Exploring the features of shareability

The stories that we have collected here represent approximately the 500 most shared pieces of content over the past year. We're going to try to deconstruct these articles to find the common traits that make them so shareable. We'll begin by looking at the image data.

Exploring image data

Let's begin by looking at the number of images that are included with each story. We'll run a value count and then plot the numbers:

dfc['img_count'].value_counts().to_frame('count')

The preceding code generates the following output:

Exploring image data

Now, let's plot that same information:

fig, ax = plt.subplots(figsize=(8,6))
y = dfc['img_count'].value_counts().sort_index() ...

Get Python Machine Learning Blueprints now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.