The Plant Health Monitoring Web Application using Machine Learning

Abstract

Plant diseases and unfavorable environmental conditions pose significant challenges to agricultural productivity, often remaining undetected until severe damage occurs. This paper presents a full-stack web application designed to monitor plant health and provide intelligent crop recommendations based on environmental conditions such as temperature, humidity, sunlight, and watering frequency. The system utilizes a Python-based machine learning backend that trains and evaluates three supervised classification models: Decision Tree, Random Forest, and Logistic Regression. The best-performing model is selected and deployed for real-time prediction through a REST API. The frontend is implemented as a responsive multi-page web interface that enables users to perform plant recommendation, health prediction, and leaf image-based disease detection. Experimental results demonstrate that the Decision Tree model achieves the highest accuracy, making it suitable for deployment. The system offers a practical, accessible, and efficient solution for plant health monitoring and agricultural decision support.

Code and data availability

The paper describes a plant health monitoring web application but provides no public dataset, code, model, or supplement of its own. The only mentioned dataset (PlantVillage) is cited prior work, and the structured CSV dataset used for training is not deposited or linked.

No evidence-backed public reproduction asset is currently recorded.