Supervised learning models to predict engagement for Facebook and Twitter posts. Developed using the Scikit-learn Python machine learning library.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
facebook
twitter
README.md

README.md

astro-ml

Description

Supervised learning models to predict engagement rate of Facebook and Twitter posts developed using the Scikit-learn python machine learning library.

Repository content

The /scripts folder contains the script to extract the social media posts using the Facebook Graph API and Twitter REST API respectively.

The /classifiers folder contains the three classifiers developed for this project: Decision Tree, Naive Bayes, and linear model with SGD training.

Publication

The findings of this project are published in the Computers in Human Behavior Journal.

Citation:

Hwong, Y. L., Oliver, C., Van Kranendonk, M., Sammut, C., & Seroussi, Y. (2017). What makes you tick? The psychology of social media engagement in space science communication. Computers in Human Behavior, 68, 480-492.

Paper can be downloaded here