Gamemood

References | 29 April 2025

GAMEMOOD: AI to Understand Feelings in the Gaming World

At the intersection of artificial intelligence and video games, a still largely unexplored space emerges: the emotional analysis of communication that takes place in virtual environments and in the context of video game streaming. Despite the enormous social and economic impact of this digital ecosystem, research on how to apply AI techniques to evaluate emotional responses on platforms such as Twitch—particularly in Spanish—is surprisingly scarce. In fact, our team has been a pioneer with the first study published to date that addresses this topic in depth (see study).

This gap opens up a major opportunity for innovation. Through rigorous technical work, we have developed a unique web application that integrates emotional artificial intelligence specifically trained in the gaming environment. Our technology is built on two main pillars:

  • Specialized linguistic corpus: We have created a corpus annotated with different levels of polarity and emotions, covering a wide variety of video game genres, including multiplayer, cooperative, competitive, and single-player titles. Some examples are League of Legends, Fortnite, World of Warcraft, and Dark Souls.
  • Advanced machine learning algorithms: Using models based on BERT (Bidirectional Encoder Representations from Transformers), we have achieved high accuracy in the emotional detection of messages in Spanish gaming streaming channels, identifying both polarity and specific emotions in real time.

All this technological foundation has been integrated into an intuitive web platform, with interactive visualizations and an interface designed to be accessible to all types of users—from developers to analysts, researchers, or content creators—without the need for advanced technical knowledge.

We are currently making progress in the development of key functionalities such as:

  1. Definition of user type–specific requirements.
  2. Design of a visual, user-friendly, and adaptive interface.
  3. Integration of AI models based on NLP (Natural Language Processing).
  4. Creation of easy-to-interpret emotional visualization dashboards.

With this application, we take a decisive and innovative step toward the future of emotional analysis in video games, opening new pathways for the design of more immersive experiences, more comprehensive market studies, and novel business models within the gaming industry.

Previous
Emotions, Data and AI: A New Era for Emotional Analysis in Video Games
Next
GAMEMOOD: Real Applications and Potential Beneficiaries in the Gaming Ecosystem