Institute of Engineering & Management,
Kolkata
Department of Computer Science
EVS PROJECT REPORT
PROJECT TOPIC :- Air Quality Monitoring with Website &
AI/ML Integration.
CSE , SECTION - C
STUDENT DETAILS:-
NAME ENROLL ROLL NO.
1. Subhajit Samanta . 12023052002182 175
2. Suvrajit Patari. 12023052002188 181
3. Amitanshu. 12023052002173 164
4. Avinash Kumar 12023052002173 167
5. Ashish Kumar Sharma 12023052002215 208
EVS Project Report
EVS Project Report: Air Quality
Monitoring with Website & AI/ML
Integration
1. Introduction
Air Quality Monitoring is a vital activity in today’s world due to the rapid urbanization and
industrialization that significantly affect environmental health.
● Air pollution is responsible for millions of deaths every year globally.
● Monitoring air quality allows governments, scientists, and citizens to take
preventive measures.
● With the advancement of technology, we can now monitor air quality using AI/ML
algorithms and make data accessible through web-based platforms.
2. Objectives of the Project
● To study air pollutants and their effects on health and the environment
● To design a website to present air quality data to users
● To apply AI/ML for forecasting air quality trends
● To promote awareness and real-time action based on pollution levels
3. What is Air Quality Monitoring?
Air Quality Monitoring is the process of measuring pollutants present in the atmosphere to
evaluate the quality of the air in a specific area.
● Common pollutants include: PM2.5, PM10, NO₂, SO₂, CO, O₃
● Measurement tools include: sensors, satellite data, and mobile monitoring stations
● AQI (Air Quality Index) is used to represent the data in an easy-to-understand
format.
4. The Air Quality Index (AQI)
AQI Range Category Impact on Health
0–50 Good Minimal or no risk
51–100 Satisfactory Minor discomfort to sensitive people
101–200 Moderate Breathing discomfort in sensitive groups
201–300 Poor Health effects for all
301–400 Very Poor Serious health risk on prolonged
exposure
401–500 Severe Emergency conditions
5. Sources of Air Pollution
● Vehicular emissions
● Industrial discharges
● Burning of biomass and garbage
● Construction activities
● Dust storms and natural causes
6. Health and Environmental Impact
● Short-term effects: Headache, nausea, throat irritation
● Long-term effects: Lung cancer, cardiovascular disease, neurological disorders
● Environmental damage: Acid rain, crop damage, reduced biodiversity
7. Tools for Monitoring Air Quality
● Government monitoring stations
● Low-cost air sensors (e.g., Arduino + MQ135 sensor)
● Mobile apps and online platforms (e.g., AQICN, BreezoMeter)
● Satellites (e.g., NASA’s MODIS, Sentinel)
8. Building a Website for Air Quality Monitoring
Creating a website is an effective way to present air quality data to the public in real-time.
Tools Required:
● Frontend: HTML, CSS, JavaScript (React or [Link])
● Backend: Python (Flask/Django), [Link]
● Database: MySQL, MongoDB
● APIs: OpenWeatherMap API, AQICN API, or data from local sensors
Features to Include:
● Live AQI display based on location
● Graphical representation of pollutants
● Historical data analysis
● Health suggestions
● Login/register for personalized alerts
How to Build the Website (Step-by-Step):
1. Design the User Interface (UI):
○ Home page with a map showing AQI
○ Dashboard with pollutants data
○ Blog/Info section on health impacts
2. Collect Data:
○ Use APIs or real-time sensor input (e.g., using Arduino + ESP8266)
3. Connect the Frontend and Backend:
○ Backend handles API requests and stores data in the database
4. Deploy the Website:
○ Use platforms like Heroku, Netlify, or Vercel
9. Using AI/ML to Monitor and Predict Air Quality
Why AI/ML?
● AI/ML can identify patterns in historical air quality data
● Predict pollution spikes based on weather, traffic, and industrial activity
● Provide personalized alerts to users based on their sensitivity
ML Algorithms for Air Quality Prediction:
● Linear Regression: For simple AQI forecasting
● Random Forests: For analyzing multiple environmental features
● Neural Networks: For complex, non-linear predictions
● LSTM (Long Short-Term Memory): For time-series prediction of AQI trends
Step-by-Step Implementation:
1. Data Collection:
○ Use APIs or local sensors to gather AQI data over time
2. Preprocessing:
○ Clean and normalize the data
○ Remove outliers and fill missing values
3. Model Training:
○ Use Python libraries like Scikit-learn, TensorFlow, or PyTorch
4. Model Deployment:
○ Integrate the trained model into the website backend
○ Use Flask/Django to provide AQI predictions through a REST API
5. Visualization:
○ Use libraries like [Link] or [Link] to display prediction graphs
10. Sample AI/ML Workflow
1. Collect AQI, weather, and traffic data from an area.
2. Train a Random Forest Regressor on the features: temperature, humidity, wind
speed, etc.
3. Predict AQI for the next 24–48 hours.
4. Display predictions on the website with suggestions like:
○ “Avoid outdoor activity between 2–5 PM”
○ “Use N95 masks today”
11. Real-World Examples of Smart Air Monitoring
● India's SAFAR network for forecasting urban air quality
● OpenSense project using mobile sensors on vehicles
● Google’s Environmental Insights Explorer providing emissions insights for cities
12. Benefits of Combining Website + AI/ML for AQI
Monitoring
● Real-time data availability
● Accessible from anywhere
● Predictive insights for users
● Educational tool for schools and communities
● Data-driven decision-making for local governments
13. Challenges in Air Quality Monitoring
● Cost of sensors and devices
● Calibration and maintenance of equipment
● Data privacy and location tracking
● Public awareness and tech literacy
14. Recommendations and Suggestions
● Use open-source sensors and software to lower costs
● Collaborate with local schools for awareness campaigns
● Promote open data initiatives
● Engage with local municipalities for smarter infrastructure
15. Conclusion
Air pollution is a growing threat, but technology gives us powerful tools to monitor
and fight it. With a smart combination of web development, data science, and AI/ML, we
can not only measure but also predict and act on air quality data. By creating an
interactive website, we bring this power to the people.
Together, we can breathe cleaner air.