Kullanıcı Tabanlı ve Öğe Tabanlı İşbirlikçi Filtreleme ile Kümeleme Algoritmalarının Değerlendirilmesi

Yıl 2021, Sayı: 28, 453 - 458, 30.11.2021

Mustafa Özgür Cingiz , Kadriye Marangoz

https://doi.org/10.31590/ejosat.1005391

Öz

Öneri sistemleri, kullanıcıların memnuniyetini ve bağlılığını arttırıp, kullanıcılara kişiselleştirilmiş sistem deneyimini yaşatabilmek için geliştirilmiştir. Öneri sistemleri sayesinde kullanıcılar tercihlerine en uygun olan sonucu en az çaba göstererek bulabilmektedirler. Kullanıcıya özel öneri sistemlerinin önemi son yıllarda giderek artmakta ve filmler, şarkılar, haberler başta olmak üzere çeşitli alanlarda uygulanmaktadır. Öneri sistemleri hafıza tabanlı ve model tabanlı olmak üzere ikiye ayrılmaktadır. Model tabanlı filtreleme yaklaşımlarından olan işbirlikçi filtreleme yöntemleri, öneri sistemlerinde yaygın olarak kullanılmaktadır. Bu çalışmada Jester veri seti içerisinde bulunan şakalar kullanıcı tabanlı ve öğe tabanlı işbirlikçi filtreleme yöntemleri ile kümelenmiştir. Sonuçlar Davies–Bouldin İndeksi, Dunn İndeksi ve Silhouette Katsayısı değerlerine göre karşılaştırılmıştır. Karşılaştırmaların sonuçlarına göre öğe tabanlı işbirlikçi filtreleme yönteminin kullanıcı tabanlı işbirlikçi filtreleme yöntemine göre daha iyi bir doğruluk sağladığı görülmüştür.

Anahtar Kelimeler

Öneri Sistemleri, İşbirlikçi Filtreleme, Spektral Çift Kümeleme., K-ortalama, Hiyerarşik Kümeleme

Evaluation of Clustering Algorithms with User Based and Item Based Collaborative Filtering

Öz

Recommendation systems have been developed to increase the satisfaction and loyalty of the users and to provide the users with a personalized system experience. Because of recommendation systems, users can find most suitable result for their preferences with least effort. The importance of user-specific recommendation systems has been increasing in recent years and has been applied in various fields, especially movies, songs, and news. Suggestion systems are divided into memory-based and model-based. Collaborative filtering methods, which are model-based filtering approaches, are widely used in recommendation systems. In this study, jokes in jester dataset were clustered with user-based and item-based collaborative filtering methods. The results were compared according to the davies–bouldin index, dunn index and silhouette score. According to the general result of the comparisons, it has been seen that item-based collaborative filtering method provides better accuracy than the user-based collaborative filtering method.

Anahtar Kelimeler

Recommender Systems, Collaborative Filtering, K-means, Hierarchical Clustering, Spectral Bi-clustering

Kaynakça

• (2021). Geeksforgeeks: https://www.geeksforgeeks.org/dunn-index-and-db-index-cluster-validity-indices-set-1/
• (2021). Scikit-Learn: https://scikit-learn.org/stable/modules/generated/sklearn.metrics.silhouette_score.html
• (2021). Scikit-Learn: https://scikit-learn.org/stable/modules/generated/sklearn.metrics.davies_bouldin_score.html
• (2021). Scikit-Learn: https://scikit-learn.org/stable/modules/biclustering.html
• (2021). Veri Bilimi Okulu: https://www.veribilimiokulu.com/hiyerarsik-kumeleme/
• (2021). The Goldberg Berkeley: https://goldberg.berkeley.edu/jester-data/
• Fathan, G., Adji, T. B., & Ferdiana, R. (2018). Impact of Matrix Factorization and Regularization Hyperparameter on a Recommender System for Movies. 2018 5th International Conference on Electrical Engineering, Computer Science and Informatics (EECSI).
• Hatipoğlu, E. (2021). The Medium website: https://medium.com/@ekrem.hatipoglu/machine-learning-clustering-k%C3%BCmeleme-k-means-algorithm-part-13-be33aeef4fc8
• Koren, Y. (2008). Factorization meets the neighborhood: a multifaceted collaborative filtering model. Proceeding of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining.
• Krishna, M. G., Haseeb, M., Siyad, B. M., Zameel, P., Budget, & Raj, S. . Experience Based Travel Planner Using Collaborative Filtering. 2021 1st Odisha International Conference on Electrical Power Engineering, Communication and Computing Technology (ODICON).
• Ponnam, L. T., Punyasamudram, S. D., Nallagulla, S. N., & Yellamati, S. (2016). Movie recommender system using item based collaborative filtering technique. 2016 International Conference on Emerging Trends in Engineering, Technology and Science (ICETETS) (pp. 1-5). IEEE
• Raghuwanshi, S., & Pateriya, R. (2019). Collaborative Filtering Techniques in Recommendation Systems. In Data,Engineering and Applications. (pp 11-21). Springer, Singapore
• Sánchez-Moreno, D., González, A. B., Dolores, M., Vicente, M., Batista, V. F., N., M., & García, M. (2016). A collaborative filtering method for music recommendation using playing coefficients for artists and users. Expert Systems with Applications, 66, 234-244
• Sarwar, B., Karypsis, G., & Konstan, J. (2001). Item-based Collaborative Filtering Recommendation Algorithms. In Proceedings of the 10th International World Wide Web Conference on World Wide Web (pp. 285-295)
• Subramaniyaswamy, V., Logesh, R., Chandrashekhar, M., Challa, A., & Vijayakumar, V. (2017). A personalised movie recommendation system based on collaborative filtering. International Journal of High Performance Computing and Networking, 10(1-2), 54-6