A Controlled Study of Relevance–Diversity Trade-offs in News Recommendation with Embeddings

• Ana Lilian Alfonso Toledo UFSM https://orcid.org/0009-0001-3544-4336
• Vitalio Alfonso Reguera UTEC https://orcid.org/0000-0002-3647-8677
• Gabriel Machado Lunardi UFSM https://orcid.org/0000-0001-6655-184X

DOI: https://doi.org/10.5753/semish.2026.20503

Resumo

News recommender systems reduce information overload, but personalization can reinforce the filter-bubble effect by repeatedly prioritizing similar content. Post-filtering diversification re-ranks top-N lists using item–item similarity, so outcomes depend on item representations. We study how replacing binary news features with TruncatedSVD embeddings affects the relevance–diversity balance in a controlled pipeline. Using a dataset of Brazilian political news, we run offline temporal replay with KNN-U, KNN-I, and SVD plus Maximal Marginal Relevance (MMR), sweeping embedding sizes. Results show embeddings boost neighborhood-based relevance but increase homogeneity, sharpening the relevance–diversity trade-off.

Referências

Carbonell, J. and Goldstein, J. (1998). The use of mmr, diversity-based reranking for reordering documents and producing summaries. In Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval, pages 335–336.

Gao, Z., Shen, T., Mai, Z., Bouadjenek, M. R., Waller, I., Anderson, A., Bodkin, R., and Sanner, S. (2022). Mitigating the filter bubble while maintaining relevance: Targeted diversification with vae-based recommender systems. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 2524–2531.

Herlocker, J. L., Konstan, J. A., Borchers, A., and Riedl, J. (1999). An algorithmic framework for performing collaborative filtering. In Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval, pages 230–237.

Koren, Y., Bell, R., and Volinsky, C. (2009). Matrix factorization techniques for recommender systems. Computer, 42(8):30–37.

Liu, N., Hu, X. E., Savas, Y., Baum, M. A., Berinsky, A. J., Chaney, A. J. B., Lucas, C., Mariman, R., de Benedictis-Kessner, J., Guess, A. M., Knox, D., and Stewart, B. M. (2025). Short-term exposure to filter-bubble recommendation systems has limited polarization effects: Naturalistic experiments on youtube. Proceedings of the National Academy of Sciences, 122(8):e2318127122.

Lunardi, G. M., Machado, G. M., Maran, V., and de Oliveira, J. P. M. (2020). A metric for filter bubble measurement in recommender algorithms considering the news domain. Applied Soft Computing, 97:106771.

Martinsson, P.-G. and Tropp, J. A. (2020). Randomized numerical linear algebra: Foundations and algorithms. Acta Numerica, 29:403–572.

Sarwar, B., Karypis, G., Konstan, J., and Riedl, J. (2001). Item-based collaborative filtering recommendation algorithms. In Proceedings of the 10th international conference on World Wide Web, pages 285–295.

Wilhelm, M., Ramanathan, A., Bonomo, A., Jain, S., Chi, E. H., and Gillenwater, J. (2018). Practical diversified recommendations on youtube with determinantal point processes. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management, pages 2165–2173.

Ziegler, C.-N., McNee, S. M., Konstan, J. A., and Lausen, G. (2005). Improving recommendation lists through topic diversification. In Proceedings of the 14th international conference on World Wide Web, pages 22–32.