I'd like some feedback on my take of Fandango Movie Ratings Guided Project


I have recently completed this project and I tried to explain the things I’ve done as thoroughly as possible. I have decided to focus mostly on explaining what I’ve done in this project, so I could use some feedback specifically on the markdowns.

Here is the last page of the project.

You can find my notebook here.

Thanks in advance! :smiling_face_with_three_hearts:


I like your work. That’s nice.

1 Like

Hi Dilara,

Nice work! Here is some specific feedback:

  • I think your introduction was strong and engaging. You do a good job summarizing Hickey’s findings and what you are going to cover.

  • Sampling and describing data - you make it clear what data you are including and your assumptions. One thing I might mention is that Hickey’s data and the 2016 data were selected with somewhat different criteria, and neither data set was selected with random sampling. So, we can’t really prove causation here. We can say that there was a difference in 2015 vs. 2016 popular ratings on Fandango. That difference may be because Fandango ‘fixed’ the bug after Hickey’s report, but we can’t actually prove this to be the case or not.

  • Plotting - you can use plt.show() at the end of your code and it will just display the graph, and hide all the info on the xticks. I also just finished this project and used https://www.dataquest.io/blog/making-538-plots/ to play around with tweaking all these little details on the plots.

  • Frequency table - you can use Series.value_counts(normalize=True) to get this table without having to do all the backend work.




Thanks a lot for the feedback! I’m pretty busy nowadays but I’ll check these out and see if I should make changes. I appreciate it. :smiling_face_with_three_hearts:



I think content wise it’s all fine as far as I can see.
I’d only improve some
of the English grammar and punctuation.

He has visualized the general case like below:

this would read better if like were replaced with >> as

He has visualized the general case as below:

Now, we’ll use Kernel Density Plots to plot each sample to compare them.>>
Now, we’ll use Kernel Density Plots in order to plot each sample to compare them.

As we can see, both distributions are strongly negatively skewed, with distribution of values for 2015>>
As we can see, both distributions are strongly negatively skewed, with the distribution of values for 2015

Regarding the function, it was only used once and so therefore seemed like a lot of code for what was accomplished in the solution by this line–

fandango_2015['Fandango_Stars'].value_counts(normalize = True).sort_index() * 100

def freq_table_maker(series_obj):
    intervals = pd.interval_range(start=2, end=5, freq=0.5)
    freq_table = pd.Series([0,0,0,0,0,0], index=intervals)

    for star in series_obj:
        for interval in intervals:
            if star in interval:
                freq_table.loc[interval] += 1
    freq_dist = round(freq_table / freq_table.sum() * 100, 2)
    return pd.DataFrame({'Percentile':freq_dist.values, 'Stars':freq_dist.index, })

print("Below are the distribution of stars for the year 2015:")
1 Like

Thanks for the feedback! Especially for the grammar improvements, yeah I could use some better, un-rushed grammar. :sweat_smile:
I will keep the function part saved somewhere on my mind to use on my later projects.