Efficient collaborative filtering recommendations with multi-channel feature vectors
- PDF / 1,289,770 Bytes
- 8 Pages / 595.276 x 790.866 pts Page_size
- 67 Downloads / 208 Views
ORIGINAL ARTICLE
Efficient collaborative filtering recommendations with multi‑channel feature vectors Heng‑Ru Zhang1,2 · Fan Min1 · Zhi‑Heng Zhang1,2 · Song Wang3 Received: 21 October 2016 / Accepted: 26 February 2018 © Springer-Verlag GmbH Germany, part of Springer Nature 2018
Abstract Feature vectors and similarity measures are the two key issues of most existing collaborative filtering (CF) algorithms. In item-based CF algorithms, the feature vector is often defined as the ratings of all users for a given item. For a recommender system with n users, m items, and c ratings, the length of the feature vector is n; hence, the time complexity of the similarity computation is O(n). Consequently, the overall time complexity is O(m2 n2 ) , which may be computationally prohibitive for recommender systems with millions of users. In this paper, we define the multi-channel feature vector (MCFV), which is a vector of channel length c, and calculate the similarity between items using the respective MCFVs. Each element of an MCFV corresponds to the number of users with respective ratings for the item. The time complexity for the similarity computation is O(c), and the overall time complexity is O(m2 nc) when the k-nearest neighbors and weighted average algorithms are used. Experiments were conducted on four movie recommender systems, where n ranges from a few hundred to half a million, and c is five. Results show that the recommendation algorithms using our new similarity measure are significantly faster than their counterparts without sacrificing prediction accuracy in terms of mean absolute error and root mean square error. Keywords Collaborative filtering · Feature-vector similarity · Multi-channel · Recommender system
1 Introduction Collaborative filtering (CF) is one of the earliest and most successful algorithms that underlie recommender systems [1–4]. A latent assumption of the CF approach is that, in a social network, those who have agreed in the past tend to agree again in the future [5]. Some recommender systems present nominal recommendations evaluated by accuracy [6, 7] or cost [8]. Others present a numerical prediction of This work is supported in part by the National Natural Science Foundation of China (Grants 61379089, 41604114), the Innovation and Entrepreneurship Foundation of Southwest Petroleum University (Grant SWPUSC16-003) and the Natural Science Foundation of the Department of Education of Sichuan Province (Grant 16ZA0060). * Fan Min [email protected] 1
School of Computer Science, Southwest Petroleum University, Chengdu 610500, China
2
School of Science, Southwest Petroleum University, Chengdu 610500, China
3
Box Inc., 900 Jefferson Ave, Redwood City, CA 94063, USA
ratings [1, 9] evaluated by the mean absolute error (MAE) [10] and root mean square error (RMSE) [11]. Still others present a recommendation sequence (order) evaluated by the standard information retrieval measures half-life [12] and discounted-cumulative-gain [13]. In these different approaches, item-based recommenders ar
Data Loading...
