ID2223-Lab-1

Air Quality Dashboard - Stockholm St. Eriksgatan 83

Sensor Location: Stockholm, Sweden - St. Eriksgatan 83 Station ID: @10523 Last Updated: Auto-updated daily at 08:00 UTC

7-Day PM2.5 Forecast

This chart shows predicted PM2.5 (fine particulate matter) levels for the next 7-10 days.

Forecast

Model Performance Monitoring

1-Day Hindcast: Predictions vs Actual Measurements

This chart compares our model’s predictions (made 1 day in advance) against actual measured air quality values.

Hindcast

About PM2.5

PM2.5 refers to fine particulate matter with a diameter of less than 2.5 micrometers. These particles can penetrate deep into the lungs and even enter the bloodstream, posing health risks.

Health Guidelines (μg/m³):

0-12: Good
12-35: Moderate
35-55: Unhealthy for sensitive groups
55+: Unhealthy

Technical Details

Model: XGBoost Regression
Features: Temperature, precipitation, wind speed, wind direction
Training Data: Historical air quality + weather data
Update Frequency: Daily predictions at 08:00 UTC
Feature Store: Hopsworks

LLM Integration

There is also a Python program to interact with the air quality ML system using natural language (text or voice), powered by a function-calling LLM.

Built for ID2223 Scalable Machine Learning @ KTH