Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal JOURNAL OF SCIENCE AND SOCIAL RESEARCH Bulletin of Computer Science Research
Andree Rizky Yuliansyah Siregar
Unknown Affiliation
Computer Science & IT Economics, Econometrics & Finance Education Social Sciences

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Analisis Stok Abu Batu Menggunakan Metode Fuzzy Sugeno Pada Perencanaan Kebutuhan Infrastruktur Andree Rizky Yuliansyah Siregar; Sri Wahyuni
Bulletin of Computer Science Research Vol. 5 No. 1 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v5i1.409

Abstract

Infrastructure needs planning requires careful consideration of the availability and needs of raw materials, such as fly ash, which is often used in various construction projects. One of the main challenges in this planning is the uncertainty in predicting the supply and demand of fly ash. To overcome this problem, this study applies the Fuzzy Sugeno method, which is an approach in Fuzzy logic, to analyze fly ash stocks and plan infrastructure needs more effectively. This method allows processing uncertain and subjective data, such as estimates of fly ash stock and demand that can vary over time. Using the Fuzzy Sugeno model, calculations are carried out to determine the optimal stock level and predict long-term fly ash needs. The results of this study are expected to provide a more accurate picture of fly ash needs planning and assist related parties in making more efficient decisions in infrastructure planning. The results of the process carried out from January 2023 to December 2023 obtained a mape value of 30%, this value is included in the reasonable category.
PREDIKSI CUSTOMER CHURN PADA LAYANAN INDIHOME MENGGUNAKAN ALGORITMA DECISION TREE (STUDI KASUS PT. TELKOM AKSES) Andree Rizky Yuliansyah Siregar; Muhammad Iqbal
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol. 8 No. 1 (2025): February 2025
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v8i1.2698

Abstract

Abstract: This study aims to predict Customer Churn for IndiHome services at PT. Telkom Akses using the Decision Tree algorithm. The research analyzes several variables that influence customers' decisions to discontinue subscriptions, including complaints, service packages, and subscription duration. The analysis results show that the Complaint attribute has the highest Gain value (0.6898), making it the main factor distinguishing customers likely to churn from those who remain. This suggests a strong correlation between complaint frequency and the customers’ decision to continue or terminate the service. Additionally, the Package attribute also has a significant impact, with a Gain of 0.2785, indicating that the selected package speed influences the likelihood of churn. Conversely, the Subscription Duration attribute has the lowest Gain (0.0411), indicating that this variable provides minimal information in predicting churn. Based on these findings, PT. Telkom Akses is recommended to consider improving complaint management and optimizing service package offerings to reduce Customer Churn rates.Keyword: Customer Churn; Decision Tree; Prediction Abstrak: Penelitian ini bertujuan untuk memprediksi Customer Churn pada layanan IndiHome di PT. Telkom Akses menggunakan algoritma Decision Tree. Penelitian ini dilakukan dengan menganalisis sejumlah variabel yang mempengaruhi keputusan pelanggan untuk berhenti berlangganan, antara lain: keluhan, paket layanan, dan lama berlangganan. Hasil analisis menunjukkan bahwa atribut Keluhan memiliki nilai Gain tertinggi (0.6898), menjadikannya faktor utama dalam membedakan pelanggan yang cenderung berhenti berlangganan (churn) dan yang tetap (tidak churn). Hal ini menunjukkan adanya korelasi yang kuat antara frekuensi keluhan dengan keputusan pelanggan untuk terus menggunakan layanan. Selain itu, atribut Paket juga memiliki pengaruh signifikan dengan Gain 0.2785, yang mengindikasikan bahwa kecepatan paket layanan memengaruhi kemungkinan pelanggan untuk churn. Sebaliknya, atribut Lama Berlangganan memiliki Gain terendah (0.0411), yang menunjukkan bahwa variabel ini kurang informatif dalam memprediksi churn. Dengan hasil ini, PT. Telkom Akses disarankan untuk mempertimbangkan pengelolaan keluhan pelanggan serta pemilihan paket layanan yang sesuai guna mengurangi tingkat Customer Churn.Kata kunci: Customer Churn; Decision Tree; Prediksi
Co-Authors Muhammad Iqbal Sri Wahyuni