EBAY (Kleinanzeigen) Used Cars Analysis - Feedback

I would appreciate if you have the time to take a look at my project and share with me your feedback and how i can be better


Files on Github https://github.com/KarimYOmar/Ebay-Used-Car-Sales-Analysis

Basics.ipynb (170.0 KB)

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


Hi @karimyousrymohamedom,

You have gone beyond the general instructions and took up the extra instructions as well which is a great thing. I just felt that you could have used the markdown cells more and added more explanations to make your project more readable and understandable for not so technical people as well. Ending the project with a proper conclusion is also important. At times people might only be interested in the conclusions.

You have converted the date columns but didn’t use it for any purpose. Maybe you can experiment with them to find some insights.

Overall you did a good job. By adding a conclusion to your work, it will become more useful. Hope this helps.

1 Like

Hi Jithins,

Thanks for your feedback, i would appreciate if you can check my other project, i tried following your notes from the previous project and apply it here, appreciate if you can share your feedback and how i can be better


1 Like

Hi @karimyousrymohamedom

could you please share the link of the thread where you have shared?
Or the notebook file itself.

The above link is taking me to DQ mission screen.

My apologiesBasics.ipynb (667.0 KB)

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

Everything looks very clean in the first look itself, which I think is great.

Maybe you can try this to create a table, if you want to try something new for listing down the column name description.

| Syntax      | Description |
| ----------- | ----------- |
| Header      | Title       |
| Paragraph   | Text        |
  • For most of the scatter plots, it was difficult to find a correlation, I agree. But if you look at the initial values, you can see something strange. Both high and low values are concentrated there. Maybe you can try to think of a reason for that.

  • You had titles for all the scatter plots, but for histogram, there are no labels and titles. So the plot itself is not self explanatory. You could have added it there too.

  • For both scatter plot and histogram, you can actually go further into the values, sort of like zooming in by setting shorter range of x values. This will in turn give you more insights.

  • When you get to the scatter matrix plot, you can probably feel that maybe the initial spike is due to the influence of smaller sample sizes. Though you are right, no big correlation can be found there.

  • I also did the bar graph like you did, ie vertical bar graph. But I have seen someone doing a horizontal bar graph. So a horizontal bar graph makes it easier to read all the names of the majors.

  • For this part below where you have to combine the data…

Comparing the number of men with the number of women in each category of majors

you can look up for groupby() function and play around with it till you achieve the same result. If you use groupby() the whole calculation under this section can be done with a single line of code. If you get time, please try it.

Like I said earlier, the whole project looks very neat. Conclusions are good. But if you spend more time on this, you can find more conclusion. Also, it would be great if you can use a few subheadings in between instead of bullet points.

I hope this helps. Happy learning.

Hi @jithins123,

I really appreciate you taking the time to share with me your feedback, it`s really motivating.

I actually do not take much time in the projects to do further analysis as i am short on time and studying during a full-time job, but i am saving your comments so when i finish the career path, i`ll visit the projects again and apply your comments, in the meantime i will try to implement it in the coming projects.

It would be great if you can share with me your projects to learn more.


1 Like

Hi @karimyousrymohamedom

There you go. Please let me know your feedback.