Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Hospital Majapahit : Jurnal Ilmiah Kesehatan Politeknik Kesehatan Mojokerto
Fany Rosita Dewi
Unknown Affiliation
Dentistry Education Medicine & Pharmacology Nursing Public Health Social Sciences Veterinary

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

Found 1 Documents
Search

 1 
Pengembangan Chatbot Deteksi Dini Diabetes Melitus Tipe 2 Berbasis Algoritma Random Forest Fany Rosita Dewi; Widya Puspitasari; Hesti Yusiani
Hospital Majapahit (JURNAL ILMIAH KESEHATAN POLITEKNIK KESEHATAN MOJOKERTO) Vol. 17 No. 2 (2025): HOSPITAL MAJAPAHIT
Publisher : LPPM Sekolah Tinggi Ilmu Kesehatan Majapahit Mojokerto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55316/hm.v17i2.1130

Abstract

Type 2 Diabetes Mellitus (T2DM) stands as one of the leading non-communicable diseases globally and significantly contributes to early mortality due to its complications. Early identification of high-risk individuals is essential in alleviating the burden of this disease. This study aimed to develop an early detection chatbot for T2DM based on machine learning using the Random Forest algorithm integrated into the Telegram platform. The research employed a Research and Development (R&D) approach following the ADDIE model (Analysis, Design, Development, Implementation, Evaluation). The dataset used was the Pima Indians Diabetes dataset, which includes eight key health variables. The model was trained and tested to predict T2DM risk and implemented in an interactive chatbot developed using Python and the Telegram Bot API. The training results indicate that the Random Forest model achieved good performance with an accuracy of 85%, precision of 83%, and recall of 80%, demonstrating the model’s reliable and sensitive capability to detect diabetes risk. The chatbot effectively collects user health data through simple conversational interactions and provides real-time prediction results. System evaluation through User Acceptance Testing (UAT) involving 35 respondents showed a very positive response, with an average score of 4.5 out of 5, indicating that the chatbot is easy to use, informative, and beneficial as an educational and early screening tool for non-communicable diseases. Recommendations for future research include using localized datasets to better represent the Indonesian population and integrating the chatbot with Electronic Medical Records (EMR) systems to enable direct clinical application by healthcare professionals
Co-Authors Hesti Yusiani Widya Puspitasari