Claim Missing Document
Check
Articles

Found 28 Documents
Search

Customer Churn Prediction Pada Sektor Perbankan Dengan Model Logistic Regression dan Random Forest Mufida, Ely; Andriansyah, Doni; Hertyana, Hylenarti
Computer Science (CO-SCIENCE) Vol. 5 No. 1 (2025): Januari 2025
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/coscience.v5i1.7576

Abstract

– Customer churn is a detrimental phenomenon in the banking sector because it can reduce revenue and increase the cost of acquiring new customers. This research aims to compare the performance of two models, Logistic Regression and Random Forest, to predict customer churn using datasets from Kaggle. The research process involves data preprocessing such as z-score normalization and dividing the dataset into training data (70%) and testing data (30%). The model was evaluated using a confusion matrix with Accuracy, precision, recall and F1-Score values. Logistic Regression achieved 76.85% Accuracy, 79% precision, 94% recall, and 86% F1-Score, showing quite good performance but susceptible to false positives. In contrast, Random Forest shows superior performance with 83.12% Accuracy, 84% precision, 96% recall, and 90% F1-Score. Random Forest is suitable for problems with high recall requirements because it is more reliable in detecting potential customer churn. To further improve model performance, it is recommended to perform hyperparameter optimization and feature importance analysis. This churn prediction model is expected to help banks reduce churn and increase customer retention.
Pemanfaatan Aplikasi Mind Mapping Dalam Pembelajaran Pendidikan Agama Islam di SMP UBQ Nurul Islam Mojokerto Andriansyah, Doni; Soraya, Irma
Al-Ulum: Jurnal Pendidikan Islam Vol 5, No 2 (2024)
Publisher : Yayasan Rahmat Islamiyah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56114/al-ulum.v5i2.11585

Abstract

Education is a fundamental need for humans, because with education humans can develop physical and intellectual potential. Moreover, Islamic education, which is a guideline for Muslims in forming an obedient attitude and in accordance with religious law. Monotonous learning, namely using traditional learning methods, reduces the quality of education because it does not keep up with progress. However, the challenges in learning at UBQ Nurul Islam Mojokerto Middle School, the lack of asking students and outdated learning methods, make it necessary to innovate in learning, one of which is mind mapping. This research uses a quantitative approach to explain the application of mind mapping in PAI learning. The results show that this mindmapping method is quite well implemented by educators. Because by using the mind mapping method students become more active and make it easier for them to understand the learning.
Pemanfaatan Aplikasi Mind Mapping Dalam Pembelajaran Pendidikan Agama Islam di SMP UBQ Nurul Islam Mojokerto Andriansyah, Doni; Soraya, Irma
Al-Ulum: Jurnal Pendidikan Islam Vol 5, No 2 (2024)
Publisher : Yayasan Rahmat Islamiyah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56114/al-ulum.v5i2.11585

Abstract

Education is a fundamental need for humans, because with education humans can develop physical and intellectual potential. Moreover, Islamic education, which is a guideline for Muslims in forming an obedient attitude and in accordance with religious law. Monotonous learning, namely using traditional learning methods, reduces the quality of education because it does not keep up with progress. However, the challenges in learning at UBQ Nurul Islam Mojokerto Middle School, the lack of asking students and outdated learning methods, make it necessary to innovate in learning, one of which is mind mapping. This research uses a quantitative approach to explain the application of mind mapping in PAI learning. The results show that this mindmapping method is quite well implemented by educators. Because by using the mind mapping method students become more active and make it easier for them to understand the learning.
Optimalisasi Presensi Sekolah Berbasis QR Code dengan Metode Rapid Application Development Rahmawati, Eva; Brawijaya, Herlambang; Andriansyah, Doni; Mufida, Elly
Computer Science (CO-SCIENCE) Vol. 5 No. 2 (2025): Juli 2025
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/coscience.v5i2.8505

