A REAL-TIME ANOMALY DETECTION SYSTEM FOR TEMPERATURE MONITORING USING ML

This paper presents a Real-Time Temperature Anomaly Detection and Monitoring System using machinelearning. The system is developed using Python, FastAPI, and Streamlit. A sensor simulator generates temperaturedata. The data is sent to the backend API built using FastAPI. An Isolation Forest model detects anomalies fromthe temperature readings. The detected results are displayed using a real-time dashboard developed with Streamlit.This project simulates an edge computing environment where data is processed locally. The system is suitable forapplications like smart agriculture and industrial temperature monitoring.