Garuda - Garba Rujukan Digital

Robust Predictive Model for Heart Disease Diagnosis Using Advanced Machine Learning Techniques Sovia, Rini; Anam, M. Khairul; Wisky, Irzal Arief; Permana, Randy; Rahmi, Nadya Alinda; Zain, Ruri Hartika
Journal of Applied Data Sciences Vol 7, No 1: January 2026
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v7i1.1092

This study presents a hybrid ensemble learning framework designed to enhance the predictive accuracy, robustness, and generalizability of heart disease classification models. The framework integrates three base classifiers: Decision Tree (DT), Gaussian Naive Bayes (GNB), and K Nearest Neighbor (KNN), which are combined using a stacking ensemble method with Logistic Regression (LR) as the meta learner. Each classifier contributes a distinct analytical perspective: DT models nonlinear relationships, GNB provides probabilistic reasoning, and KNN captures similarity-based patterns. Logistic Regression aggregates their outputs to produce a unified predictive decision. To mitigate class imbalance commonly observed in clinical datasets, the Synthetic Minority Oversampling Technique (SMOTE) is applied to generate synthetic samples of the minority class, improving the model’s ability to recognize underrepresented cases. Hyperparameter optimization is performed using the Optuna framework, which applies the algorithm to efficiently explore parameter configurations. The proposed model was evaluated on a publicly available heart disease dataset and achieved an accuracy of 99.61%, precision of 99.62%, recall of 99.59%, F1 score of 99.60%, and specificity of 99.58%, corresponding to a false positive rate of only 0.42 percent. These results demonstrate the framework’s strong ability to accurately identify heart disease cases while minimizing misclassification. The integration of SMOTE, stacking, and Optuna optimization contributes to its superior performance and robustness. Consequently, this approach shows strong potential for integration into clinical decision support systems to assist healthcare professionals in reliable and timely diagnosis.

Machine Learning Predicts the Level of Disease Spread Dhio Saputra; Irzal Arief Wisky; Sarjon Defit
Jurnal Penelitian Pendidikan IPA Vol 10 No 4 (2024): April
Publisher : Postgraduate, University of Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29303/jppipa.v10i4.7070

The aim of the research is predictive analysis of the spread of disease. Variable analysis at the population level in a region and the total disease events detected in the community. These variables can show the accuracy and certainty of the status of the resulting analysis. The concept of Machine Learning analysis is proposed to develop previous analysis models. The methods used include the K-Means cluster, Naïve Bayes, and Decision Tree (DT). There are two stages in the analysis process: pre-processing and classification. The discussion presented by K-Means provides a classification analysis pattern. The patterns obtained will be passed on to the classification process using Naïve Bayes and DT. Naïve Bayes results provide quite significant results with an accuracy rate of 83.33%. DT can also describe the results of information and knowledge analysis in the form of decision trees. DT produces decision trees that can provide knowledge and information analysis. The DT results provide an accuracy rate of 91.76% so these results can be used as consideration in decision making. The resulting information and knowledge can be used as a guide in making policies for handling health in the community.

Analisis Kepuasan Pelayanan Publik Menggunakan Metode Naïve Bayes Pada Dinas Kependudukan Dan Pencatatan Sipil Kabupaten Agam Irzal Arief Wisky; Eva Rianti; Afika Syahira
Jurnal Sains Informatika Terapan Vol. 4 No. 3 (2025): Jurnal Sains Informatika Terapan (Oktober, 2025)
Publisher : Riset Sinergi Indonesia (RISINDO)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62357/jsit.v4i3.784

This study aims to analyze public satisfaction with services provided by the Department of Population and Civil Registration (Disdukcapil) of Agam Regency using the Naïve Bayes method. The research data were collected from 252 respondents through a public satisfaction survey covering nine service indicators based on the Ministry of Administrative Reform Regulation No.16 of 2014. The Naïve Bayes algorithm was applied to classify satisfaction levels into four categories: very dissatisfied, dissatisfied, satisfied, and very satisfied. The results indicate that the developed web-based system can accurately predict public satisfaction levels, with the highest probability value of 0.1127 falling under the “very satisfied” category. These findings demonstrate that the service quality at Disdukcapil Agam Regency has been well implemented, and the application of the Naïve Bayes method is effective in supporting the evaluation and continuous improvement of public service performance.