Abstract

High School attendance systems play an important role in monitoring student attendance and enforcing discipline in the academic environment. However, many schools still use manual methods such as written attendance lists or teacher name calling, which are inefficient, time-consuming, and prone to manipulation and fraud. These methods present challenges for teachers and administrative staff, leading to inaccurate recording, data loss, and falsification of attendance. To address these issues, this study proposes the development of a QR Code-based school attendance system using the Rapid Application Development (RAD) methodology. RAD was chosen because of its ability to produce prototypes quickly and allow for iterative system improvements according to user needs. The proposed system allows students to scan a unique QR Code to automatically record their attendance, thereby reducing human intervention and minimizing errors. The expected outcomes of this study include increased accuracy, efficiency, and security in recording student attendance. The RAD approach is predicted to accelerate the development process without sacrificing ease of use and system reliability. In addition, this system is expected to be able to prevent fraud in attendance, because QR Code-based authentication provides a more secure validation mechanism. Through a series of trials and evaluations, this study aims to prove that the integration of RAD with QR Code technology can improve the effectiveness of attendance recording compared to conventional methods. Based on the results of the trials and evaluations, it can be concluded that the QR Code-based attendance system with the RAD approach has been proven to improve the efficiency, accuracy, and security of the attendance system in schools.
Customer Churn Prediction Pada Sektor Perbankan Dengan Model Logistic Regression dan Random Forest Mufida, Ely; Andriansyah, Doni; Hertyana, Hylenarti; mufida, elly
Computer Science (CO-SCIENCE) Vol. 5 No. 1 (2025): Januari 2025
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/coscience.v5i1.7576

Abstract

– Customer churn is a detrimental phenomenon in the banking sector because it can reduce revenue and increase the cost of acquiring new customers. This research aims to compare the performance of two models, Logistic Regression and Random Forest, to predict customer churn using datasets from Kaggle. The research process involves data preprocessing such as z-score normalization and dividing the dataset into training data (70%) and testing data (30%). The model was evaluated using a confusion matrix with Accuracy, precision, recall and F1-Score values. Logistic Regression achieved 76.85% Accuracy, 79% precision, 94% recall, and 86% F1-Score, showing quite good performance but susceptible to false positives. In contrast, Random Forest shows superior performance with 83.12% Accuracy, 84% precision, 96% recall, and 90% F1-Score. Random Forest is suitable for problems with high recall requirements because it is more reliable in detecting potential customer churn. To further improve model performance, it is recommended to perform hyperparameter optimization and feature importance analysis. This churn prediction model is expected to help banks reduce churn and increase customer retention.
Pembelajaran Kontruktikvis pada Pembelajaran Fiqih di SMP UBQ Nurul Islam Mojokerto Guna Meningkatkan Berpikir Kritis Siswa Andriansyah, Doni
Nuris Journal of Education and Islamic Studies Vol. 4 No. 2: 2024
Publisher : STAI Nurul Islam Mojokerto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52620/jeis.v4i2.71

