Datathons Solutions Datathon – Sofia Air 2.0 – Solution – Team Chameleons Posted 14.04.201916.05.2020 vrategov, caseyp, kali Mentors: ekaterinamarina, paspaldzhiev, pk1k 3votes Article (1) Share this
Great job. I would suggest you to add a legend to the maps. For the stage 2 (construction sites) you could try to take into account the stage of the construction sites based on its starting and ending date.
I would agree with pepe’s comment for state of construction sites.
Otherwise, using color overlay with “blurring” over map for pollution spreading is quite effective way to present results. I will repeat one of my comments to other team “It would be even better (I know you didn’t had time) to take into account surounding buildings and how they impact pollution spread to get more custom “shape” of spread instead of combination of circles” …. I would use custome scale for coloring within “heat map” instead of default ones because something which is bright red doesn’t necessarily means we have so much larger polution effect than in yellow/green zones. It is just default color scale which spans from minimal to maximum available measured value. Applying custom color scale which coresponds with some official polution scale rather that min/max value combination would tell story even better and visually showed all clues where you need to focus for and “smart-city” initiative.
Thanks for the feedbacl. In regards to the state of construction, I think this was taken care in the provided dataset. Every construction type has a default length which they pollute the air for. For example, “small housing” is assumed to affect the air in the next n months after starting. We were told that all examples which exceeded this length in comparison to our sample period were filtered out from the construction sites dataset.
Thanks for the feedback. I completely aggree in regards to the color scale.