Skip to content

KavyaSoni123/AQI-predictor-model

Repository files navigation

AQI Prediction Model

This project focuses on predicting the Air Quality Index (AQI) using various machine learning techniques. The main goal is to develop a robust model that accurately predicts AQI values based on historical data, helping users understand air quality patterns and take necessary precautions.

Table of Contents

Project Overview

Air quality has a significant impact on public health, and predicting AQI can help communities prepare for hazardous conditions. This project implements a machine learning pipeline to predict AQI using various data preprocessing and modeling techniques.

Features

  • Data preprocessing and handling missing values using advanced imputation techniques.
  • Data visualization using Matplotlib and Seaborn.
  • Implementation of multiple machine learning models, including Random Forest, Gradient Boosting, and XGBoost.
  • Selection of the best performing model based on evaluation metrics.

Data Preprocessing

The data preprocessing steps include:

  • Handling Missing Values: Missing values in the dataset were filled using the MICE (Multiple Imputation by Chained Equations) technique. This was implemented using the Iterative Imputer from scikit-learn with Ridge, Elastic Net, and Bayesian Regression as estimators.

  • Data Visualization: Data insights and visual patterns were explored using Pandas, Numpy, Matplotlib, and Seaborn.

Machine Learning Models

Several machine learning models were tested to find the best performing one:

  • Random Forest (Selected Model): Performed best among all models.
  • XGBoost
  • Gradient Boosting

The Random Forest model was selected based on evaluation metrics such as accuracy, mean squared error, and R² score.

Installation

Clone the repository and install the required dependencies.

git clone https://github.com/KavyaSoni123/AQI-predictor-model.git
cd aqi-prediction-model
pip install -r requirements.txt

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages