Claim Missing Document
Check
Articles

Found 36 Documents
Search

Data Mining with Logistic Regression and Support Vector Machine for Hepatitis Disease Diagnosis Amrin, Amrin; Rudianto, Rudianto; Sismadi , Sismadi
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 8 No. 2 (2025): Issues January 2025
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v8i2.13218

Abstract

Hepatitis is a chronic and dangerous disease that can lead to death. Making early predictions to detect hepatitis is very important because many people still underestimate the disease. These predictions can be made by collecting patient data or health examination results, so that preventive measures can be taken faster and better. Early diagnosis of the disease is important for prompt management and treatment. The right stage of diagnosis activities and accurate disease prediction in time can save many patients. The magnitude of this disease problem in Indonesia can be known from various studies, studies, and disease observation activities. In this study, researchers will apply and compare data mining classification methods, namely the Logistic Regression method and Support Vector Machine to diagnose hepatitis disease. Based on the research, it is known that the Logistic Regression method has an accuracy rate of 84.62% and an under the curve (AUC) value of 0.841, then the Support Vector Machine method has an accuracy rate of 87% and an AUC value of 0.865. From the t-test results, it can be seen that there is no significant difference between the Logistic Regression and Support Vector Machine methods, because the value = 0.520>0.05. This shows that the Logistic Regression method has almost the same performance as the Support Vector Machine method. Hopefully the results of this research can help doctors determine a diagnosis more quickly and reduce the possibility of misdiagnosis so that early detection of hepatitis can be carried out more widely, especially in remote areas with limited health facilities
Analisa Komparasi Model Data Mining Algoritma C4.5, CHAID, dan Random Forest Untuk Penilaian Kelayakan Kredit Amrin, Amrin; Pahlevi, Omar; Rianto, Harsih
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.v5i1.6208

Abstract

Credit has now become a trend in society. The problem with credit is the improper history of credit card usage. The resulting impact can lead to bad credit. If customers fail to pay off debts that have been agreed upon with the bank, they can increase their credit risk. This study aims to conduct a comparative analysis of three data mining classification methods: the C4.5 algorithm, Chi-Squared Automatic Interaction Detection (CHAID), and Random Forest. The goal is to classify creditworthiness status. The researcher used 481 vehicle credit records with "bad" and "good" reviews. In this study, the independent variables used are dependent status, age, marital status, occupation, income, employment status, company status, last education, length of stay, house condition, and down payment. For creditworthiness assessment, the C4.5 model shows a truth accuracy rate of 91.90% with an area under the curve (AUC) value of 0.915. The CHAID model shows a truth accuracy rate of 63.83% with an AUC value of 0.661, and the Random Forest model shows a truth accuracy rate of 78.60% with an AUC value of 0.907. The evaluation results show that both the Random Forest and C4.5 algorithms have high accuracy rates and AUC values.
SOFTWARE DEFECT PREDICTION TRENDS: A BIBLIOMETRIC ANALYSIS OF MACHINE AND DEEP LEARNING Harsih Rianto; Omar Pahlevi; Desmulyati; Amrin; Ade Surya Budiman; Budi Supriyadi
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 11 No. 3 (2026): JITK Issue February 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v11i3.7351

Abstract

This study provides a comprehensive bibliometric mapping of global research trends and emerging frontiers in Software Defect Prediction (SDP), emphasizing the integration of machine learning (ML) and deep learning (DL) approaches. Unlike previous bibliometric surveys that focused narrowly on metric-based or short-term analyses, this work offers a broader and more integrated perspective on the intellectual evolution, collaboration patterns, and thematic directions in SDP research. Using data retrieved from the Scopus database and analyzed through Bibliometrix and VOSviewer, the study systematically applied the PRISMA protocol to ensure transparency and replicability. A total of 1,549 publications were examined, revealing a steady increase in scientific output dominated by China, India, and the United States. Thematic and keyword analyses identified five core clusters that trace the paradigm shift from traditional statistical models to advanced ML- and DL-driven predictive frameworks. Emerging topics such as transfer learning, cross-project prediction, and explainable AI (XAI) were identified as promising frontiers shaping the next phase of software quality prediction research. Beyond mapping academic progress, this study contributes strategic insights for researchers seeking to identify research gaps, industry practitioners developing intelligent defect prediction tools, and policymakers designing AI-driven software quality initiatives
Model Waterfall Untuk Rancang Bangun Sistem Informasi Pengadaan Mesin EDC Pada E-Channel Operations Perbankan Amrin Amrin; Muhammad Reza Aldiansyah
INSANtek Vol. 2 No. 2 (2021): November 2021
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/instk.v2i2.668

