Perbandingan Kinerja Akurasi Model Mesin Learning Untuk Prediksi Penyakit Jantung

Juyus Muhammad Adinulhaq(1*), Muhammad Sam'an(2)


(1) Program Studi S1 Informatika Universitas Muhammadiyah Semarang
(2) Program Studi Informatika, Universitas Muhammadiyah Semarang
(*) Corresponding Author

Abstract


This research aims to comprehensively analyze heart disease-related data through Exploratory Data Analysis (EDA), identification of correlations between numerical variables, and cluster analysis to uncover patterns in the data. Furthermore, using various machine learning algorithms, such as Logistic Regression, Support Vector Classifier, Decision Tree Classifier, Random Forest Classifier, K-Nearest Neighbors, and Gaussian Naive Bayes, a heart disease prediction model was built. The model evaluation shows that Naive Bayes has the highest test accuracy of 90%, followed by RandomForestClassifier and KNeighborsClassifier which have 85% test accuracy. These findings indicate a good ability to predict heart disease, but further analysis is needed to ensure good generalization to unseen data. This research makes an important contribution to the development of heart disease prediction models and can support early detection and appropriate intervention strategies.

Full Text:

PDF

Article Metrics

Abstract view : 147 times
PDF - 32 times

DOI: https://doi.org/10.26714/jkti.v1i2.12918

Refbacks

  • There are currently no refbacks.


=======================================================================================

Penerbit:

  • JKTI | Jurnal Komputer dan Teknologi Informasi
  • Program Studi S1 Informatika,  Unimus| Universitas Muhammadiyah Semarang
  • Sekretariat: Gedung Kuliah Bersama II (GKB II) Lantai 7,  Jl. Kedungmundu Raya No 18 Semarang
  • email: informatika@unimus.ac.id, Phone: + +62 813 2504 3677
  • e-ISSN: 2986-7592

Paper Template: Download

View My Stats

------------------------------------------------------------------------------------------------------------------------------------------------------