طرّاحی سیستم پیشنهاددهنده ترکیبی برای حل مشکل شروع سرد به کمک خوشه‌بندی و الگوریتم‌ ازدحام ذرّات آشوبی

نویسندگان

1 دانشگاه بین المللی امام رضا(ع)

2 دانشگاه صنعتی سجاد

چکیده

امروزه یکی از مهم‌ترین چالش‌های افزایش اطلاعات، یافتن اطلاعات مورد علاقه از بین انبوه داده‌هاست. به این موضوع در طراحی سایت‌های تعاملی همواره توجّه شده‌است. سیستم‌های پیشنهاددهنده برای حل این مسئله به وجود آمده‌اند تا به کاربران برای رسیدن به اطلاعات مورد نظرشان کمک کنند؛ امّا این سیستم‌ها محدویت‌هایی دارند. یکی از مهم‌ترین چالش‌های پیش‌روی سیستم‌های پیشنهاددهنده، مشکل شروع سرد است. این مشکل زمانی به وجود می‌آید که یک کاربر (قلم‌داده) جدید وارد سیستم می‌شود. عدم وجود اطلاعات قبلی از این کاربر (قلم‌داده) باعث می‌شود سیستم نتواند به طور عادی پیشنهادها را تولید کند. در این مقاله برای حل مشکل شروع سرد کاربر، روش جدیدی به کمک ترکیب مدل‌های مبتنی برمحتوا و فیلترمشارکتی ارائه شده است. در این روش لیست پیشنهادی، دارای ویژگی‌هایی مانند کیفیت بالای قلم‌داده‌های پیشنهادی و تنوّع ‌آن‌ها است که دامنه‌ی اطلاعات دریافتی از کاربر را به سرعت گسترش می‌دهد، به همین دلیل کاربران را سریع‌تر از حالت شروع سرد خارج می‌کند. همچنین با استفاده از اطلاعات دموگرافیک کاربر، سعی شده قلم‌داده‌های لیست پیشنهادی به نحوی انتخاب شوند که به علایق کاربر نزدیک‌تر باشند تا دقت بیشتر شود.  نتایج ارزیابی روش پیشنهادی نشان می‌دهد میزان خطای MAE و RMSE نسبت به روش‌های موجود تا حد مطلوبی کاهش یافته‌است.
 

کلیدواژه‌ها


عنوان مقاله [English]

Design a Hybrid Recommender System Solving Cold-start Problem Using Clustering and Chaotic PSO Algorithm

نویسندگان [English]

  • Nafisah Moosarrezayi 1
  • Javad Hamidzadeh 2
1
2
چکیده [English]

One of the main challenges of increasing information in the new era, is to find information of interest in the mass of data. This important matter has been considered in the design of many sites that interact with users. Recommender systems have been considered to resolve this issue and have tried to help users to achieve their desired information; however, they face limitations. One of the most important challenges that they face is cold-start problem, which is raised when a new user/item entered into the system, while no previous information is available for it. The lack of previous knowledge of the new user/item, will causes the system fails generating its suggestions normally. In this paper, to solve the problem of cold-start user/item a new method is presented using combining content-based models and collaborative filtering. Moreover, demographic data is used to recommend the nearest items to cold-start users/items' interests. Compared to existing methods, the evaluation results show that the proposed method reduces the MAE and RMSE error.

کلیدواژه‌ها [English]

