Melola Discovery

← Melola Discovery

Akses terbuka Jurnal SINTA 1

Customer Decision Prediction Using Deep Neural Network on Telco Customer Churn Data

Jurnal Elektronika dan Telekomunikasi · Vol. 21 No. 2 (2021) · hlm. 122–127

DOI
10.14203/jet.v21.122-127
ISSN
1411-8289, 2527-9955
Terbit
2021-12-31
Penerbit
Indonesian Institute of Sciences
Bahasa
en
Lisensi
creativecommons.org/licenses/by-nc-sa/4.0/

Abstrak

Customer churn is the most important problem in the business world, especially in the telecommunications industry, because it greatly influences company profits. Getting new customers for a company is much more difficult and expensive than retaining existing customers. Machine learning, part of data mining, is a sub-field of artificial intelligence widely used to make predictions, including predicting customer churn. Deep neural network (DNN) has been used for churn prediction, but selecting hyperparameters in modeling requires more time and effort, making the process more challenging for the researcher. Therefore, the purpose of this study is to propose a better architecture for the DNN algorithm by using a hard tuner to obtain more optimal hyperparameters. The tuning hyperparameter used is random search in determining the number of nodes in each hidden layer, dropout, and learning rate. In addition, this study also uses three variations of the number of hidden layers, two variations of the activation function, namely rectified linear unit (ReLu) and Sigmoid, then uses five variations of the optimizer (stochastic gradient descent (SGD), adaptive moment estimation (Adam), adaptive gradient algorithm (Adagrad), Adadelta, and root mean square propagation (RMSprop)). Experiments show that the DNN algorithm using hyperparameter tuning random search produces a performance value of 83.09 % accuracy using three hidden layers, the number of nodes in each hidden layer is [20, 35, 15], using the RMSprop optimizer, dropout 0.1, the learning rate is 0.01, with the fastest tuning time of 21 seconds. Better than modeling using k-nearest neighbor (K-NN), random forest (RF), and decision tree (DT) as comparison algorithms.

Kata kunci

Jaringan kolaborasi para penulis

31 penulis · 124 tautan kepenulisan bersama · 80 lintas organisasi

Arahkan kursor atau ketuk sebuah nama untuk menyorot rekan penulisnya. Klik dua kali untuk membuka halaman penulis.

  • Indonesian Institute of Sciences
  • University of Brawijaya
  • National Research and Innovation Agency
  • Universitas Nusa Mandiri
  • Indonesia Research Institute for Tea and Cinchona
  • Sebelas Maret University
  • Padjadjaran University
  • Organisasi lain
  • Organisasi berbeda
  • Organisasi sama
  • Afiliasi tidak diketahui
  • Garis lebih tebal = lebih banyak artikel bersama · lingkaran lebih besar = lebih banyak artikel
Tampilkan sebagai tabel
Rekan penulisOrganisasiArtikel bersama
Asri Rizki YulianiIndonesian Institute of Sciences4
Endang SuryawatiNational Research and Innovation Agency4
Ade RamdanIndonesian Institute of Sciences4
Vicky ZilvanIndonesian Institute of Sciences3
R. Budiarianto Suryo KusumoIndonesian Institute of Sciences3
Ana HeryanaNational Research and Innovation Agency3
Raden Sandra YuwanaIndonesian Institute of Sciences3
Fani FauziahIndonesia Research Institute for Tea and Cinchona3
Dikdik KrisnandiIndonesian Institute of Sciences3
Ahmad Afif SupiantoUniversity of Brawijaya2
Fitra Septia NugrahaUniversitas Nusa Mandiri1
Ade RamdanIndonesian Institute of Sciences1
Tibyani TibyaniUniversity of Brawijaya1
Mochammad IzzuddinUniversity of Brawijaya1
Syaiful AnamUniversity of Brawijaya1
Onky PrasetyoUniversity of Brawijaya1
Rika SustikaIndonesian Institute of Sciences1
Vitria Puspitasari RahadiIndonesia Research Institute for Tea and Cinchona1
Syifa Auliyah HasanahPadjadjaran University1
Jimmy Abdel KadarNational Research and Innovation Agency1
Purwoko AdhiNational Research and Innovation Agency1
M. Faizal AmriIndonesian Institute of Sciences1
Mawadatul MaulidahUniversitas Bina Sarana Informatika1
Aldi JakariaUniversitas Nusa Mandiri1
Heri SyahrianIndonesia Research Institute for Tea and Cinchona1
Oka MahendraIndonesian Institute of Sciences1
Rico DahlanIndonesian Institute of Sciences1
Hasih PratiwiSebelas Maret University1
Arnetta Listiana DewiSebelas Maret University1

