Delhi Pollution Dataset
Home About Publication Ethical Documentation Sensor Calibration Novelty Dataset
To download PM- dataset, click here...
Download
To download Anomalies - dataset, click here...
Download
To download PM-dataset along with meteorological parameters, click here...
Download

Delhi Pollution Dataset

Summary of dataset

The stastics and plots given below describing the dataset gives high level overview of the properties of the dataset along its variables. The below table provdie basic stastics on the different variables present in this dataset.

1. Description of Nominal variables
Name Type Count Unique Entries Most frequent Least frequent Missing
deviceID string 12542182 13 00000000c37f0aa8 (1393279) 000000008f525c6e (427436) 0%
uid string 12542182 12542182 multiple detected multiple detected 0%
dateTime string 12542182 12542182 2021-01-12 20:42:05 (12) multiple detected 0%
2. Description of Ordinal variables
Name Type Count Unique Entries Most frequent Least frequent Missing
dateTime string 12542182 12542182 2021-01-12 20:42:05 (12) multiple detected 0%
3. Description of Continuous variables
Name Type Count Min Median Max Mean std Missing Zeros
lat number 12542182 28.48655 28.57932 28.72000 28.57855 0.050018 0% 0%
long number 12542182 77.10014 77.24073 77.32076 77.25250 0.03945 0% 0%
pm 1_0 number 12542182 1 110 1730.5 120.34796 57.27231 0% 0%
pm2_5 number 12542182 1 183 1792 207.92479 114.36323 0% 0%
pm 10 number 12542182 1 199 1903 226.11056 123.8647 0% 0%

Histogram and CDF plots for varying PM

Below are the histograms and CDFs for PM1.0, PM2.5 and PM10 respectively. These show how these values were distributed throughout the dataset.

Average PM2.5 maps for each month in dataset

These maps show the average PM2.5 values in each month for the points in our trajectories. Lighter colored points indicate lesser pollution levels while darker colored points indicate mode pollution levels.

1. This shows average PM2.5 values recorded by our sensors for the month of November. November has relatively lesser pollution levels across the city with only a small number of regions showing heavy pollution levels
2. This shows average PM2.5 values recorded by our sensors for the month of December. December shows an increase in the pollution levels with a cluster of really dark points in the center of our area of study.
3. This shows average PM2.5 values recorded by our sensors for the month of January. It shows that January doesn’t really have lower pollution levels. It has it’s own regions where densely located dark points can be found.