Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal Infotech: Jurnal Informatika & Teknologi TEKNOSAINS: Jurnal Sains, Teknologi dan Informatika
Tri Prasetyo
Universitas Pamulang, Tangerang Selatan, Banten, Indonesia
Computer Science & IT Control & Systems Engineering Energy Engineering Industrial & Manufacturing Engineering

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Prediksi penyakit ginjal kronis menggunakan algoritma random forest Tri Prasetyo
INFOTECH : Jurnal Informatika & Teknologi Vol 7 No 1 (2026): INFOTECH: Jurnal Informatika & Teknologi
Publisher : LPPMPK - Universitas Muhammadiyah Cileungsi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37373/infotech.v7i1.2095

Abstract

Penyakit ginjal kronis (Chronic Kidney Disease / CKD) merupakan penyakit degeneratif yang memerlukan deteksi dini agar penanganan dapat dilakukan secara lebih efektif. Penelitian ini bertujuan untuk mengembangkan model prediksi CKD menggunakan algoritma Random Forest dengan memanfaatkan dataset publik dari Kaggle yang terdiri dari 400 sampel dan 25 atribut klinis. Tahapan penelitian meliputi preprocessing data, yang mencakup penanganan nilai hilang melalui imputasi, pengodean variabel kategorik menggunakan Label Encoding, serta standardisasi fitur numerik untuk mendukung kestabilan proses pemodelan. Model dilatih dan diuji menggunakan pembagian data sebesar 80:20. Hasil evaluasi menunjukkan bahwa model Random Forest menghasilkan kinerja klasifikasi yang sangat tinggi pada dataset penelitian, dengan nilai akurasi, precision, recall, F1-score, dan AUC yang mencapai nilai maksimum. Meskipun demikian, capaian kinerja tersebut perlu diinterpretasikan secara hati-hati karena ukuran dataset relatif kecil dan berpotensi menimbulkan overfitting, sehingga belum sepenuhnya mencerminkan kinerja model pada kondisi klinis nyata. Sebagai kontribusi ilmiah, penelitian ini menyajikan penerapan Random Forest dengan preprocessing yang terstruktur dan evaluasi multi-metrik untuk prediksi CKD, yang menunjukkan potensi penggunaan machine learning sebagai alat bantu skrining awal berbasis data klinis. Penelitian lanjutan disarankan untuk menerapkan teknik validasi silang serta menggunakan dataset yang lebih besar dan beragam guna meningkatkan kemampuan generalisasi model.
Expert system for diagnosing measles in children based on android Jupron Jupron; Tri Prasetyo; Cendra Harmon; Gigih Amrillah Ibnurhus
TEKNOSAINS : Jurnal Sains, Teknologi dan Informatika Vol 13 No 1 (2026): TEKNOSAINS: Jurnal Sains, Teknologi dan Informatika
Publisher : LPPMPK- Universitas Muhammadiyah Cileungsi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37373/tekno.v13i1.1602

Abstract

Measles is a contagious disease caused by the Morbillivirus, and it can be fatal, especially in children. In Indonesia, cases of measles have significantly increased due to low vaccination coverage and limited access to healthcare services in rural areas. The main problem faced is the lack of early detection of measles by communities that do not have direct access to medical professionals. To address this challenge, an Android-based expert system has been developed that allows users to perform early diagnosis of measles independently based on the symptoms experienced. This system uses the Forward Chaining inference method to determine the possible diagnosis from the combination of symptoms chosen by the user. The system development process is carried out using Agile and Rapid Application Development (RAD) approaches to ensure that development can be conducted quickly and flexibly according to user needs. System testing is conducted using the Blackbox Testing method to evaluate outputs based on specific inputs, and the Whitebox Testing method assesses the internal logic of the system. The test results show that this expert system has an accuracy rate of 92.5% in diagnosing measles symptoms, based on test data that has been validated by medical professionals. In addition to diagnosis functionality, the system is also equipped with features for storing diagnosis history, food recommendations, and education about measles. This application is expected to be an effective solution in raising awareness and knowledge of the public about measles, as well as encouraging them to take medical action promptly to prevent more serious complications, especially in areas with limited health facilities.
Co-Authors Cendra Harmon Gigih Amrillah Ibnurhus Jupron Jupron