Abstract

Dalam era globalisasi sekarang ini, teknologi informasi melaju dengan cepatnya. Adapun komputer yang merupakan peralatan yang diciptakan untuk mempermudah pekerjaan manusia, saat mencapai kemajuan baik di dalam pembuatan hardware maupun software. E-Channel Operations Perbankan membutuhkan sekali adanya suatu sistem informasi yang menunjang dan memberikan pelayanan yang memudahkan bagi para karyawannya. Sistem pengadaan yang digunakan E-Channel Operations Perbankan masih dilakukan secara manual, yaitu dengan membuat surat permintaan setiap kali melakukan pengadaan mesin baru. Perancangan sistem informasi pengadaan mesin EDC berbasis web merupakan solusi yang terbaik untuk memecahkan permasalahan- permasalahan yang ada pada perusahaan ini, serta dengan sistem yang terkomputerisasi dapat tercapai suatu kegiatan yang efektif dan efisien dalam menunjang aktifitas pada perusahaan ini. Metode yang digunakan untuk pengembangan perangkat lunak adalah metode waterfall. Hasil dari penelitian ini adalah dengan adanya aplikasi pengadaan mesin EDC, dapat mempermudah pegawai yang terkait dalam proses pengadaan secara real time, pembuatan laporan dapat dilakukan secara cepat dan tepat.serta penyimpanan data menjadi lebih aman.
Improving Liver Disease Diagnosis Accuracy Using Synthetic Minority Oversampling Technique and Particle Swarm Optimization Harsih Rianto; Amrin amrin
JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Vol. 10 No. 1 (2026): Issues July 2026
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jite.v10i1.17943

Abstract

This study aims to improve liver disease diagnosis accuracy using the Synthetic Minority Oversampling Technique (SMOTE) and Particle Swarm Optimization (PSO). The dataset used in this study is the Indian Liver Patient Dataset (ILPD), which consists of 583 patient records with 10 input attributes and one class attribute. The main problem in this dataset is the imbalance of class distribution between liver and non-liver patients, which may affect the performance of classification models. The research stages include data preprocessing, missing value imputation, stratified 70:30 train-test splitting, applying SMOTE to the training data, feature selection using PSO, and evaluating several classification models, namely Decision Tree (DT), Support Vector Machine (SVM), Random Forest (RF), Naive Bayes (NB), and Logistic Regression (LR). The experimental results show that SMOTE-RF achieved the best overall performance with an AUC of 0.8610, accuracy of 0.7784, F1-score of 0.7783, and MCC of 0.5570. Meanwhile, SMOTE-PSO-DT improved the Tree model by selecting five important attributes, namely Direct Bilirubin, SGOT, Total Proteins, Albumin, and A/G Ratio. These results indicate that SMOTE effectively improves classification performance, while PSO helps simplify the model through feature selection.
Optimasi Kinerja Decision Tree C4.5 dengan Metode Backward Elimination pada Sistem Penilaian Kredit Amrin Amrin; Andi Diah Kuswanto; Ahmad Fauzi
IMTechno: Journal of Industrial Management and Technology Vol. 7 No. 1 (2026): Vol. 7 No. 1 (2026): Januari 2026
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/imtechno.v7i1.11087

Abstract

Penggunaan kredit telah menjadi fenomena yang meluas di masyarakat, namun di balik popularitasnya terdapat risiko signifikan berupa peningkatan kredit macet. Masalah ini sering kali berakar pada histori pengelolaan keuangan yang buruk, di mana kegagalan nasabah untuk memenuhi kewajiban pembayaran utang secara langsung meningkatkan risiko kredit bagi lembaga pemberi pinjaman. Untuk mengatasi tantangan ini, penelitian ini mengimplementasikan dan membandingkan dua model klasifikasi untuk menilai kelayakan kredit yaitu algoritma Decision Tree C4.5 dan C4.5 yang dioptimasi dengan teknik seleksi fitur Backward Elimination. Menggunakan dataset 481 catatan kredit kendaraan yang diklasifikasikan sebagai "baik" dan "buruk", model ini dikembangkan dengan sebelas variabel independen, termasuk status tanggungan, usia, pendidikan, pekerjaan, pendapatan, dan uang muka. Hasil pengujian menunjukkan bahwa model C4.5 tanpa optimasi mencapai akurasi sebesar 91,90% dengan nilai Area Under Curve (AUC) 0,915. Sebaliknya, model yang mengintegrasikan Backward Elimination menunjukkan peningkatan kinerja yang signifikan, dengan akurasi mencapai 94,80% dan AUC sebesar 0,973. Temuan ini mengkonfirmasi bahwa penerapan optimasi Backward Elimination secara efektif meningkatkan kemampuan prediktif model klasifikasi kelayakan kredit.