Impression : Jurnal Teknologi dan Informasi
Vol. 5 No. 2 (2026): Juli 2026

Klasifikasi Depresi Menggunakan Analisis Algoritma Random Forest Pada Data Kesehatan Mental

Iqbal Giffari Ritonga (Universitas Efarina)
Hartono (Universitas Medan Area)
Syaipuddin Syaipuddin (Universitas Efarina)
Ari Rizkita (Universitas Efarina)



Article Info

Publish Date
31 Jul 2026

Abstract

Depresi merupakan gangguan kesehatan mental yang dapat menurunkan kualitas hidup dan produktivitas, sehingga deteksi dini diperlukan untuk mendukung pencegahan dan penanganan. Penelitian ini bertujuan mengklasifikasikan depresi menggunakan algoritma Random Forest berdasarkan faktor demografis, gaya hidup, dan kesehatan mental. Dataset terdiri dari 502 data dengan 11 atribut yang dibagi menjadi 401 data latih dan 101 data uji menggunakan train-test split 80:20. Kebaruan penelitian terletak pada analisis feature importance untuk mengidentifikasi faktor dominan dalam prediksi depresi. Hasil menunjukkan Random Forest mencapai akurasi 91,09% dengan precision, recall, dan F1-score yang tinggi pada kedua kelas. Academic Pressure menjadi fitur paling dominan, diikuti suicidal thoughts, usia, jam belajar, kepuasan belajar, dan stres finansial. Confusion matrix juga menunjukkan kesalahan klasifikasi yang rendah, dengan dua kasus false negative. Hasil ini menunjukkan bahwa Random Forest berpotensi mendukung pengembangan sistem deteksi dini depresi berbasis machine learning.   Depression is a mental health disorder that can reduce quality of life and productivity, making early detection important for prevention and intervention. This study aims to classify depression using the Random Forest algorithm based on demographic, lifestyle, and mental health factors. The dataset consisted of 502 records with 11 attributes, divided into 401 training and 101 testing data using an 80:20 train-test split. The novelty lies in applying feature importance analysis to identify dominant factors in depression prediction. The results show that Random Forest achieved 91.09% accuracy, with high precision, recall, and F1-score for both classes. Academic Pressure was the most influential feature, followed by suicidal thoughts, age, study hours, learning satisfaction, and financial stress. The confusion matrix also indicated low classification errors, with only two false-negative cases. These findings indicate that Random Forest has potential to support the development of machine learning-based early depression detection systems.

Copyrights © 2026






Journal Info

Abbrev

jti

Publisher

Subject

Automotive Engineering Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Electrical & Electronics Engineering

Description

Impression accepts articles in the fields of Electrical Engineering, Mechanical Engineering, Civil Engineering, Marine Technology Industrial Engineering, Marine Fisheries Technology, Agricultural Technology, Informatics Engineering, Information Systems, Computer, Expert systems, Decision Support ...