Embedded Files
Collection
Collection
To collect our data, we visited several parking garages and measured air quality at both the ground level and the top level of each structure. At every location, we recorded measurements such as particulate matter (PM1, PM2.5, and PM10), temperature, humidity, and other environmental data. We also measured the height of each parking garage so we could compare how elevation affected air quality. By collecting data from multiple locations and heights, we were able to analyze how air pollution changes with altitude and different environments.
AIR QUALITY REFLECTION
Which location had the best air quality? Explain your reasoning.
The top of the parking garages generally had the best air quality because they had lower levels of pollutants like CO₂ and particulate matter. This is likely because there was better airflow and ventilation at higher elevations.
Which location had the worst air quality?
The ground level of the parking garages had the worst air quality because pollutants from vehicles accumulated there, resulting in higher CO₂ and particulate matter levels.
Which variables appeared to be related?
PM1, PM2.5, and PM10 were strongly related since they all measure different sizes of particulate matter. Temperature and humidity also showed some relationship with air quality measurements.
Why did you choose your prediction target?
We chose PM2.5 as our prediction target because it is one of the most important indicators of air quality and can have significant impacts on human health.
Which measurements did you use as inputs for the machine learning model? Why?
We used temperature, humidity, PM1, and PM10 as inputs because these variables showed relationships with PM2.5 and helped improve the accuracy of the prediction.
How well did the model perform?
The model performed very well because the predicted PM2.5 values closely matched the actual measured values, showing that it was able to make accurate predictions.
What additional data would improve the model?
Collecting more data from different locations, weather conditions, and times of day would improve the model. Additional measurements such as wind speed, traffic volume, and precipitation could also make the predictions more accurate.
Data on matlab
Data on matlab
We imported the air quality data collected during our mobile survey and stationary monitoring into MATLAB for analysis. MATLAB allowed us to organize the data, remove errors, and combine measurements such as temperature, humidity, particulate matter (PM1, PM2.5, and PM10), CO₂, air pressure, and GPS location into one dataset. We then used MATLAB to create graphs, maps, and statistical models that helped us identify trends, compare locations, and better understand how different environmental factors relate to air quality. This made it much easier to interpret the data and communicate our findings.
Page updated
Google Sites
Report abuse