Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal TRILOGI: Jurnal Ilmu Teknologi, Kesehatan, dan Humaniora
Br Tarigan, Febrinata Silvianna
Unknown Affiliation
Humanities Computer Science & IT Engineering Public Health Other

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Tinjauan Literatur Sistematis (2019–2025) Kinerja Decision Tree dan Neural Network (Deep Learning) serta Perbandingannya dengan Naive Bayes dan SVM Syahputra, Fahmy; Sabrina, Elsa; Br Tarigan, Febrinata Silvianna; Sarumaha, Matius Irvan; Rahmadhani, Alfi; Simanjorang, Sandha Calista; Gorat, Loveyanni Marito Benedikta
TRILOGI: Jurnal Ilmu Teknologi, Kesehatan, dan Humaniora Vol 6, No 4 (2025)
Publisher : Universitas Nurul Jadid

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33650/trilogi.v6i4.13429

Abstract

This study presents a Systematic Literature Review (2019–2025) comparing the performance of Decision Tree and Neural Network (Deep Learning) models, alongside their relative performance against Naive Bayes and Support Vector Machine (SVM). The review synthesizes empirical findings across multiple application domains—including healthcare, education, industry, and finance—focusing on commonly reported classification metrics such as accuracy, precision, recall, and F1-score. The synthesis indicates that Decision Trees are frequently preferred for structured/tabular data due to their high interpretability and transparent decision rules, which are valuable for accountable decision-making. In contrast, Neural Networks/Deep Learning tend to outperform on unstructured data (e.g., medical images and text) and complex non-linear patterns, albeit often with reduced explainability. In several studies, Naive Bayes remains competitive as a lightweight baseline, while SVM continues to be effective for high-dimensional feature spaces and specific classification settings. Overall, the review highlights that algorithm selection should be driven by data characteristics, problem complexity, interpretability requirements, and computational constraints, since no single algorithm consistently dominates across all scenarios.
Co-Authors Gorat, Loveyanni Marito Benedikta Rahmadhani, Alfi SABRINA, ELSA Sarumaha, Matius Irvan Simanjorang, Sandha Calista Syahputra, Fahmy