Understanding Facebook Audience Behaviour

North Africa


Client Brief

  • The client wanted to have a better understanding of how people engage on their Facebook pages

  • The main objective was to understand which topics generate engagement and how users interact with the page and with each other


Solution

  • Using Natural Language Processing techniques, the posts published by the client's pages were clustered by topics

  • For each topic, user engagement (likes, reactions, shares, comments) was analysed to identify the successful topics

  • The textual content (users comments) was also analysed to know more about their feeling about topics (sentiment analysis) and what they discuss (word clouds)

  • Network analysis was also conducted on users' comments to know who answers to who in the comments


Deliverable

  • Final presentation containing all the data visualisations (word clouds, sentiment graphs, networks...) and conclusions