Implementasi Metode K-Means Clustering Untuk Pengelompokan Pasien Berdasarkan Tipe Penyakit Di Dr. Drs. M. Hatta Brain Hospital, Bukittinggi Irzal Arief Wisky; Ridha Utami
Jurnal Teknik dan Teknologi Tepat Guna Vol. 4 No. 3 (2026): Jurnal Teknik dan Teknologi Tepat Guna
Publisher : Riset Sinergi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62357/j-t3g.v4i3.1357

Volume data pasien yang terus meningkat di rumah sakit membutuhkan manajemen dan analisis data yang efisien untuk mendukung pengambilan keputusan dalam layanan kesehatan. Studi ini bertujuan untuk mengimplementasikan algoritma K-Means Clustering dalam sistem informasi berbasis web untuk mengelompokkan pasien berdasarkan karakteristik penyakit di Rumah Sakit Otak Dr. Drs. M. Hatta (RSOMH) Bukittinggi. Penelitian ini menggunakan pendekatan penelitian dan pengembangan yang melibatkan pengumpulan data, analisis sistem, desain sistem, implementasi, dan pengujian. Aplikasi ini dikembangkan menggunakan PHP dan MySQL, sementara algoritma K-Means digunakan untuk mengklasifikasikan data pasien berdasarkan kesamaan atribut klinis. Sistem secara otomatis melakukan klasterisasi dengan menghitung jarak Euclidean dan memperbarui sentroid klaster secara iteratif hingga konvergensi tercapai. Hasil implementasi menunjukkan bahwa sistem yang diusulkan berhasil mengelompokkan pasien ke dalam klaster yang bermakna, memudahkan analisis data pasien yang lebih cepat dan akurat dibandingkan metode konvensional. Selain itu, sistem ini menyediakan informasi berguna untuk manajemen rumah sakit dalam memantau pola penyakit, mengoptimalkan alokasi sumber daya, dan mendukung pengambilan keputusan layanan kesehatan berbasis datan.

Penerapan Data Mining Untuk Pengembangan Destinasi Wisata Menggunakan Algoritma K-Means Clustering Irzal Arief Wisky; Febria Nika Putri; Mutiana Pratiwi
Jurnal Sains Informatika Terapan Vol. 5 No. 2 (2026): Jurnal Sains Informatika Terapan (Juni, 2026)
Publisher : Riset Sinergi Indonesia (RISINDO)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62357/jsit.v5i2.1337

One of the strategic industries that supports community wellbeing and regional economic development is tourism. In West Sumatra's Solok Regency, Alahan Panjang is home to a wide range of natural tourist sites, each with unique characteristics and possibilities for growth. Due to differences in the number of tourists, attractions, accessibility, and amenities, it is still difficult to identify development priorities for each destination. Based on their similarities, this study uses the K-Means Clustering algorithm and the Data Mining approach to categorize tourist attractions. The study uses tourist destination data from Alahan Panjang that includes variables for tourism components and visitor statistics. To ensure a consistent range of values, the data were subjected to Min-Max normalization before clustering. Three types of tourist sites are distinguished by the clustering findings: high-priority, priority, and supporting locations. Additionally, an information system built on PHP and MySQL integrated the clustering results to assist tourism managers and municipal governments in assessing tourist potential and establishing development goals. Through data-driven and objective tourism analysis, the suggested method is anticipated to increase the effectiveness of decision-making.

Penerapan Data Mining Untuk Pengembangan Destinasi Wisata Menggunakan Algoritma K-Means Clustering Irzal Arief Wisky; Febria Nika Putri; Mutiana Pratiwi
Jurnal Sains Informatika Terapan Vol. 5 No. 2 (2026): Jurnal Sains Informatika Terapan (Juni, 2026)
Publisher : Riset Sinergi Indonesia (RISINDO)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62357/jsit.v5i2.1337

One of the strategic industries that supports community wellbeing and regional economic development is tourism. In West Sumatra's Solok Regency, Alahan Panjang is home to a wide range of natural tourist sites, each with unique characteristics and possibilities for growth. Due to differences in the number of tourists, attractions, accessibility, and amenities, it is still difficult to identify development priorities for each destination. Based on their similarities, this study uses the K-Means Clustering algorithm and the Data Mining approach to categorize tourist attractions. The study uses tourist destination data from Alahan Panjang that includes variables for tourism components and visitor statistics. To ensure a consistent range of values, the data were subjected to Min-Max normalization before clustering. Three types of tourist sites are distinguished by the clustering findings: high-priority, priority, and supporting locations. Additionally, an information system built on PHP and MySQL integrated the clustering results to assist tourism managers and municipal governments in assessing tourist potential and establishing development goals. Through data-driven and objective tourism analysis, the suggested method is anticipated to increase the effectiveness of decision-making.

