ANALISIS SEGMENTASI KONSUMEN MENGGUNAKAN ALGORITMA K-MEANS CLUSTERING BERDASARKAN MODEL RFM SEBAGAI REKOMENDASI STRATEGI PEMASARAN (STUDI KASUS DI PT. XYZ)

Deni Riswana, - (2023) ANALISIS SEGMENTASI KONSUMEN MENGGUNAKAN ALGORITMA K-MEANS CLUSTERING BERDASARKAN MODEL RFM SEBAGAI REKOMENDASI STRATEGI PEMASARAN (STUDI KASUS DI PT. XYZ). S1 thesis, Universitas Pendidikan Indonesia.

Baca Full Text klik disini
[img] Text
S_BD_TSK_1905735_Title.pdf
Download (227kB)
[img] Text
S_BD_TSK_1905735_Chapter1.pdf
Download (126kB)
[img] Text
S_BD_TSK_1905735_Chapter2.pdf
Restricted to Staf Perpustakaan
Download (314kB)
[img] Text
S_BD_TSK_1905735_Chapter3.pdf
Download (101kB)
[img] Text
S_BD_TSK_1905735_Chapter4.pdf
Restricted to Staf Perpustakaan
Download (878kB)
[img] Text
S_BD_TSK_1905735_Chapter5.pdf
Download (49kB)
[img] Text
S_BD_TSK_1905735_Appendix.pdf
Restricted to Staf Perpustakaan
Download (287kB)
Official URL: http://repository.upi.edu

Abstract

Sejumlah besar perusahaan di dunia telah mengadopsi teknik data mining, perusahaan tidak hanya berfokus pada pengembangan produk saja, akan tetapi juga berfokus pada penggalian data konsumen yang dapat diidentifikasi melalui data transaksi konsumen. Berdasarkan pendekatan tersebut, peneliti bertujuan untuk melakukan analisis segmentasi konsumen dengan tujuan dapat mengidentifikasi karakteristik konsumen berdasarkan pola-pola yang terbentuk. Pada penelitian ini, peneliti menggunakan software RapidMiner dengan menggunakan algoritma K-Means clustering berdasarkan model RFM (Recency, Frequency, Monetary). Kemudian penerapan visualisasi data melalui Google Looker Studio bertujuan mempermudah dan mempercepat proses visualisasi. Metode penelitian yang diterapkan adalah Cross Industry Standard Process for Data Mining (CRISP-DM). Sumber data pada penelitian ini didapatkan dari database internal perusahaan. Hasil penelitian menunjukkan bahwa terdapat 3 kelompok (cluster) dan didapatkan nilai indeks Davies Bouldin sebesar 0.185 yang menguji validitas model cluster. Cluster Low loyalty terdiri dari 4222 konsumen, medium loyalty terdiri dari 4133 konsumen, dan high loyalty terdiri dari 4449 konsumen. Pola dan karakteristik dari setiap segmen konsumen dapat menjadi wawasan dan pengetahuan baru untuk pengembangan strategi pemasaran yang tepat dan terarah A significant number of companies worldwide have embraced data mining techniques, wherein companies not only develop products and services but also delve into consumer needs identifiable through transactional data and consumer behaviors. Built upon this approach, the researcher aims to conduct consumer segmentation analysis with the goal of identifying consumer characteristics based on emerging patterns. In this study, the researcher employs the RapidMiner software, utilizing the K-Means clustering algorithm based on the RFM (Recency, Frequency, Monetary) model. Subsequently, the application of data visualization through Google Looker Studio aims to simplify and expedite the visualization process. The applied research method adheres to the Cross Industry Standard Process for Data Mining (CRISP-DM). The data source utilized originates from the company's database. The research findings reveal the presence of three distinct clusters, yielding a Davies Bouldin index score of 0.185, validating the cluster model. The "Low loyalty" cluster consists of 4222 consumers, the "medium loyalty" cluster includes 4133 consumers, and the "high loyalty" cluster encompasses 4449 consumers. Patterns and characteristics unique to each consumer segment can provide insights and novel knowledge for the development of marketing strategies tailored to the attributes of each segment.

Item Type: Thesis (S1)
Additional Information: Link Google Scholar: https://scholar.google.com/citations?view_op=list_works&hl=en&user=6I1VyZ0AAAAJ ID SINTA Dosen Pembimbing: Oding Herdiana: 6745912 Rangga Gelar Guntara: 6738149
Uncontrolled Keywords: Data Mining, K-Means Clustering, Segmentasi Konsumen, Strategi Pemasaran
Subjects: L Education > L Education (General)
Divisions: UPI Kampus Tasikmalaya > S1 Bisnis Digital
Depositing User: Deni Riswana
Date Deposited: 26 Sep 2023 15:31
Last Modified: 26 Sep 2023 15:31
URI: http://repository.upi.edu/id/eprint/108153

Actions (login required)

View Item View Item