  • Recommender systems
  • Cold-Start Problem
  • Content-Based Method
  • Collaborative Filtering Method
  • Quality
  • Diversity
[1] E. R. Núñez-Valdéz, J. M. C. Lovelle, O. S. Martínez, V. García-Díaz, P. O. de Pablos, & C. E. M. Marín, “Implicit feedback techniques on recommender systems applied to electronic books,” Computers in Human Behavior, vol. 28 ,no. 4, pp. 1186-1193, 2012. [2] S. Roy, & S. C. Guntuku, “Latent Factor Representations for Cold-Start Video Recommendation,” In Proceedings of the 10th ACM Conference on Recommender Systems, pp. 99-106, ACM, 2016. [3] A. B. Barragáns-Martínez, E. Costa-Montenegro, J. C. Burguillo, M. Rey-López, F. A. Mikic-Fonte, & A. Peleteiro, “A hybrid content-based and item-based collaborative filtering approach to recommend TV programs enhanced with singular value decomposition,” Information Sciences, vol. 180, no. 22, pp. 4290-4311, 2010. [4] S. K. Lee, Y. H. Cho, & S. H. Kim, “Collaborative filtering with ordinal scale-based implicit ratings for mobile music recommendations,” Information Sciences, vol. 180, no. 11, pp. 2142-2155, 2010. [5] J. E. S. U. S. Bobadilla, F. Serradilla, & A. Hernando, “Collaborative filtering adapted to recommender systems of e-learning,” Knowledge-Based Systems, vol. 22, no. 4, pp. 261-265, 2009 [6] J. J. Castro-Schez, R. Miguel, D. Vallejo, & L. M. López-López, “A highly adaptive recommender system based on fuzzy logic for B2C e-commerce portals,” Expert Systems with Applications, vol. 38, no. 3, pp. 2441-2454, 2011. [7] T. T. S. Nguyen, H. Y. Lu, & J. Lu, “Web-page recommendation based on web usage and domain knowledge,” IEEE Transactions on Knowledge and Data Engineering, vol. 26, no. 10, pp. 2574-2587, 2014. [8] L. H. Son, “Dealing with the new user cold-start problem in recommender systems: A comparative review,” Information Systems, vol. 58, pp. 87-104, 2016. [9] C. C. Aggarwal, “Recommender Systems,” Springer, 2016. [10] L. Safoury, & A. Salah, “Exploiting user demographic attributes for solving cold-start problem in recommender system,” Lecture Notes on Software Engineering, vol. 1, no. 3, pp. 303, 2013. [11] F. Peng, J. Lu, Y. Wang, R. Yi-Da Xu, C. Ma, & J. Yang, “N-dimensional Markov random field prior for cold-start recommendation,” Neurocomputing, vol. 191, pp. 187-199, 2016. [12] C. B. Huang, & S. J. Gong, “Employing rough set theory to alleviate the sparsity issue in recommender system.” International Conference on Machine Learning and Cybernetics, Vol. 3, pp. 1610-1614, IEEE, 2008. [13] J. Xu, Y. Yao, H. Tong, X. Tao, & J. Lu, “Ice-breaking: Mitigating cold-start recommendation problem by rating comparison,” In Proceedings of the 24th International Conference on Artificial Intelligence, pp. 3981-3987, AAAI Press, 2015. [14] A. Hernando, J. Bobadilla, F. Ortega, & A. Gutiérrez, “A probabilistic model for recommending to new cold-start non-registered users,” Information Sciences, vol. 376, pp. 216-232, 2017. [15] S.Feil, M. Kretzer, K. Werder, & A. Maedche, “Using Gamification to Tackle the Cold-Start Problem in Recommender Systems,” In Proceedings of the 19th ACM Conference on Computer Supported Cooperative Work and Social Computing Companion, pp. 253-256, ACM, 2016. [16] L. H. Son, “HU-FCF: a hybrid user-based fuzzy collaborative filtering method in recommender systems,” Expert Systems with Applications: An International Journal, vol. 41, no. 15, pp. 6861-6870, 2014. [17] L. H. Son, “HU-FCF++: A novel hybrid method for the new user cold-start problem in recommender systems,” Engineering Applications of Artificial Intelligence, vol. 41, pp. 207-222, 2015. [18] M. Aharon, O. Anava, N. Avigdor-Elgrabli, D. Drachsler-Cohen, S. Golan, & O. Somekh, “ExcUseMe: Asking Users to Help in Item Cold-Start Recommendations,” In Proceedings of the 9th ACM Conference on Recommender Systems, pp. 83-90, ACM, 2015. [19] I. Fernández-Tobías, M. Braunhofer, M. Elahi, F. Ricci, & Cantador, “I Alleviating the new user problem in collaborative filtering by exploiting personality information,” User Modeling and User-Adapted Interaction, pp. 1-35, 2015. [20] I. Barjasteh, R. Forsati, F. Masrour, A. H. Esfahanian, & H. Radha, “Cold-start item and user recommendation with decoupled completion and transduction,” In Proceedings of the 9th ACM Conference on Recommender Systems, pp. 91-98, ACM, 2015. [21] A. L. V. Pereira, & E. R. Hruschka, “Simultaneous co-clustering and learning to address the cold start problem in recommender systems,” Knowledge-Based Systems, vol. 82, pp. 11-19, 2015. [22] GroupLens Research, MovieLens, 2014, Available at: http://grouplens.org/datasets/ movielens/. [23] H. J. Ahn, “A new similarity measure for collaborative filtering to alleviate the new user cold-starting problem,” Information Sciences, vol. 178, no. 1, pp. 37-51, 2008. [24] B. Alatas, E. Akin, & A. B. Ozer, “Chaos embedded particle swarm optimization algorithms,” Chaos, Solitons & Fractals, vol. 40, no. 4, pp. 1715-1734, 2009. [25] P. J. Rousseeuw, ‘‘Silhouettes: A graphical aid to the interpretation and validation of cluster analysis’’ Journal of computational and applied mathematics, vol. 20, pp. 53-65, 1987.