Case Study

NLP Solution for Customer Feedback Analysis

The Client

An international game developer and publisher. The client specialises in developing free-to-play online action games and is a market leader with over 60 million registered players.

The Challenges

  • The gaming community are quite vocal across multiple platforms, resulting in a lot of valuable content, however the client needed a more effective way of processing all of this data.

  • Developing a solution that can also pick up on gamer slang, which is notoriously difficult to do, and picking up actionable insights from this.

The Solutions

  • The first stages were training the semantic representation and using these to build and train two more models, one for topic prediction and the other for sentiment analysis.
  • ‍Built a solution for downloading all of the comments in a timely manner. 
  • Trained models for sentiment analysis and topic prediction.

The Results

  • Resulted in a 20 fold reduction in cost, and time spent on manual topic retrieval and classification of data collected.

  • The company saw significant improvement in the working process of their social research department, and in the quality of processed data within a couple of months after deployment.
Learn More
Thank you! A member of our team will be in touch as soon as possible.
Oops! Something went wrong while submitting the form.