АНАЛІТИЧНИЙ ОГЛЯД КОНТЕНТНОЇ ТА КОЛАБОРАТИВНОЇ ФІЛЬТРАЦІЙ У РЕКОМЕНДАЦІЙНИХ СИСТЕМА

Автор(и)

  • Максим ШЕВЧЕНКО, асп. Київський національний університет імені Тараса Шевченка, Київ, Україна Автор

DOI:

https://doi.org/10.17721/AIT.2025.1.07

Ключові слова:

алгоритм Роккіо, векторна модель, гібридна фільтрація, колаборативна фільтрація, контентна фільтрація, рекомендаційні системи.

Анотація

Вступ. За стрімкого зростання обсягу цифрового контенту рекомендаційні системи стають ключовим інструментом для надання персоналізованих пропозицій. Вони сприяють відкриттю нових фільмів, музики та товарів, підтримуючи зацікавленість користувачів у використанні платформ. Актуальність дослідження алгоритмів рекомендаційних систем зумовлена необхідністю вдосконалення їхньої роботи для задоволення індивідуальних уподобань користувачів. Ця робота являє собою огляд та аналітичне дослідження алгоритмів рекомендаційних систем. Метою цієї роботи є систематизація, класифікація та критичний аналіз двох основних підходів у рекомендаційних системах: фільтрації на основі вмісту (контентної) та колаборативної фільтрації.

Методи. Огляд існуючих методів рекомендаційних систем, порівняльне й аналітичне дослідження.

Результати. Проаналізовано алгоритми рекомендаційних систем. Дано формальне визначення задачі рекомендацій, де вподобання користувачів моделюються як функціональна залежність від властивостей об’єктів. У межах фільтрації на основі вмісту розглянуто використання класифікаційних алгоритмів, таких як наївний баєсів класифікатор, і дерев рішень, а також алгоритму Роккіо, який застосовує релевантний зворотний зв’язок для оновлення профілю користувача. Проведено аналіз сильних і слабких сторін різних мір подібності між векторами. У колаборативній фільтрації досліджено memory-based підхід (user-based та item-based методи) і model-based техніки з акцентом на алгоритмі k-NN. Для подолання недоліків окремих методів запропоновано гібридний підхід, який об’єднує їхні переваги. Представлено способи інтеграції систем у гібридну модель, що дає змогу покращити точність рекомендацій.

Висновки. Результати роботи виокремлюють особливості зазначених методів фільтрації, демонструють вплив реалізації алгоритмів і вхідних даних на точність рекомендацій і час відповіді. Аналіз недоліків підкреслює значення комбінованого використання алгоритмів фільтрації для підвищення ефективності рекомендаційних систем, що робить гібридний підхід перспективним напрямом для подальших досліджень і впровадження.

Завантажити

Дані для завантаження поки недоступні.

Посилання

Anisimov, A. V., Marchenko, O. O., & Kysenko, V. K. (2011). A method for the computation of the semantic similarity and relatedness between natural language words. Cybernetics and Systems Analysis, 47, 515–522. https://doi.org/10.1007/s10559-011-9334-2

Anisimov, A. V., Marchenko, O. O., & Vozniuk, T. G. (2014). Determining Semantic Valences of Ontology Concepts by Means of Nonnegative Factorization of Tensors of Large Text Corpora. Cybernetics and Systems Analysis, 50, 327–337. https://doi.org/10.1007/s10559-014-9621-9

Baxla, M. A. (2014). Comparative study of similarity measures for item based top n recommendation [Unpublished thesis, National Institute of Technology Rourkela]. CORE. https://files.core.ac.uk/download/53190130.pdf

Belhaouari, S. B., Fareed, A., Hassan, S., & Halim, Z. (2023). A collaborative filtering recommendation framework utilizing social networks. Machine Learning with Applications, 14, 1–20. https://doi.org/10.1016/j.mlwa.2023.100495

Burke, R. (2007). Hybrid Web Recommender Systems. In P. Brusilovsky, A. Kobsa, & W. Nejdl (Eds.), Lecture Notes in Computer Science, 4321. The Adaptive Web (pp. 377–408). Springer. https://doi.org/10.1007/978-3-540-72079-9_12

Çano, E. (2017). Hybrid Recommender Systems: A Systematic Literature Review. Intelligent Data Analysis, 21, 1487–1524. https://doi.org/10.3233/IDA-163209

