Title :
Session-Based Recommendation -- Case Study on Tencent Weibo
Author :
Chen-Ling Chen ; Chia-Hui Chang
Author_Institution :
Dept. of Comput. Sci. & Inf. Eng., Nat. Central Univ., Jhongli, Taiwan
Abstract :
Ten cent Weibo is one of the largest micro-blogging websites in China. There are more than 200 million registered users on Ten cent Weibo, generating over 40 million messages each day. Recommending appealing items to users is a mechanism to reduce the risk of information overload. The task of this paper is to predict whether or not a user will follow an item that has been recommended to the user by Ten cent Weibo. This paper contains two parts: predicting users´ interests and distinguish whether the user is busy or available to browse recommended items. We apply several model based collaborative filtering as well as content-based filtering to capture users´ interests. Besides, we built an occupied model to distinguish users´ state and combined with recommendations methods as the final result. In the paper, we used session-based hamming loss as performance measure. The hamming loss was greatly reduced (40%) with occupied model from 0.187 to 0.13.
Keywords :
collaborative filtering; content-based retrieval; human computer interaction; recommender systems; social networking (online); China; Tencent Weibo; content-based filtering; information overload risk reduction; microblogging websites; model based collaborative filtering; session-based hamming loss; session-based recommendation; user interest prediction; Collaboration; Data models; Filtering; Frequency modulation; Predictive models; Testing; Training data; collaborative filtering; factorization machine; matrix factorization; social network recommendation;
Conference_Titel :
Technologies and Applications of Artificial Intelligence (TAAI), 2013 Conference on
Conference_Location :
Taipei
Print_ISBN :
978-1-4799-2528-5
DOI :
10.1109/TAAI.2013.49
