Prediksi Pelanggan Churn Menggunakan Pendekatan RFM dan Extreme Gradient Boosting untuk Rekomendasi Strategi Perusahaan

Putris, Nadhifa Afrinia Dwi (2023) Prediksi Pelanggan Churn Menggunakan Pendekatan RFM dan Extreme Gradient Boosting untuk Rekomendasi Strategi Perusahaan. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06111940000073-Undergraduate_Thesis.pdf] Text
06111940000073-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2025.

Download (4MB) | Request a copy

Abstract

Pelanggan merupakan salah satu aset penting dalam perkembangan perusahaan bisnis. Pelanggan dapat saja meninggalkan perusahaan dan berpindah menggunakan produk atau jasa perusahaan lain karena alasan tertentu. Peristiwa ini disebut dengan customer churn. Apabila perusahaan kehilangan pelanggannya maka akan berpengaruh pada pendapatan perusahaan yang semakin menurun. PT Kasir Pintar Internasional mengalami perubahan dinamis terhadap pelanggannya dan terdapat banyaknya kompetitor pada bisnis teknologi kasir UMKM. Oleh karena itu, pada penelitian ini dilakukan segmentasi pelanggan menggunakan pendekatan analisis recency, frequency, dan monetary (RFM) sebagai perhatian perusahaan terhadap karakteristik pelanggan dan memprediksi klasifikasi customer churn menggunakan metode Extreme Gradient Boosting (XGBOOST). Berdasarkan hasil evaluasi pada metode XGBoost dapat menjadi masukan untuk proses analitik preskriptif dengan memperhatikan fitur-fitur penting yang berpengaruh pada hasil evaluasi. Selanjutnya akan diberikan rekomendasi strategi dan tindakan untuk mengatasi permasalahan churn. Hasil yang diperoleh adalah pelanggan dibagi menjadi tiga segmentasi sesuai dengan skor RFM dan segmentasi ketiga berhasil memperoleh nilai accuracy, presisi, recall, dan f1-score tertinggi, masing-masing sebesar 0.98833, 0.98768, 0.98899, dan 0.98833. Diketahui karakteristik segmentasi, representasi data, dan pendekatan perilaku dengan analisis RFM berpengaruh pada peningkatan keakuratan model dalam prediksi churn. Untuk rekomendasi strategi diberikan berdasarkan tiga fitur penting tertinggi, yaitu berdasarkan kota, perolehan GMV, dan perolehan total transaksi pelanggan.
==================================================================================================================================
Customers are one of the important assets in the development of business companies. Customers may leave the company and switch to using other companies’ products or services for certain reasons. This event is called customer churn. If the company loses its customers, it will affect the company’s decreasing revenue. PT Kasir Pintar Internasional experiences dynamic changes in its customers and there are many competitors in the MSME cashier technology business. Therefore, in this study, customer segmentation is carried out using the recency, frequency, and monetary (RFM) analysis approach as a company’s attention to customer characteristics and predicting customer churn classification using the Extreme Gradient Boosting (XGBOOST) method. Based on the evaluation results in the XGBoost method, it can be an input for the prescriptive analytic process by paying attention to important features that affect the evaluation results. The results obtained are customers divided into three segmentations according to the RFM score and the third segmentation successfully obtained the highest accuracy, precision, recall, and f1-score values, respectively 0.98833, 0.98768, 0.98899, and 0.98833. It is known that the segmentation characteristics, data representation, and behavioral approach with RFM analysis have an effect on improving the accuracy of the model in the analysis churn prediction. Strategy recommendations are based on the three highest important features, namely by city, GMV acquisition, and total customer transaction acquisition.

Item Type: Thesis (Other)
Uncontrolled Keywords: Customer Churn, Pelanggan, Perusahaan, Segmentasi, Customer, Company, Segmentation, XGBoost
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science)
Q Science > QA Mathematics > QA76.F56 Data structures (Computer science)
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User: Nadhifa Afrinia Dwi Putris
Date Deposited: 06 Oct 2023 02:02
Last Modified: 06 Oct 2023 02:02
URI: http://repository.its.ac.id/id/eprint/100757

Actions (login required)

View Item View Item