Scribd
Upload
35 views12 pages

Air Quality Forecasting Using Machine Learning

Uploaded by

Senthil Sk
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
35 views12 pages

Air Quality Forecasting Using Machine Learning

Uploaded by

Senthil Sk
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 12

AIR QUALITY FORECASTING

USING MACHINE LEARNING

MENTOR Team Members


Ms.S Lavanya 1.Senthil Kumar S
2.Anandhaselvam M
INTRODUCTION
● A cutting-edge approach that leverages advanced algorithms and data analysis
techniques to predict the levels of air pollutants and air quality in a given area.

● Rapid population growth and economic development in developing countries,


like India, have led to various environmental problems.

● Recent studies have linked exposure to atmospheric pollutants to adverse


health effects, including asthma, lung inflammation, and approximately 30
million asthma-related deaths, especially among children.

● Particulate matter (PM2.5 or PM10) in the air is associated with various


respiratory, cardiovascular, and blood diseases.
INTRODUCTION
● Over the past decade, there have been significant advancements in air quality
monitoring and forecasting techniques, focusing on real-time monitoring and
statistical models based on historical data.

● In recent years, machine learning techniques have emerged as strong


alternatives to traditional statistical models in the field of forecasting.

● These models, often referred to as black-box or data-driven models, are


nonparametric and nonlinear, relying solely on historical data to learn
stochastic dependencies between past and future events.

● Overall, ML is a powerful tool for building AI applications that can learn from
data and improve their performance over time.
OBJECTIVES

● The main goal is to forecast air quality levels, including concentrations of


pollutants like O3, CO2, NO2, PM2.5 and others.

● Continuous refinement and improvement of machine learning models can lead


to more accurate and reliable air quality forecasts over time.
LITERATURE SURVEY

S.NO AUTHOR TITLE CONFERENCE INFERENCE

1. Liu, Zhe et "Short-term air International Conference on Investigated the use of deep
al. quality forecasting Machine Learning (ICML) learning methods like CNN and
using deep LSTM for short-term air
learning methods." quality prediction.

2 Pratik "DeepAR: IEEE International Conference Utilized a deep autoregressive


Dubal, Probabilistic on Data Mining (ICDM) model (DeepAR) to predict air
Sankirtan forecasting with quality with probabilistic
Bhatt autoregressive forecasts.
recurrent networks
5

3 Mehwish Dildar Air quality International Conference on Reviewed various machine


1 , Shumaila
Akram prediction using Artificial Intelligence and learning techniques for air
machine learning Applications (AIAPP) quality prediction, highlighting
techniques their pros and cons.
LITERATURE SURVEY
S.NO AUTHOR TITLE CONFERENCE INFERENCE

4 LISHEN A hybrid model based on IEEE International Proposed a hybrid model combining
G WEI LSTM and SVR for daily Conference on Data LSTM and SVR for accurate daily
1 , KUN PM2.5 concentration Science and Advanced PM2.5 concentration forecasting.
DING prediction. Analytics (DSAA)

5 Josef Short-term air quality International Conference Investigated the use of deep learning
Steppan, forecasting using deep on Pattern Recognition methods like CNN and LSTM for
Sten learning methods. (ICPR) short-term air quality prediction.
Hanke

6 Sumaiya A review of air quality International Conference Provided a comprehensive review of


Dabeer, prediction models. on Environmental various air quality prediction models,
Maha Science and Technology including machine
6 learning-based
Mohamm (ICEST) approaches.
ed Khan,
LITERATURE SURVEY
S.NO AUTHO TITLE CONFERENCE INFERENCE
R

7 Andre An integrated International Conference on Computer Developed an integrated approach


Esteva1 machine Science and Applications (ICCSA). using machine learning models to
*, Brett learning predict short-term air quality
Kuprel1* approach for effectively.
short-term air
quality
prediction

8 Md Long short- IEEE International Conference on Big Demonstrated the application of


Shahin term memory Data (BigData) LSTM networks to predict air quality
Ali a , neural using large-scale datasets.
Md network for 7

Sipon air quality


Miah prediction
using big data
EXISTING SYSTEM

• The existing system typically involves a combination of data


collection, preprocessing, feature engineering, model training, and
evaluation.

• Provide accurate and timely information to help indivuals regarding air


quality and its potential health impacts.

• Uses MLR and supervised machine learning for real-time AQI


prediction, with rigorous performance evaluation.

• Employed ARIMA time series modeling for future AQI forecasts,


achieving high accuracy and efficiency.
PROPSED WORK

• Used FB Prophet, a significant forecasting model, specializes in time


series data forecasting.

• Its strength lies in its ability to handle stationarity within the data,
making it a powerful and versatile library for time series forecasting.

• Prophet supports custom seasonalities where we can define specific


seasonal patterns that are relevant to dataset, such as monthly or
quarterly patterns.

• It simplifies the process of time series forecasting.


REQUIREMENT ANALYSIS

HARDWARE REQUIREMENTS:
● Processor (CPU)
● Video Card (GPU)
● Memory (RAM)
● Storage (Drives)
SOFTWARE REQUIREMENTS:
● Datasets
● Development tool - Notebook - Google Collab

10
PLANNING
THANK YOU

You might also like