First Look’s sentiment analysis adds another piece of the technical puzzle

First Look Games has introduced its ‘Sentiment Analysis’ tool which it says provides “another piece of the technical innovation puzzle” to further its ambition of being “the ultimate vehicle to market online casino games”.

Aiming to help developers better understand how their games are being received by affiliates, the tool uses natural language processing and artificial intelligence to read and decipher the sentiment behind reviews written by all of the affiliates that have signed up to and utilised FLG.

Each affiliate review will then be given a sentiment score from one (clearly negative) through to five (clearly positive), with the average for each game collated and shared back to the studio. Developers will also receive an average score for all their games.

“Ever since we launched our platform, our game studio partners have been seeking ways in which to improve the quality of the review their games get from affiliates,” commented Tom Galanis, managing director of First Look Games.

“Our accuracy management service has delivered that objectively, assisting affiliates in accessing and where necessary, correcting key information. Through the integration of natural language processing, we are now using artificial intelligence to read and decipher the sentiment behind the reviews affiliates make of our studio partners’ games.

“Sentiment analysis is a great addition to our suite of tools that allow our studio partners to have a clear understanding of how their games are being received by affiliates and players.” 

Furthermore, the score will also be factored into the First Look Games index, which ranks all of the games listed in its library. This is currently based on the number of game reviews made by affiliates, player audience reach and the number of free-to-play demo game sessions played.

The tool has been built as an extension to the group’s accuracy management service, which identifies key game information that affiliates review inaccurately and automatically communicates the appropriate correction on behalf of the studio.