Deutschman, Z. (2023, August 7). Recommender Systems: Machine Learning Metrics and Business Metrics. Neptune AI. https://neptune.ai/blog/recommender-systems-metrics

Fkih, F. (2022). Similarity measures for Collaborative Filtering-based Recommender Systems: Review and experimental comparison. Journal of King Saud University - Computer and Information Sciences, 34(9), 7645–7669. https://doi.org/10.1016/j.jksuci.2021.09.014

Gaurav, P. (2023, February 14). Step by Step Content-Based Recommender system. Medium. https://medium.com/@prateekgaurav/step-by-step-content-based-recommendation-system-823bbfd0541c

Gershman, A., Meisels, A., Luke, K.-H., Rokach, L., Schclar, A., & Sturm, A. (2010). A Decision Tree Based Recommender System. In G. Eichler, P. Kropf, U. Lechner, P. Meesad, & H. Unger (Eds.), 10th International Conference on Innovative Internet Community Services: Vol. 165. Lecture Notes in Informatics (pp. 170–179). Gesellschaft für Informatik. https://dl.gi.de/server/api/core/bitstreams/ca0e5035-3a82-48a1-8eb8-8f49ee374161/content

Gosh, S., Nahar, N., Wahab, M. A., Biswas, M., Hossain, M. S., & Andersson, K. (2021). Recommendation system for E-commerce Using Alternating Least Squares (ALS) on Apache Spark. In P. Vasant, I. Zelinka, & G. W. Weber (Eds.), Intelligent Computing and Optimization: Vol. 1324. Advances in Intelligent Systems and Computing (pp. 880–893). Springer. https://doi.org/10.1007/978-3-030-68154-8_75

Grover, P. (2017, December 28). Various Implementations of Collaborative Filtering. Medium. https://towardsdatascience.com/various-implementations-of-collaborative-filtering-100385c6dfe0

Gunes, I., Kaleli, C., Bilge, A., & Polat, H. (2014). Shilling attacks against recommender systems: a comprehensive survey. Artificial Intelligence Review, 42, 767–799. https://doi.org/10.1007/s10462-012-9364-9

Herimanto, H, Samosir, K., & Ginting, F. (2024). A Comparative Analysis of Content-Based Filtering and TF-IDF Approaches for Enhancing Sports Recommendation Systems. Innovation in research of informatics, 6(2), 90–97. https://doi.org/10.37058/innovatics.v6i2.12404

Joy, J., & Renumol, V. G. (2020). Comparison of Generic Similarity Measures in E-learning Content Recommender System in Cold-Start Condition. In IEEE Bombay Section Signature Conference (pp. 175–179). Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/IBSSC51096.2020.9332162

Kumar, S. (2022, September 25). Collaborative Filtering based Recommender Systems for Implicit Feedback Data. Sumit’s Diary. https://blog.reachsumit.com/posts/2022/09/explicit-implicit-cf/

Liu, Y., Wang, S., Khan, M. S., & He, J. (2018). A novel deep hybrid recommender system based on auto-encoder with neural collaborative filtering. Big Data Mining and Analytics, 1(3), 211–221. https://doi.org/10.26599/BDMA.2018.9020019

Mandal, S., & Maiti, A. (2018). Explicit Feedbacks Meet with Implicit Feedbacks: A Combined Approach for Recommender system. In L.M. Aiello, H. Cherifi, P. Lió, L.M. Rocha, C. Cherifi, R. Lambiotte (Eds.), 7th International Conference on Complex Networks and their Applications: Vol. 813. Studies in Computational Intelligence (pp. 169–181). Springer. https://doi.org/10.1007/978-3-030-05414-4_14

Marchenko, O. O. (2016). A Method for Automatic Construction of Ontological Knowledge Bases. Development of a Semantic-Syntactic Model of Natural Language. Cybernetics and Systems Analysis, 52, 20–29. https://doi.org/10.1007/s10559-016-9795-4

Marchenko, O., & Shevchenko, M. (2024). Influence of distance measures and data characteristics on time performance in content-based and collaborative filtering datasets. In A. Anisimov, V. Snytyuk, A. Chris, A. Pester, F. Mallet, I. Krak, N. Cogan, O. Chertov, O. Marchenko, S. Bozóki, T. Needham, V. Tsyganok, & V. Vovk (Eds.), Information Technology and Implementation: Vol. 3909. Central Europe University Repository Workshop Proceedings (pp. 99-108). CEUR-WS. https://ceur-ws.org/Vol-3909/Paper_8.pdf

Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., & Galstyan, A. (2021). A Survey on Bias and Fairness in Machine Learning. ACM Computing Surveys, 54(6), 115. https://doi.org/10.1145/3457607

Meteren, R., & Someren, M. (2000). Using Content-Based Filtering for Recommendation. In Proceedings of the machine learning in the new information age: Vol. 30. MLnet/ECML2000 workshop (pp. 47–56). ICS. https://users.ics.forth.gr/~potamias/mlnia/paper_6.pdf

Analytics Vidhya. (2024, October 14). Movie Recommendation and Rating Prediction using K-Nearest Neighbors. https://www.analyticsvidhya.com/blog/2020/08/recommendation-system-k-nearest-neighbors/

Nguyen, A. (2016). Singular Value Decomposition in Recommender Systems [Honors project, Texas Christian University]. TCU Digital Repository. https://repository.tcu.edu/server/api/core/bitstreams/7483e691-6fc0-4a82-9185-3adeb00cde44/content

Qazi, M., Fung, G. M., Meissner, K. J., & Fontes, E. R. (2017). An insurance recommendation system using Bayesian networks. In 11th ACM Conference on Recommender Systems (pp. 274–278). Association for Computing Machinery. https://doi.org/10.1145/3109859.3109907

Ricci, F. (2002). Content-Based Filtering and Hybrid Methods. EIA. http://eia.udg.es/arl/Agentsoftware/3-ContentBasedHybrid.pdf

Roy, D., & Dutta, M. (2022). A systematic review and research perspective on recommender systems. Journal of Big Data, 9, 59. https://doi.org/10.1186/s40537-022-00592-5

Sabiri, B., Khtira, A., El Asri, B., & Rhanoui, M. (2025). Hybrid Quality-Based Recommender Systems: A Systematic Literature Review. Journal of Imaging, 11(1), 12. https://doi.org/10.3390/jimaging11010012

Saranya, K. G., Sadasivam, G. S., & Chandralekha, M. (2016). Performance Comparison of Different Similarity Measures for Collaborative Filtering Technique. Indian Journal of Science and Technology, 9(29), 1–8. https://doi.org/10.17485/ijst/2016/v9i29/91060

Steck, H. (2019). Markov Random Fields for Collaborative Filtering. Advances in Neural Information Processing Systems, 32, 5473–5484. https://doi.org/10.48550/arXiv.1910.09645

Sun, S.-B., Zhang, Z.-H., Dong, X.-L., Zhang, H.-R., Li, T.-J., Zhang, L., & Min, F. (2017). Integrating Triangle and Jaccard similarities for recommendation. PLoS ONE, 12(8), e0183570. https://doi.org/10.1371/journal.pone.0183570

Vijay, H. (2020, April 11). Recommendation System using kNN. Auriga. https://aurigait.com/blog/recommendation-system-using-knn/

Wayesa, F., Betalo, M. L., Asefa, G. & Kedir, A. (2023). Pattern-based hybrid book recommender system using semantic relationships. Scientific Report, 13, 3693. https://doi.org/10.1038/s41598-023-30987-0

Wijewickrema, M., Petras, V., & Dias, N. (2019). Selecting a text similarity measure for a content-based recommender system: A comparison in two corpora. The Electronic Library, 37(3), 506–527. https://doi.org/10.1108/EL-08-2018-0165

Xia, Z., Sun, A., Xu, J., Peng, Y., Ma, R., & Cheng, M. (2024). Contemporary Recommendation Systems on Big Data and Their Applications: A Survey. IEEE Access, 12, 196914–196928. https://doi.org/10.1109/ACCESS.2024.3517492

Yadav, V., Shukla, R., Tripathi, A., & Maurya, A. (2021). A New Approach for Movie Recommender System using K-means Clustering and PCA. Journal of Scientific & Industrial Research, 80(2), 159–165. https://doi.org/10.56042/JSIR.V80I02.40102

Yang, Q., Liu, Y., Chen, T., & Tong, Y. (2019). Federated Machine Learning: Concept and Applications. ACM Transactions on Intelligent Systems and Technology, 10(2), 12. https://doi.org/10.48550/arXiv.1902.04885

Zisopoulos, Z., Karagiannidis, S., Demirtsoglou, G., & Antaris, S. (2008, October). Content-Based Recommender systems. ResearchGate. https://www.researchgate.net/publication/236895069_Content-Based_Recommendation_Systems

Завантаження

Опубліковано

2025-11-17

Номер

Розділ

Огляди й обговорення