A Preliminary Study on a Conceptual Game Feature Generation and Recommendation System

TL;DR

This paper presents a system that generates game feature suggestions from text prompts using machine learning models trained on extensive game descriptions, aiming to assist game designers at a conceptual level.

Contribution

It introduces a novel feature generation system combining word embeddings and generator models, and evaluates its effectiveness through a user study comparing different approaches.

Findings

GPT-2 outperformed human suggestions in some cases

Human suggestions received the most overall votes

The system demonstrates potential as a game design assistant

Abstract

This paper introduces a system used to generate game feature suggestions based on a text prompt. Trained on the game descriptions of almost 60k games, it uses the word embeddings of a small GLoVe model to extract features and entities found in thematically similar games which are then passed through a generator model to generate new features for a user's prompt. We perform a short user study comparing the features generated from a fine-tuned GPT-2 model, a model using the ConceptNet, and human-authored game features. Although human suggestions won the overall majority of votes, the GPT-2 model outperformed the human suggestions in certain games. This system is part of a larger game design assistant tool that is able to collaborate with users at a conceptual level.