Efficient Pairwise Association Rules for Personalized Recommendations: Leveraging Caching and Asynchronous Model Updates

Mortazavi, Seyed Mohammad; Feyzi, Farid

doi:10.22124/cse.2025.30749.1108

Document Type : Original Article

Authors

¹ Ahrar Institute of Technology and Higher Education

² University of Guilan

https://doi.org/10.22124/cse.2025.30749.1108

Abstract

Recommender systems based on content-based and collaborative filtering techniques face significant challenges, including the cold-start problem and privacy concerns due to their reliance on user profiles and product metadata. This study presents an optimized pairwise association rules (PAR) algorithm that addresses these limitations by operating independently of personal user data while maintaining recommendation accuracy. The proposed solution incorporates three key enhancements: (1) a privacy-preserving design using only transactional co-occurrence patterns, (2) a caching mechanism for modular training models that reduces recommendation latency by up to 102%, and (3) asynchronous execution for efficient resource management. Evaluations on a dataset of 20,000 food items demonstrate the algorithm's effectiveness, showing 18.7% higher nDCG scores than conventional methods while maintaining sub-second response times even with large-scale catalogs. The PAR algorithm proves particularly robust in sparse-data scenarios and cold-start conditions, offering a practical alternative to traditional approaches.

Keywords

References

[1] Panniello, U., Hill, S., & Gorgoglione, M. (2016). The impact of profit incentives on the relevance of online recommendations. Electronic Commerce Research and Applications, 20, 87–104. https://doi.org/10.1016/j.elerap.2016.10.003

[2] Li, Y.-M., Wu, C.-T., & Lai, C.-Y. (2013). A social recommender mechanism for e-commerce: Combining similarity, trust, and relationship. Decision Support Systems, 55(3), 740–752. https://doi.org/10.1016/j.dss.2013.02.009

[3] Agarwal, V., & Bharadwaj, K. K. (2013). A collaborative filtering framework for friends recommendation in social networks based on interaction intensity and adaptive user similarity. Social Network Analysis and Mining, 3(3), 359–379. https://doi.org/10.1007/s13278-012-0083-7

[4] Symeonidis, P., & Mantas, N. (2013). Spectral clustering for link prediction in social networks with positive and negative links. Social Network Analysis and Mining, 3(4), 1433–1447. https://doi.org/10.1007/s13278-013-0128-6

[5] Papadimitriou, A., Symeonidis, P., & Manolopoulos, Y. (2012). Fast and accurate link prediction in social networking systems. Journal of Systems and Software, 85(9), 2119–2132. https://doi.org/10.1016/j.jss.2012.04.019

[6] Sahebi, S., & Cohen, W. (2011). Community-Based Recommendations: A Solution to the Cold Start Problem.

[7] Guo, G., Zhang, J., & Yorke-Smith, N. (2015). Leveraging multiviews of trust and similarity to enhance clustering-based recommender systems. Knowledge-Based Systems, 74, 14–27. https://doi.org/10.1016/j.knosys.2014.10.016

[8] Bao, J., Zheng, Y., Wilkie, D., & Mokbel, M. (2015). Recommendations in location-based social networks: A survey. GeoInformatica, 19(3), 525–565. https://doi.org/10.1007/s10707-014-0220-8

[9] Chekkai, N., Chikhi, S., & Kheddouci, H. (2012). A weighted-graph based approach for solving the cold start problem in collaborative recommender systems. 000759–000764. https://doi.org/10.1109/ISCC.2012.6249390

[10] Hu, Y., Shi, W., Li, H., & Hu, X. (2017). Mitigating Data Sparsity Using Similarity Reinforcement-Enhanced Collaborative Filtering. ACM Transactions on Internet Technology, 17, 1–20. https://doi.org/10.1145/3062179

[11] Lemire, D., & Maclachlan, A. (2007). Slope One Predictors for Online Rating-Based Collaborative Filtering. Proceedings of the 2005 SIAM International Conference on Data Mining, SDM 2005, 5. https://doi.org/10.1137/1.9781611972757.43

[12] Trihatmaja, R., & Wardhana Asnar, Y. D. (2018). Improving the Performance of Collaborative Filtering Using Outlier Labeling, Clustering, and Association Rule Mining. 2018 5th International Conference on Data and Software Engineering (ICoDSE), 1–6. https://doi.org/10.1109/ICODSE.2018.8705883

[13] Chen, Y., Li, F., & Fan, J. (2015). Mining association rules in big data with NGEP. Cluster Computing, 18(2), 577–585. https://doi.org/10.1007/s10586-014-0419-3

[14] Marín, N., Molina, C., Serrano, J., & Vila, M. (2008). A Complexity Guided Algorithm for Association Rule Extraction on Fuzzy DataCubes. Fuzzy Systems, IEEE Transactions On, 16, 693–714. https://doi.org/10.1109/TFUZZ.2007.905909

[15] Osadchiy, T., Poliakov, I., Olivier, P., Rowland, M., & Foster, E. (2019). Recommender system based on pairwise association rules. Expert Systems with Applications, 115, 535–542. https://doi.org/10.1016/j.eswa.2018.07.077

Computational Sciences and Engineering

Efficient Pairwise Association Rules for Personalized Recommendations: Leveraging Caching and Asynchronous Model Updates

References

References

Volume 4, Issue 2
September 2024
Pages 237-257

Efficient Pairwise Association Rules for Personalized Recommendations: Leveraging Caching and Asynchronous Model Updates

References

References

Volume 4, Issue 2September 2024Pages 237-257

Volume 4, Issue 2
September 2024
Pages 237-257