Afiliasi berasal dari OpenAlex; metadata OAI-PMH jurnal tidak memuatnya. Penulis tanpa ID OpenAlex dicocokkan berdasarkan nama, sehingga dua orang bisa tergabung atau satu orang bisa terpisah.

Sumber metadata: OpenAlex API (jurnal SINTA, ADR 0048) https://openalex.org/S4210212760, pertama di-harvest 21 Nov 2026, terakhir terlihat 21 Nov 2026.
Metadata asli (oai_dc)
<oai_dc:dc xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/"><dc:title xml:lang="en">Customer Decision Prediction Using Deep Neural Network on Telco Customer Churn Data</dc:title><dc:creator>Hiya Nalatissifa</dc:creator><dc:creator>Hilman F. Pardede</dc:creator><dc:subject>customer churn prediction</dc:subject><dc:subject>telecommunications customers</dc:subject><dc:subject>deep learning architectures</dc:subject><dc:subject>hyperparameter optimization</dc:subject><dc:subject>random search</dc:subject><dc:subject>hidden layer size</dc:subject><dc:subject>dropout</dc:subject><dc:subject>learning rate</dc:subject><dc:subject>activation functions</dc:subject><dc:subject>optimizer comparison</dc:subject><dc:subject>RMSProp</dc:subject><dc:subject>ReLU</dc:subject><dc:subject>customer retention</dc:subject><dc:description xml:lang="en">Customer churn is the most important problem in the business world, especially in the telecommunications industry, because it greatly influences company profits. Getting new customers for a company is much more difficult and expensive than retaining existing customers. Machine learning, part of data mining, is a sub-field of artificial intelligence widely used to make predictions, including predicting customer churn. Deep neural network (DNN) has been used for churn prediction, but selecting hyperparameters in modeling requires more time and effort, making the process more challenging for the researcher. Therefore, the purpose of this study is to propose a better architecture for the DNN algorithm by using a hard tuner to obtain more optimal hyperparameters. The tuning hyperparameter used is random search in determining the number of nodes in each hidden layer, dropout, and learning rate. In addition, this study also uses three variations of the number of hidden layers, two variations of the activation function, namely rectified linear unit (ReLu) and Sigmoid, then uses five variations of the optimizer (stochastic gradient descent (SGD), adaptive moment estimation (Adam), adaptive gradient algorithm (Adagrad), Adadelta, and root mean square propagation (RMSprop)). Experiments show that the DNN algorithm using hyperparameter tuning random search produces a performance value of 83.09 % accuracy using three hidden layers, the number of nodes in each hidden layer is [20, 35, 15], using the RMSprop optimizer, dropout 0.1, the learning rate is 0.01, with the fastest tuning time of 21 seconds. Better than modeling using k-nearest neighbor (K-NN), random forest (RF), and decision tree (DT) as comparison algorithms.</dc:description><dc:publisher>Indonesian Institute of Sciences</dc:publisher><dc:date>2021-12-31</dc:date><dc:type>info:eu-repo/semantics/article</dc:type><dc:format>application/pdf</dc:format><dc:identifier>https://doi.org/10.14203/jet.v21.122-127</dc:identifier><dc:identifier>10.14203/jet.v21.122-127</dc:identifier><dc:source>Jurnal Elektronika dan Telekomunikasi; Vol 21 No 2 (2021); 122-127</dc:source><dc:source>1411-8289</dc:source><dc:source>2527-9955</dc:source><dc:language>en</dc:language><dc:relation>https://www.jurnalet.com/jet/article/download/427/367</dc:relation><dc:rights>https://creativecommons.org/licenses/by-nc-sa/4.0/</dc:rights></oai_dc:dc>