Mining Patterns in BoardGameGeek Data: An Exploratory Analysis for Board Game Recommendation Systems

• Eric Oliveira Leal UNIRIO
• Daniel Villaça de Oliveira Silva UNIRIO
• Jefferson Elbert Simões UNIRIO

DOI: https://doi.org/10.5753/sbgames.2026.26410

Resumo

Introduction: The availability of a large-scale dataset from BoardGameGeek (BGG) enables the exploration of board game metadata to extract patterns that can be useful for recommendation systems. However, the high dimensionality and heterogeneity of these attributes complicate the selection of which characteristics have the greatest descriptive power and representativeness. Objective: This work aims to analyze this dataset to identify which attributes are most informative and how their relationships can be used to support the design of board game recommendation models. Methodology: The approach combined web scrapping techniques to reconstruct the dataset in 2025 with quantitative analyses of Shannon entropy, as well as correlation and conditional expectation analyses on popularity metrics. Results: The entropy analysis showed that Categories are the most efficient attribute in terms of exploiting their theoretical potential (86%), whereas Families exhibit higher absolute entropy (9.31 bits) but a highly asymmetric distribution. Mechanics emerge as the central unit of technical description, displaying behavior close to a power law, which is favorable for computationally modeling game profiles. These findings suggest that combining Categories and Mechanics, while carefully handling sparse Families, can provide a solid starting point for defining features and similarity functions in graph-based recommendation systems.

Palavras-chave: Board games, BoardGameGeek, Metadata, Data Analysis, Recommendation Systems

Referências

Harrington, J. (2024). (re)platforming subcultures: Sinosphere board game geeks. Global Media and China, 0(0):20594364241310553.

Jorro-Aragoneses, J. L., Cantador, I., e Bellogín, A. (2024). Context-aware board game recommendations. Bari, Italy), RecSys.

Kim, J., Wi, J., Jang, S., e Kim, Y. (2020). Sequential recommendations on board-game platforms. Symmetry, 12(2).

Lima, H. e Falcão, T. (2025). Relações entre mecânicas de jogos de tabuleiro e os pilares do pensamento computacional. In Anais do XXXVI Simpósio Brasileiro de Informática na Educação, pages 1570–1579, Porto Alegre, RS, Brasil. SBC.

Piette, É., Stephenson, M., Soemers, D. J. N. J., e Browne, C. (2021). General board game concepts. CoRR, abs/2107.01078.

Putra, D. e Wibowo, A. (2022). Sentiment analysis for board game review using deep learning and sentiment lexicon. Int. J. Emerg. Technol. Adv. Eng, 12(6):56–62.

Samarasinghe, D., Barlow, M., Lakshika, E., Lynar, T., Moustafa, N., Townsend, T., e Turnbull, B. (2021). A data driven review of board game design and interactions of their mechanics. IEEE Access, 9:114051–114069.

Shannon, C. E. (1948). A mathematical theory of communication. The Bell System Technical Journal, 27(3):379–423.

Silva, E., Araújo, C., e Júnior, J. S. (2025). Métodos de game design aplicados a jogos de tabuleiro: Uma revisão sistemática. In Anais do XXIV Simpósio Brasileiro de Jogos e Entretenimento Digital, pages 219–232, Porto Alegre, RS, Brasil. SBC.

Xiao, T. (2025). Analysis of factors influencing board game ownership based on a gradient boosting model. Theoretical and Natural Science, 130:16–24.