Perbandingan Penerapan Teknik Popularitas Item dalam Sistem Rekomendasi

Noor Ifada; Yunus Safi'i

doi:10.33795/jip.v11i3.5200

Authors

Noor Ifada Universitas Trunojoyo Madura
Yunus Safi'i Universitas Trunojoyo Madura

DOI:

https://doi.org/10.33795/jip.v11i3.5200

Keywords:

bias popularitas, item popularity, efek long tail, sistem rekomendasi

Abstract

Tantangan utama dalam era big data adalah pengguna seringkali mengalami kesulitan untuk menentukan atau memilih item yang relevan dari banyaknya data yang tersedia. Permasalahan ini dapat diatasi dengan pengimplementasian sistem rekomendasi untuk menyaring informasi dan memberikan rekomendasi yang sesuai dengan preferensi pengguna. Walaupun sistem rekomendasi telah berkembang pesat dalam mengatasi tantangan informasi yang melimpah, masalah popularitas item masih menjadi isu utama. Item yang populer mendominasi rekomendasi, sedangkan item yang kurang populer mengalami kesulitas mendapatkan ekspousr kepada pengguna. Dalam konteks rekomendasi penjualan, hal ini dapat menghambat penjualan item yang kurang popular karena minimnya rekomendasi kepada pelanggan. Sejumlah penelitian telah dilakukan untuk mengatasi masalah ini, tetapi belum ada studi yang secara komprehensif membahas penerapan teknik popularitas item. Penelitian ini bertujuan melakukan studi tentang tren dari penelitian yang menerapkan teknik popularitas item dalam sistem rekomendasi, yaitu sejumlah 20 literatur yang dipublikasikan dari tahun 2012 sampai 2022. Hasil studi menunjukkan bahwa topik popularitas item mulai populer diteliti sejak tahun 2018 hingga 2022, yaitu sejumlah 16 literatur. Jenis literatur didominasi oleh prosiding yaitu sejumlah 15 literatur, dan sisanya berupa jurnal. Dataset yang paling banyak digunakan adalah MovieLens dan LastFM, yaitu untuk sistem rekomendasi film dan lagu. Fokus masalah yang paling mendominasi adalah penyelesaian permasalahan bias popularitas, yaitu sebanyak 12 literatur. Sedangkan luaran rekomendasi didominasi oleh rekomendasi top-N, yaitu sebanyak 18 literatur.

Downloads

References

Abdollahpouri, H., & Mansoury, M. (2020). Multi-sided Exposure Bias in Recommendation. Proceedings of The ACM KDD 1st International Workshop on Industrial Recommendation Systems, Virtual, 157–165, doi: https://doi.org/10.48550/arXiv.2006.15772.

Abdollahpouri, H., Mansoury, M., Burke, R., Mobasher, B., & Malthouse, E. (2021). User-centered evaluation of popularity bias in recommender systems. Proceedings of The 29th ACM Conference on User Modeling, Adaptation and Personalization, Utrecht, Netherlands, 119-129, doi: https://doi.org/10.1145/3450613.3456821.

Bertani, R. M., Bianchi, R. A. C., & Costa, A. H. R. (2020). Combining novelty and popularity on personalised recommendations via user profile learning. Expert Systems with Applications, 146 113149, doi: https://doi.org/10.1016/j.eswa.2019.113149.

Chen, C., Zhang, M., Liu, Y., & Ma, S. (2018). Missing data modeling with user activity and item popularity in recommendation. In Y. Tseng, et al. (Ed.), Information Retrieval Technology. AIRS 2018. Lecture Notes in Computer Science (Vol. 11292, pp. 113-125). Switzerland Springer International Publishing, doi: https://doi.org/10.1007/978-3-030-03520-4_11.

Chen, S.-H., Sou, S.-I., & Hsieh, H.-P. (2020). HPCF: hybrid music group recommendation system based on item popularity and collaborative filtering. Proceedings of The 2020 International Computer Symposium (ICS), Tainan, Taiwan, 43-49, doi: https://doi.org/10.1109/ICS51289.2020.00019.

Chen, T., Tian, H., & Zhu, X. (2014). An Enhanced Collaborative Filtering with Flexible Item Popularity Control for Recommender Systems. Proceedings of The 2014 International Conference on Social Computing (SocialCom '14), Beijing, China, 1-6, doi: https://doi.org/10.1145/2639968.2640076.

de Sousa Silva, D. V., & Durão, F. A. (2020). Dynamic clustering personalization for recommending long tail items. Proceedings of The 2020 15th Conference on Computer Science and Information Systems (FedCSIS), Sofia, Bulgaria, 417-425, doi: https://doi.org/10.15439/2020F157.

