Data Cleaning:-
- in the quality columns, we are replacing NA with the previous valid occuring value(because the values are super stable)
- in value columns we can initialize the value of previous two fields to zero, then replace NA with the average of the two values.
- take the lat long value of each unique station, and then just replace those where ever the values dont match
Visualizations:-
- the polution distribution with lat long, station-wise in 1 day(hourly breakdown), and within 1 week intervals(at the median of the 1 day dist)
- pollution trends in a month