Home // eKNOW 2026, The Eighteenth International Conference on Information, Process, and Knowledge Management // View article

Genre Prediction Using RNNs and LLM-Enhanced Video Game Review Data

Authors:
Gabriel Young
Susan Gauch

Keywords: Classic ML; AI/LLM; NLP; Video Games; Reviews.

Abstract:
Large Language Models (LLMs) are powerful tools for engaging with textual data, carrying many advantages over classical Natural Language Processing (NLP) and Machine Learning (ML) approaches. However, a classical ML model can still be faster, more efficient to run, and accessible than an LLM. We seek to capture the benefits of LLM-based text comprehension and preserve them within a classical ML model through a hybrid approach. The LLM operates on text to identify relevant information and associations within our problem space, then the ML model trains on the LLM output. The model may learn from the LLM and provide a more efficient alternative to querying the LLM directly for future data. Using review data from the video game marketplace Steam, we conduct a series of experiments toward this end. We prompt the LLM to surface various information from the raw data and train Recurrent Neural Networks (RNNs) to predict a single genre of the games, "Role-Playing Game" (“RPG”). We then evaluate the performance of the trained RNN models on the raw data, checking for generalizability and performance loss/improvement. Results are promising. At baseline, using raw review data, a balanced (50% RPG, 50% non-RPG) dataset, and no LLM assistance, a shallow RNN can predict the genre under test with an average accuracy of 64.1%. The maximum accuracy of the LLM on this same dataset is 84.1%. Our other models under test lie between these two bounds and demonstrate merits from engaging the LLM during their training.

Pages: 27 to 32

Copyright: Copyright (c) IARIA, 2026

Publication date: May 24, 2026

Published in: conference

ISSN: 2308-4375

ISBN: 978-1-68558-387-3

Location: Venice, Italy

Dates: from May 24, 2026 to May 28, 2026