Genai-Powered Diabetes Prediction and Conversational Health Support System
Fuente:
Zenodo
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Recurso digital |
| Langue: | anglais |
| Publié: |
Zenodo
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866901559188652032 |
|---|---|
| author | Ranjith RB Mandapalli Sai Dishitha Reddy Privietha P Preethy Jemims P Yovan Felix A |
| author_facet | Ranjith RB Mandapalli Sai Dishitha Reddy Privietha P Preethy Jemims P Yovan Felix A |
| contents | <p><strong><span>Abstract</span><span>—</span></strong><span lang="EN-IN"> The rising global rate of diabetes has made it crucial to create tools that predict the risk of developing the disease and offer clear, actionable health guidance. In this work, the researcher presents a web-based platform that combines a machine learning prediction engine with an interactive health assistant. Predictive model uses Decision Tree Classifier and is trained on a publicly available dataset of PIMA Indians medical records. It assesses risk based on key health indicators like HbA1c levels, glucose, insulin, BMI, and age. The platform’s interface, developed with Streamlit and ReactJS, focuses on simplicity and accessibility. Scheduled Backend operations, such as updating model and performance tracking, are managed by Cron Jobs. A conversational chatbot, powered by the Gemini generative AI framework, provides personalized, real-time responses to users' health questions. Additionally, a built-in Knowledge Centre offers structured and easy-to-understand resources on diabetes prevention and management. Evaluation results show that the system effectively identifies at-risk individuals early, and the chatbot consistently provides relevant and clear guidance. These features together create a practical and user-friendly tool for early screening and education.</span></p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_18136559 |
| institution | Zenodo |
| language | eng |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Genai-Powered Diabetes Prediction and Conversational Health Support System Ranjith RB Mandapalli Sai Dishitha Reddy Privietha P Preethy Jemims P Yovan Felix A Predictive analysis for diabetes, decision tree–based classification, AI-powered conversational systems, Streamlit framework <p><strong><span>Abstract</span><span>—</span></strong><span lang="EN-IN"> The rising global rate of diabetes has made it crucial to create tools that predict the risk of developing the disease and offer clear, actionable health guidance. In this work, the researcher presents a web-based platform that combines a machine learning prediction engine with an interactive health assistant. Predictive model uses Decision Tree Classifier and is trained on a publicly available dataset of PIMA Indians medical records. It assesses risk based on key health indicators like HbA1c levels, glucose, insulin, BMI, and age. The platform’s interface, developed with Streamlit and ReactJS, focuses on simplicity and accessibility. Scheduled Backend operations, such as updating model and performance tracking, are managed by Cron Jobs. A conversational chatbot, powered by the Gemini generative AI framework, provides personalized, real-time responses to users' health questions. Additionally, a built-in Knowledge Centre offers structured and easy-to-understand resources on diabetes prevention and management. Evaluation results show that the system effectively identifies at-risk individuals early, and the chatbot consistently provides relevant and clear guidance. These features together create a practical and user-friendly tool for early screening and education.</span></p> |
| title | Genai-Powered Diabetes Prediction and Conversational Health Support System |
| topic | Predictive analysis for diabetes, decision tree–based classification, AI-powered conversational systems, Streamlit framework |
| url | https://doi.org/10.5281/zenodo.18136559 |