CUSTOMER CLUSTER GROUPING IN STARBUCKS APPLICATION USING K-MEANS ALGORITHM AND PRINCIPAL COMPONENT ANALYSIS

Hidayatulloh, Alfian (2022) CUSTOMER CLUSTER GROUPING IN STARBUCKS APPLICATION USING K-MEANS ALGORITHM AND PRINCIPAL COMPONENT ANALYSIS. ["eprint_fieldopt_thesis_type_tugasakhir" not defined] thesis, University of Technology Yogyakarta.

[img] Text
ALFIAN HIDAYATULLOH_5180411382_ABSTRAK.pdf

Download (15kB)

Abstract

Starbucks is an American coffee shop chain headquartered in Seattle, Washington. Since the millennial era, competition in the coffee business is no longer based on brand names but product quality and price. Product quality and prices that match market demand automatically raise the brand name in the eyes of consumers. The purpose of this research is to analyze and practice customer segmentation. By implementing customer segmentation, Starbucks can better understand the needs and wants of its customers. It also allows Starbucks to offer different offers and promotions based on customer preferences. To find out the characteristics of each customer, the authors designed the FMT (frequency, monetary, tenure) model. Principal Component Analysis (PCA) method is used to reduce the number of variables (dimension reduction) without eliminating critical information in the dataset. Based on the PCA results, the authors found that using 3 out of 5 variables already represented 89% of the total information in the FMT model. Based on the results of the clustering process using the K-Means algorithm in the Google Collaboratory, the model can classify customers into four segments: bronze customers, silver customers, gold customers, and diamond customers. In addition, the Elbow method test results show Within-Cluster Sum of Squares (WCSS), a measure of the proximity between objects in a cluster, of 979.80. A smaller WCSS value indicates the best number of clusters. However, this is done by considering the magnitude of the WCSS value decrease in each number of clusters. Keywords: Customer Grouping, FMT Model, PCA, K-Means Algorithm, Elbow Method

Item Type: Thesis (["eprint_fieldopt_thesis_type_tugasakhir" not defined])
Subjects: T Technology > T Technology (General)
Divisions: Fakultas Sains Dan Teknologi > S1 Informatika
Depositing User: Kaprodi S1 Informatika UTY
Date Deposited: 24 Nov 2022 04:30
Last Modified: 24 Nov 2022 04:30
URI: http://eprints.uty.ac.id/id/eprint/11188

Actions (login required)

View Item View Item