Fu, Z., Xian, Y., Geng, S., De Melo, G., & Zhang, Y. (2021). Popcorn: Human-in-the-loop popularity debiasing in conversational recommender systems. Proceedings of The 30th ACM International Conference on Information & Knowledge Management, Virtual Event Queensland, Australia, 494-503, doi: https://doi.org/10.1145/3459637.3482461.

Gao, X., Ji, Q., Mi, Z., Yang, Y., & Guo, Y. (2018). Similarity measure based on punishing popular items for collaborative filtering. Proceedings of The 2018 International Conference on Computer, Information and Telecommunication Systems (CITS), Alsace, Colmar, France, 1-5, doi: https://doi.org/10.1109/CITS.2018.8440147.

Hou, L., Pan, X., & Liu, K. (2018). Balancing the popularity bias of object similarities for personalised recommendation. The European Physical Journal B, 91, 47, doi: https://doi.org/10.1140/epjb/e2018-80374-8.

Ifada, N., Ummamah, & Sophan, M. K. (2020). Hybrid Popularity Model for Solving Cold-start Problem in Recommendation System. Proceedings of The 2020 International Conference on Sustainable Information Engineering and Technology (SIET), Malang, Indonesia, 40-44, doi: https://doi.org/10.1145/3427423.3427425.

Kowald, D., Schedl, M., & Lex, E. (2020). The unfairness of popularity bias in music recommendation: A reproducibility study. In J. Jose, et al. (Ed.), Advances in Information Retrieval. ECIR 2020. Lecture Notes in Computer Science (Vol. 12036, pp. 35-42). Switzerland: Springer International Publishing, doi: https://doi.org/10.1007/978-3-030-45442-5_5.

Lin, A., Wang, J., Zhu, Z., & Caverlee, J. (2022). Quantifying and mitigating popularity bias in conversational recommender systems. Proceedings of The 31st ACM International Conference on Information & Knowledge Management, Atlanta GA, USA, 1238-1247, doi: https://doi.org/10.1145/3511808.3557423.

Steck, H. (2011). Item popularity and recommendation accuracy. Proceedings of The fifth ACM conference on Recommender systems (RecSys '11), Chicago Illinois, USA, 125-132, doi: https://doi.org/10.1145/2043932.2043957.

Sun, C., & Xu, Y. (2019). Topic Model-Based Recommender System for Longtailed Products Against Popularity Bias. Proceedings of The 2019 IEEE Fourth International Conference on Data Science in Cyberspace (DSC), Hangzhou, China, 250-256, doi: https://doi.org/10.1109/DSC.2019.00045.

Valcarce, D., Parapar, J., & Barreiro, Á. (2016). Item-based relevance modelling of recommendations for getting rid of long tail products. Knowledge-Based Systems, 103, 41-51, doi: https://doi.org/10.1016/j.knosys.2016.03.021.

Wang, J., & Luo, N. (2016). A New Hybrid Popular Model for Personalized Tag Recommendation. Journal of Computers, 11(2), 116-123, doi: https://doi.org/10.17706/jcp.11.2.116-123.

Wei, F., Chen, S., Jin, J., Zhang, S., Zhou, H., & Wu, Y. (2022). Adaptive alleviation for popularity bias in recommender systems with knowledge graph. Security and Communication Networks, 2022, 4264489, doi: https://doi.org/10.1155/2022/4264489.

Yalcin, E. (2022). PopHybrid: a novel item popularity-aware hybrid approach for long-tail recommendation. Proceedings of The 2022 International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA), Ankara, Turkey, 1-6, doi: https://doi.org/10.1109/HORA55278.2022.9800006.

Yin, H., Cui, B., Li, J., Yao, J., & Chen, C. (2012). Challenging the long tail recommendation. Proceedings of The VLDB Endowment (The 38th International Conference on Very Large Data Bases), Istanbul, Turkey, 896-907, doi: https://doi.org/10.48550/arXiv.1205.6700.

Yomeldi, H. (2020). Decision Making in Internet of Things (IoT): A Systematic Literature Review. ITEJ (Information Technology Engineering Journals), 5(1), 51-65, doi: https://doi.org/10.24235/itej.v5i1.40.

Zhang, Y., Feng, F., He, X., Wei, T., Song, C., Ling, G., & Zhang, Y. (2021). Causal intervention for leveraging popularity bias in recommendation. Proceedings of The 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, Virtual Event Canada, 11-20, doi: https://doi.org/10.1145/3404835.3462875.

Zheng, Y., Gao, C., Li, X., He, X., Li, Y., & Jin, D. (2021). Disentangling user interest and conformity for recommendation with causal embedding. Proceedings of The Web Conference 2021 (WWW), Ljubljana, Slovenia, 2980-2991, doi: https://doi.org/10.1145/3442381.3449788.