Analisis Faktor-Faktor yang Mempengaruhi Churn Rate pada Perusahaan Telekomunikasi Menggunakan Metode Support Vector Machines (Studi Kasus: PT Telekomunikasi XYZ)

Arifin, Samsul (2018) Analisis Faktor-Faktor yang Mempengaruhi Churn Rate pada Perusahaan Telekomunikasi Menggunakan Metode Support Vector Machines (Studi Kasus: PT Telekomunikasi XYZ). Masters thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
09211650053011-Master_Thesis.pdf - Published Version
Restricted to Repository staff only

Download (3MB) | Request a copy

Abstract

Dalam dunia telekomunikasi, pelanggan adalah aset yang paling berharga bagi perusahaan untuk keberlangsungan proses bisnis. Perpindahan pelanggan tentunya tidak diharapkan yang menuntut perusahaan perlu melakukan analisis dengan melakukan segmentasi pelanggan untuk mempermudah dilakukannya segmentasi bisnis dalam mendukung pengambilan keputusan yang tepat dan terarah, keputusan yang tepat diharapkan menghasilkan value yang dapat meningkatkan kinerja dan profit perusahaan. Dalam melakukan analisis data pelanggan, data masa lalu adalah mutlak dibutuhkan dengan mengambil variabel yang dapat menggambarkan keadaan pelanggan di masa yang akan datang, dengan melakukan analisa potensi churn dan non churn.. Dalam hal ini dilakukan kalsifikasi data mining menggunakan metode Support Vector Machine. Metode SVM dipilih karena akurasi yang cukup tinggi dalam melakukan klasifikasi. Dari pengujian ini akan terbentuk performance yang menggambarkan hasil klasifikasi berupa churn dan non churn. Penelitian ini mengambil 7 variabel sebagai uji coba di mana hasil yang telah dlakukan, terdapat 4 variabel dependen yang mempengaruhi churn rate secara signifikan dengan nilai X2 lebih besar dari Xtabel atau nilai Pvalue lebih kecil dari 0.05 yaitu variabel Handset, Usage Data in kb, Voice in Minutes dan Reload. Model yang dihasilkan kemudian diuji dengan SVM, dengan performance lebih baik dibandingkan dengan data awal. Akurasi terbaik ada pada 78.2% dengan proporsi 70 data training dan 30 data testing. Sedangkan pada data sebelum model, akurasi terbaik 73.9%, sehingga model yang dihasilkan lebih baik jika dibandingkan data awal. =================== In the telecommunications industry, the customer is the most valuable asset for the company to sustain the business process. The moving out of customers is certainly not expected so that companies need to analyze the customers profile to facilitate the conduct of business segmentation in support of decision making, the right decision is expected to generate value that can improve the performance and profit of the company. In analyzing customer data, past data is absolutely necessary with variables that can describe the customer value in the future, whether the customer is potentially churn or non churn. In this case, using Support Vector Machine method for classification. The SVM method was chosen because of its high accuracy in classification prediction. From this test will formed a performance that describes the results of the classification of churn and non churn. In this research, there are 7 attributs have been used for testing. there are 3 attributes that significantly influence the churn rate of total 7 attribute which have been tested. Usage Data in kb and Voice in Minutes and Reload where these attributes have a performance value smaller than 5% of the total performance of the overall attribute. While the attribute is approaching the significant in SMS that is equal to 9%. From these results, the telecommunications companies XYZ should maintain Data and Voice services in minimizing churn rate. This study took 7 variables as a trial where the results have been used for testing, there are 4 dependent variables that affect or influence churn rate significantly with X2 value greater than Xtabel or value of Pvalue smaller than 0.05. Those variables are Handset, Usage Data in kb, Voice in Minutes and Reload. The resulting model have been tested with SVM, with better performance than the initial data. The best accuracy is 78.2% with the proportion of 70 training data and 30 data testing. While in the data before the model, the best accuracy is 73.9%, so the resulting model is better when compared to the initial data.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Data mining, SVM, Pelanggan Churn, Telekomunikasi, Data mining; SVM; Customer Churn; Telecommunications
Subjects: T Technology > T Technology (General) > T57.5 Data Processing
T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing
Divisions: Faculty of Business and Management Technology > Management Technology > (S2) Master Theses
Depositing User: Arifin Samsul
Date Deposited: 21 Nov 2018 02:32
Last Modified: 21 Nov 2018 02:32
URI: http://repository.its.ac.id/id/eprint/52968

Actions (login required)

View Item View Item