Abstract

Penelitian ini bertujuan untuk mengungkapkan implementasi teori belajar konstruktivisme pada pembelajaran Fiqih untuk melatih berpikir kritis peserta didik di kelas 7 SMP UBQ Nurul Islam. Metode penelitian ini bersifat kualitatif-deskriptif dengan pendektan field research. Teknik pengumpulan data yang digunakan dalam penelitian ini menggunakan observasi, wawancara, dan dokumentasi. Data yang terkumpul kemudian dianalisis menggunakan teknik analisis Miles & Hubbermann. Hasil dari penelitian ini diperoleh perencanaan, pelaksanaan, dan faktor pendukung dan penghambat implementasi teori pembelajaran konstruktivis. Adapun perencanaan pembelajaran konstruktivis yakni menyiapkan modul ajar, media ajar, dan perangkat pembelajaran. Pada tahap pelaksanaan terdapat kegiatan pendahuluan dengan membaca do’a-do’a dan apersepsi yang terdapat pertanyaan-pertanyaan pemicu siswa untuk berpikir, kegiatan inti berupa membentuk kelompok diskusi dan membuat peta konsep kemudian siswa diberikan tugas mengamati, menanya, menalar, menganalisis, dan mengkomunikasikan hasil diskusi. Adapun faktor pendukung penerapan pembelajaran konstruktivis ini diantaranya adalah: dukungan dari pihak sekolah, kompetensi guru, semangat belajar siswa, dan fasilitas. Sedangkan faktor penghambatnya adalah keterbatasan waktu atau kurangnya jam pelajaran.
Klasifikasi Kualitas Buah Pisang Berdasarkan Waktu Panen dan Tingkat Kematangan Menggunakan Metode SVM dan KNN Doni Andriansyah; Mufida, Elly
SATIN - Sains dan Teknologi Informasi Vol 10 No 1 (2024): SATIN - Sains dan Teknologi Informasi
Publisher : STMIK Amik Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33372/stn.v10i1.1131

Abstract

Tanaman pisang atau Musa Paradisiaca merupakan tanaman yang masuk kedalam golongan klimakterik, sehingga memerlukan perhatian khusus pasca panen. Tingkat kematangan buah pisang saat dipanen sangat mempengaruhi daya simpan dan kualitas buah. Waktu panen sangat penting untuk mendapatkan buah yang matang dan berkualitas. Penelitian menggunakan algoritma SVM dan KNN dengan tujuan untuk mengetahui algortima terbaik dalam klasifikasi kualitas buah pisang. Kumpulan data yang digunakan merupakan data publik mengenai kualitas buah pisang dengan jumlah data sebanyak 8.000 baris data, dan dengan delapan atribut kolom. Dalam pengolahan data hanya menggunakan atribut kolom waktu panen dan tingkat kematangan serta dilakukan proses pengacakan terhadap kumpulan data agar model dapat belajar lebih baik dan mencegah data dari bias. Hasil penelitian menunjukkan klasifikasi dengan SVM memiliki nilai akurasi sebesar 73,4%, lebih baik dari hasil klasifikasi KNN yang hanya mencapai 69,6%.
Prediksi Kualitas Udara Daerah Tangerang Selatan Melalui Parameter ISPU dengan Metode LSTM Andriansyah, Doni; Hertyana, Hylenarti; mufida, elly
Jurnal INSAN Journal of Information System Management Innovation Vol. 5 No. 2 (2025): Desember 2025
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/j-insan.v5i2.10352

Abstract

Indeks Standar Pencemar Udara (ISPU) adalah angka yang tidak mempunyai satuan yang menggambarkan kondisi mutu udara ambien di suatu lokasi, berdasarkan dampaknya terhadap kesehatan manusia, estetika, serta makhluk hidup lainnya. Tanggerang Selatan adalah salah satu kota di Provinsi Banten, Indonesia, yang merupakan bagian dari wilayah Jabodetabek. Tangerang Selatan berkembang pesat sebagai kota urban dan satelit Jakarta, dan menjadi area industri ringan dan komersial. Penelitian ini menggunakan model Long Short-Term Memory (LSTM) untuk memprediksi angka ISPU pada kota Tanggerang Selatan. Berdasarkan hasil penelitian, model LSTM mampu memprediksi kualitas udara di Tangerang Selatan dengan akurasi yang cukup tinggi. Model berhasil mengikuti pola data historis selama 30 hari dengan baik dan menghasilkan prakiraan 7 hari ke depan yang cukup mendekati nilai aktual. Hal ini ditunjukkan oleh nilai evaluasi yang memuaskan, yaitu MSE sebesar 15,15, RMSE sebesar 3,89, dan MAE sebesar 2,86, mengindikasikan rata-rata kesalahan prediksi yang relatif kecil. Perbedaan antara nilai prediksi dan aktual kualitas udara harian berada dalam rentang yang dapat diterima.