Implementasi Metode K-Means Clustering Untuk Pengelompokan Pasien Berdasarkan Tipe Penyakit Di Dr. Drs. M. Hatta Brain Hospital, Bukittinggi Irzal Arief Wisky; Ridha Utami
Jurnal Teknik dan Teknologi Tepat Guna Vol. 4 No. 3 (2026): Jurnal Teknik dan Teknologi Tepat Guna
Publisher : Riset Sinergi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62357/j-t3g.v4i3.1357

Volume data pasien yang terus meningkat di rumah sakit membutuhkan manajemen dan analisis data yang efisien untuk mendukung pengambilan keputusan dalam layanan kesehatan. Studi ini bertujuan untuk mengimplementasikan algoritma K-Means Clustering dalam sistem informasi berbasis web untuk mengelompokkan pasien berdasarkan karakteristik penyakit di Rumah Sakit Otak Dr. Drs. M. Hatta (RSOMH) Bukittinggi. Penelitian ini menggunakan pendekatan penelitian dan pengembangan yang melibatkan pengumpulan data, analisis sistem, desain sistem, implementasi, dan pengujian. Aplikasi ini dikembangkan menggunakan PHP dan MySQL, sementara algoritma K-Means digunakan untuk mengklasifikasikan data pasien berdasarkan kesamaan atribut klinis. Sistem secara otomatis melakukan klasterisasi dengan menghitung jarak Euclidean dan memperbarui sentroid klaster secara iteratif hingga konvergensi tercapai. Hasil implementasi menunjukkan bahwa sistem yang diusulkan berhasil mengelompokkan pasien ke dalam klaster yang bermakna, memudahkan analisis data pasien yang lebih cepat dan akurat dibandingkan metode konvensional. Selain itu, sistem ini menyediakan informasi berguna untuk manajemen rumah sakit dalam memantau pola penyakit, mengoptimalkan alokasi sumber daya, dan mendukung pengambilan keputusan layanan kesehatan berbasis datan.

Robust Predictive Model for Heart Disease Diagnosis Using Advanced Machine Learning Techniques Rini Sovia; M. Khairul Anam; Irzal Arief Wisky; Randy Permana; Nadya Alinda Rahmi; Ruri Hartika Zain
Journal of Applied Data Sciences Vol 7, No 1: January 2026
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v7i1.1092

This study presents a hybrid ensemble learning framework designed to enhance the predictive accuracy, robustness, and generalizability of heart disease classification models. The framework integrates three base classifiers: Decision Tree (DT), Gaussian Naive Bayes (GNB), and K Nearest Neighbor (KNN), which are combined using a stacking ensemble method with Logistic Regression (LR) as the meta learner. Each classifier contributes a distinct analytical perspective: DT models nonlinear relationships, GNB provides probabilistic reasoning, and KNN captures similarity-based patterns. Logistic Regression aggregates their outputs to produce a unified predictive decision. To mitigate class imbalance commonly observed in clinical datasets, the Synthetic Minority Oversampling Technique (SMOTE) is applied to generate synthetic samples of the minority class, improving the model’s ability to recognize underrepresented cases. Hyperparameter optimization is performed using the Optuna framework, which applies the algorithm to efficiently explore parameter configurations. The proposed model was evaluated on a publicly available heart disease dataset and achieved an accuracy of 99.61%, precision of 99.62%, recall of 99.59%, F1 score of 99.60%, and specificity of 99.58%, corresponding to a false positive rate of only 0.42 percent. These results demonstrate the framework’s strong ability to accurately identify heart disease cases while minimizing misclassification. The integration of SMOTE, stacking, and Optuna optimization contributes to its superior performance and robustness. Consequently, this approach shows strong potential for integration into clinical decision support systems to assist healthcare professionals in reliable and timely diagnosis.

Title

Found 28 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 28 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 28 Documents
Search