Enviro Detect: IOT Integrated DNN Emission Analyzer

Bhukya Raja Kumar 1, M. Bhagya Lakhmi 2, Samakari Sreedhar 3, M. Bhuvanachandra 4,
K. Jaagruthi 5 ,Rachamreddy Dharani 6
1,2,3,4,5,6 Computer Science and Information Technology, Siddharth Institute of Engineering & Technology

Abstract- This project focuses on developing apredictive system for air quality monitoring using machine learning techniques. The goal is to forecast future hazardous gas levels and AQI values based on historical air quality data. The system analyzes pollutants like CO, NO2, Ozone, and PM2.5, which are key factors affecting air quality. By utilizing advanced machine  earning algorithms such as LSTM + GRU Hybrid, CNN-LSTM, and Stacked Bi LSTM with Dropout, the system can predict AQI values and categorize them into categories likeGOOD, Moderate, Unhealthy, and Hazardous.  Additionally, the system integrates Explainable AI (XAI) techniques, specifically SHAP plots, toprovide transparency in the model's decision-making process, enabling users to understand how predictions are made. The front-end of the system is built using HTML, CSS, and JavaScript, while the back-end is powered by Flask, ensuring efficient data processing and model deployment. This project aims to offer accurate predictions for air quality, providing essential information to users about potential healthrisks and empowering them to take proactive measures to reduce exposure to hazardous gases. The system aims to contribute to better environmental management by helping both individuals and authorities monitor and predict air quality effectively. Keywords: AQI, air quality prediction, machine learning, Flask, LSTM, GRU, CNN, SHAP, XAI,environment.