Claim Missing Document
Check
Articles

Found 3 Documents
Search

Metode Kernel Distance Classifier Terhadap Klasifikasi Penyakit Jantung Aprianto, Kasiful
JITCE (Journal of Information Technology and Computer Engineering) Vol. 5 No. 02 (2021)
Publisher : Universitas Andalas

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25077/jitce.5.02.70-74.2021

Abstract

This study compares the Support Vector Machine (SVM) and Kernel Distance Classification (KDC) methods to classify heart disease. SVM works by transforming data into higher dimensions using the kernel and classifying data linearly using a hyperplane. Meanwhile, KDC works by finding points that represent each classification from the data that has been transformed into a higher dimension using the kernel, and the new data is predicted based on the closest distance from the point of each classification. The results show that the accuracy produced by SVM is 81.11%. The accuracy produced by the SVM model is better than that produced by the KDC model of 80.47% with a difference of 0.64%, even though both models use kernel transformation.
Determinants of User Acceptance of the Halodoc Application: An Analysis of User Experience and User Satisfaction Aprianto, Kasiful; Ibrahim, Andi
Journal Medical Informatics Technology Volume 3 No. 2, June 2025
Publisher : SAFE-Network

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37034/medinftech.v3i2.41

Abstract

Halodoc is one of the leading mobile health (mHealth) applications in Indonesia, offering services such as online doctor consultations, medicine delivery, and health information. This study examines the factors influencing user acceptance of the Halodoc app, focusing on the roles of user experience and satisfaction. The research involved a survey of 81 Halodoc users, followed by validity and reliability testing of the research instruments. Results showed that most items had high validity, with correlation values ranging from 0.775 to 0.851 for user acceptance, and above 0.75 for user experience (except one item). Reliability was also high, with Cronbach’s Alpha values exceeding 0.8 across categories. The highest average score was found in user satisfaction (21.77), indicating consistently high levels of satisfaction. Significant correlations were observed among user acceptance, user experience, service quality, and user satisfaction—most notably between user acceptance and satisfaction (0.8314). Regression analysis identified user experience and satisfaction as significant predictors of user acceptance, accounting for 74.4% of the variance. In contrast, service quality did not show a significant effect. The final regression model after stepwise elimination confirmed the strong influence of user experience (coefficient = 0.3513) and satisfaction (coefficient = 0.4399). These findings highlight the importance of enhancing user experience and satisfaction to increase user acceptance of mHealth applications like Halodoc.
Classifying Heart Disease through Fusion of Multi-Source Datasets: Integration of Feature Selection and Explainable Machine Learning Techniques Aprianto, Kasiful; Anasanti, Mila Desi
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 19, No 3 (2025): July
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.92395

Abstract

This study delves into heart disease classification through integrated feature selection and machine learning methodologies, utilizing three datasets comprising 4,728 participants and 11 features, with 4.27% missing data. Employing machine learning, we used XGBoost to achieve 0.95 accuracy for one feature, while Random Forest (RF) demonstrated accuracies of 0.92 and 0.99 for the remaining two features. Comparing 11 classification models, RF and XGBoost classified heart disease with 0.97 and 0.99 accuracy, respectively, using all available features. Applying Feature Elimination with Simultaneous Perturbation Feature Selection and Ranking (SpFSR) revealed that RF attained 0.99 accuracy by selecting only four features (cholesterol level, age, resting electrocardiographic measurements, and maximum heart rate), while XGBoost dropped to 0.91. Constructing an RF model with four features enhanced interpretability without compromising accuracy. Explainable Machine Learning (XAI) techniques, including Permutation Importance and SHAP Summary Plot analyses, gauged feature impact on heart disease prediction. The resting electrocardiographic measurements feature held the highest value (0.40 ± 0.01), followed by maximum heart rate (0.32 ± 0.01), cholesterol level (0.28 ± 0.01), and age (0.26 ± 0.005). These results underscore the significance of each feature in diagnosing heart disease via machine learning.