Act fast, special offers end soon! Up to $294 is savings when you get Premium today.
Get offer codes

Project Review: Finding the Best Markets to Advertise In - UpdatedVersion

Hi DQ Community

this submission is for the guided project: “Finding the Best Markets to Advertise In”

(project updates: commented codes have been taken out, spelling mistakes have been corrected (at least most of them I hope :open_mouth: ), and itty bitty changes).

link to last page of the project

FindingTheBestMarketsToAdvertiseIn.ipynb (381.5 KB)

Any feedback is crucial and important.

Although if you didnt like it keep that to yourself :smiley: kidding… let me know what exactly is problematic, I will Think over it, decline it, call you names and block You! :sunglasses: kidding again… I will try to improve that area so that I can upload a better version of it on github.

Click here to view the jupyter notebook file in a new tab


It looks like you had a lot of fun with this project! I really enjoyed looking at the variety of graphs (the color palette is nice too). I thought the use of standard deviation to remove the outliers was a really great addition. Your project has inspired me!

There’s a lot of commented-out code in cells. Was there a reason it was kept in the project? Just curious.

Thanks so much for sharing with us!


hey @april.g

Thanks for noticing the commented sections! I left them intentionally for YOU to help me! :grin: :rofl:

Just in case someone looks up and suggests better ways to do the same and figure out my stupidity of doing them. - or I get different and better ideas to work on them later on. While working I was getting all these ideas but then I was delaying this project further and further to the point of dilly-dallying. not so sure for now, but I hope I will add few additions to them. May be use for the 2018 survey data. :crossed_fingers: (long term plan perhaps :stuck_out_tongue: )

Thanks for taking your time and going through the project :heavy_heart_exclamation:

I should take them off before I upload on GitHub right?


I figured it was for something like that. It’s nice to see someone’s thought processes. If the intention is to leave it in for the learning process, or to adjust later, you could always leave a note saying why you ended up abandoning or want to revisit it later. (I always think I’ll remember but I don’t!) I would think it’s best to leave that out for a job portfolio sample so it looks more polished?

1 Like


that is great advice. :+1: Will surely follow it.

And yeah I was actually uploading it on GH but I was like let’s see where all I can improve and then will do it. :wink:

1 Like

This project looks super good @Rucha!! :heart_eyes:

It’s so well polished and professional looking. Nice work on the graphs and the markdown! Definitely agree with @april.g - it’s so visible how much fun you had while working on this project. A very ideal ideal submission suiting your fame as Community Champion and a Community Moderator. :wink:

1 Like

Hey @april.g
I forgot earlier, thank you for all those compliments. :heavy_heart_exclamation:

hey @nityesh
I have no words for your encouragement. This should be okay I guess. BIG Thank You!

1 Like

No comments for the project… … :heart_eyes:


A great job @Rucha. It’s a very professional analysis, I really enjoyed reading it!


hey @Zubair and @artur.sannikov96

thanks a lot :smile: :heavy_heart_exclamation:

Hey @Rucha,
Great project, I just started mines and I found a lot of inspiration in yours.

As a suggestion, I wouldn’t combine horizontal and vertical bar charts together. I would keep the same orientation if are a combination of subplots, is faster for the reader to interpret the chart and as you can see they may overlap.

1 Like

nice, I was looking for a project where someone did a proper frequency table for JobRoleInterest columns (isolating each individual course when they were combined into lists).
before I found yours, I’ve managed to do it this way:

data['JobRoleInterest'] = data['JobRoleInterest'].str.replace(pat=r",\s+", repl=',', regex=True)
data['JobRoleInterest'] = data['JobRoleInterest'].str.lstrip()
data[data['JobRoleInterest'] == 'nan'] = None
data['JobRoleInterest'].str.split(',', expand=True).stack().value_counts(normalize=True, dropna=True)
1 Like