMDC, RFM (Recency, Frequency Monetary) and LTV (Life Time Value). It is important to combine the clustering algorithm to algorithm Classification, Association, and CPV to get the potential value of each cluster.