Segmentasi pelanggan kartu halo telkomsel berbasis k-means di wilayah sumbagsel. Segmentasi pelanggan Kartu Halo Telkomsel di Sumbagsel menggunakan K-Means untuk memahami perilaku pengguna dan membuat strategi pemasaran efektif berbasis data. Temukan pola penggunaan layanan digital.
Perusahaan telekomunikasi dalam menjalankan operasionalnya harus mengelola banyak data pelanggan, karena penggunaan layanan digital terus bertambah. Namun, data tersebut biasanya hanya digunakan untuk urusan administratif, sehingga kemampuan untuk menganalisis data guna memahami perilaku pelanggan belum sepenuhnya dimanfaatkan. Penelitian ini secara khusus bertujuan untuk mengelompokkan pelanggan berdasarkan pola penggunaan layanan guna mengidentifikasi karakteristik setiap segmen secara lebih sistematis. Data yang digunakan merupakan data pelanggan pascabayar Kartu Halo wilayah Sumbagsel tahun 2025 yang diperoleh dari PT Telkomsel Smart Office Palembang dengan total 4.645 data. Metode yang digunakan adalah Knowledge Discovery in Databases (KDD), yang terdiri dari beberapa tahap yaitu preprocessing, transformation, dan data mining. Proses pengelompokan dilakukan dengan menggunakan algoritma K-Means, bantuan perangkat lunak RapidMiner. Penelitian menunjukkan bahwa pelanggan bisa dibagi menjadi tiga kelompok dengan kebiasaan menggunakan layanan yang berbeda. Evaluasi menggunakan Davies-Bouldin Index menunjukkan bahwa model clustering memiliki kualitas yang baik dalam memisahkan antarkelompok. Secara ilmiah, penelitian ini membantu dalam penerapan metode clustering untuk membagi pelanggan dalam industri telekomunikasi. Secara nyata, hasil ini bisa dipakai untuk membantu membuat strategi pemasaran yang lebih tepat dan didasarkan pada data.
This paper, "Segmentasi Pelanggan Kartu Halo Telkomsel Berbasis K-Means di Wilayah Sumbagsel," addresses a highly relevant and timely issue within the telecommunications industry: the strategic utilization of vast customer data. The authors aim to move beyond mere administrative data usage to systematically segment postpaid Kartu Halo customers in the Sumbagsel region based on their service consumption patterns. Employing the Knowledge Discovery in Databases (KDD) methodology, with K-Means clustering facilitated by RapidMiner, the study processes a substantial dataset of 4,645 customer records. The core finding indicates the successful identification of three distinct customer groups characterized by varying service usage habits. A significant strength of this research lies in its practical applicability and robust methodology. By leveraging real-world data from PT Telkomsel Smart Office Palembang, the study provides tangible insights into customer behavior, a critical step often overlooked in data-rich environments. The KDD framework ensures a systematic approach from preprocessing to data mining, enhancing the reliability of the results. The choice of the K-Means algorithm is appropriate for customer segmentation, and the validation using the Davies-Bouldin Index attests to the quality and distinctiveness of the identified clusters. This scientific application of clustering directly contributes to bridging the gap between raw data and actionable business intelligence, promising more targeted and data-driven marketing strategies for Telkomsel. While the study clearly demonstrates its value, further elaboration on the specific characteristics and profiles of the three identified segments would significantly enhance its impact. Detailing the distinguishing features of each group—for instance, high-value users, budget-conscious users, or specific service-centric users—would provide clearer strategic implications beyond a general statement of "different service usage habits." Additionally, the abstract mentions using data from "tahun 2025," which appears to be a typographical error, and should ideally be clarified to reflect actual past data. Future work could also explore the temporal stability of these segments or integrate additional demographic or churn-related variables to enrich the segmentation. Nevertheless, this research represents a commendable effort in transforming raw customer data into valuable strategic assets for the telecommunications sector.
You need to be logged in to view the full text and Download file of this article - Segmentasi Pelanggan Kartu Halo Telkomsel Berbasis K-Means di Wilayah Sumbagsel from MALCOM: Indonesian Journal of Machine Learning and Computer Science .
Login to View Full Text And DownloadYou need to be logged in to post a comment.
By Sciaria
By Sciaria
By Sciaria
By Sciaria
By Sciaria
By Sciaria