secureDETECTING THE MOST ENGAGING CONTENT ON SOCIAL MEDIA
Solution Head Banner

DETECTING THE MOST ENGAGING CONTENT ON SOCIAL MEDIA

4 Jun · 1 min read

Challenge: Marketers needed a tool that would help them define what content worked on social media for different industries.

Solution: We analyzed social media posts on Facebook and Twitter and built a tool to help social media managers identify which posts perform best in their industry.

User Group: Marketers

We analyzed social media posts on Facebook and Twitter and built a tool to help social media managers identify which posts perform best in their industry. The predictive insights help clients source and create compelling content without a huge investment of time.

Problem: Marketers needed a tool that would help them define what content worked on social media for different industries. The goal of the project was to demonstrate the relationship between user content engagement and its semantic, syntactic features and come up with a system that automatically detects the most engaging content from across the web for any given category based on a defined taxonomy.

Step one: Building a dataset

  • We gathered data from Twitter and other social media platforms, then defined the taxonomy per industry.
  • We worked with human tagging specialists to label data (tagging 10K mentions in three category levels — Level 1: 10 categories, level 2: 46, level 3: 29).

Step two: Designing the model

  • The client knew their requirements. To make sure the model brought real business value, we took the time to define their specific needs, determine the KPIs, and design the tests.

Step three: Achieving a solid level of accuracy

  • We combined different machine learning techniques with expert knowledge and human input and trained the algorithm beyond the assumed level of accuracy.
  • We can now integrate the models into new product features.

Leave a comment

Comment as

Comment as

0 comments