iRec: Um framework para modelos interativos em Sistemas de Recomendação

• Thiago Silva UFSJ
• Adriano Pereira UFMG
• Leonardo Rocha UFSJ

DOI: https://doi.org/10.5753/ctd.2023.229296

Resumo

Nowadays, most e-commerce and entertainment services have adopted interactive Recommender Systems (RS) to guide the entire journey of users into the system. This task has been addressed as a Multi-Armed Bandit problem where systems must continuously learn and recommend at each iteration. However, despite the recent advances, there is still a lack of consensus on the best practices to evaluate such bandit solutions. Several variables might affect the evaluation process, but most of the works have only been concerned with the accuracy of each method. Thus, this master dissertation proposes an interactive RS framework named iRec. It covers the whole experimentation process by following the main RS guidelines. The iRec provides three modules to prepare the dataset, create new recommendation agents, and simulate the interactive scenario. Moreover, it also contains several state-of-the-art algorithms, a hyperparameter tuning module, distinct evaluation metrics, different ways of visualizing the results, and statistical validation.

Referências

Barraza-Urbina, A., Koutrika, G., d’Aquin, M., and Hayes, C. (2018). Bears: Towards an evaluation framework for bandit-based interactive recommender systems. REVEAL 18, October 6-7, 2018, Canada.

Dacrema, M. F., Boglio, S., Cremonesi, P., and Jannach, D. (2021). A troubling analysis of reproducibility and progress in recommender systems research. ACM TOIS, 39(2).

Saito, Y., Shunsuke, A., Megumi, M., and Yusuke, N. (2020). Open bandit dataset and pipeline: Towards realistic and reproducible off-policy evaluation. arXiv preprint arXiv:2008.07146.

Sanz-Cruzado, J., Castells, P., and López, E. (2019). A simple multi-armed nearest-neighbor bandit for interactive recommendation. In Proceedings of the 13th ACM Conference on Recommender Systems, pages 358–362.

Shams, S., Anderson, D., and Leith, D. (2021). Cluster-based bandits: Fast cold-start for recommender system new users.

Silva, N., Silva, T., Werneck, H., Rocha, L., and Pereira, A. (2023). User cold-start problem in multi-armed bandits: When the first recommendations guide the user’s experience. ACM Trans. Recomm. Syst., 1(1).

Silva, N., Werneck, H., Silva, T., Pereira, A. C., and Rocha, L. (2021). A contextual approach to improve the user’s experience in interactive recommendation systems. In Proceedings of the Brazilian Symposium on Multimedia and the Web, pages 89–96.

Silva, N., Werneck, H., Silva, T., Pereira, A. C., and Rocha, L. (2022a). Multi-armed bandits in recommendation systems: A survey of the state-of-the-art and future directions. Expert Systems with Applications, 197:116669.

Silva, T., Silva, N., Mito, C., Pereira, A. C. M., and Rocha, L. (2022b). Interactive poi recommendation: Applying a multi-armed bandit framework to characterise and create new models for this scenario. In Proceedings of the Brazilian Symposium on Multimedia and the Web, WebMedia ’22, page 211–221, New York, NY, USA. Association for Computing Machinery.

Silva, T., Silva, N., Werneck, H., Mito, C., Pereira, A. C., and Rocha, L. (2022c). Irec: An interactive recommendation framework. In Proceedings of the 45th International ACM SIGIR, page 3165–3175, New York, NY, USA. Association for Computing Machinery.

Silva, T., Silva, N., Werneck, H., Pereira, A. C., and Rocha, L. (2020). The impact of first recommendations based on exploration or exploitation approaches in recommender systems’ learning. In Proceedings of the Brazilian Symposium on Multimedia and the Web, pages 173–180.

Strong, E., Kleynhans, B., and Kadioglu, S. (2019). Mabwiser: A parallelizable contextual multi-armed bandit library for python. In 31st IEEE International Conference on Tools with Artificial Intelligence, ICTAI 2019, Portland, OR, USA, November 4-6, 2019, pages 909–914. IEEE.

Strong, E., Kleynhans, B., and Kadioglu, S. (2021). MABWiser: parallelizable contextual multi-armed bandits. Int. J. Artif. Intell. Tools, 30(4).

Sun, Z., Yu, D., Fang, H., Yang, J., Qu, X., Zhang, J., and Geng, C. (2020). Are we evaluating rigorously? benchmarking recommendation for reproducible evaluation and fair comparison. In Fourteenth ACM conference on recommender systems, pages 23–32.

Wu, Q., Iyer, N., and Wang, H. (2018). Learning contextual bandits in a non-stationary environment. In The 41st International ACM SIGIR, pages 495–504.

Wu, Q., Zhang, H., Gao, X., He, P., Weng, P., Gao, H., and Chen, G. (2019). Dual graph attention networks for deep latent representation of multifaceted social effects in recommender systems. In The World Wide Web Conference, pages 2091–2102.

Zhou, S., Dai, X., Chen, H., Zhang, W., Ren, K., Tang, R., He, X., and Yu, Y. (2020). Interactive recommender system via knowledge graph-enhanced reinforcement learning. In Proceedings of the 43rd International ACM SIGIR.

Zou, L., Xia, L., Gu, Y., Zhao, X., Liu, W., Huang, J. X., and Yin, D. (2020). Neural interactive collaborative filtering. In Proceedings of the 43rd International ACM SIGIR, pages 749–758.