Neural News Recommendation with Topic-Aware News Representation

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 1154–1159 Florence, Italy, July 28 - August 2, 2019. c⃝2019 Association for Computational Linguistics 1154 Neural News Recommendation with Topic-Aware News Representation Chuhan Wu1, Fangzhao Wu2, Mingxiao An3, Yongfeng Huang1, and Xing Xie2 1Department of Electronic Engineering, Tsinghua University, Beijing 100084, China 2Microsoft Research Asia, Beijing 100080, China 3University of Science and Technology of China, Hefei 230026, China wuch15@mails.tsinghua.edu.cn, {fangzwu,xingx}@microsoft.com, anmx@mail.ustc.edu.cn, yfhuang@tsinghua.edu.cn Abstract News recommendation can help users find in- terested news and alleviate information over- load. The topic information of news is critical for learning accurate news and user represen- tations for news recommendation. However, it is not considered in many existing news rec- ommendation methods. In this paper, we pro- pose a neural news recommendation approach with topic-aware news representations. The core of our approach is a topic-aware news en- coder and a user encoder. In the news encoder we learn representations of news from their ti- tles via CNN networks and apply attention net- works to select important words. In addition, we propose to learn topic-aware news repre- sentations by jointly training the news encoder with an auxiliary topic classification task. In the user encoder we learn the representations of users from their browsed news and use at- tention networks to select informative news for user representation learning. Extensive exper- iments on a real-world dataset validate the ef- fectiveness of our approach. 1 Introduction Online news platforms such as Google News and MSN News have attracted hundreds of millions of users to read news online (Das et al., 2007; Lavie et al., 2010). Massive news are generated everyday, making it impossible for users to read all news to find their interested content (Phelan et al., 2011). Thus, personalized news recommendation is very important for online news platforms to help users find their interested news and alleviate infor- mation overload (IJntema et al., 2010). Learning accurate representations of news and users is critical for news recommendation (Wu et al., 2019b,a). Several deep learning based meth- ods have been proposed for this task (Okura et al., 2017; Wang et al., 2018; Kumar et al., 2017; Khat- tar et al., 2018; Zheng et al., 2018). For example, Title Topic James Harden's incredible heroics lift Rockets over Warriors Sports These Are Some of The Safest Airlines in the World Travel Weekend snowstorm forecast from Midwest to East Coast Unlabeled Figure 1: Three example news articles. Okura et al. (2017) proposed to learn news repre- sentations from news bodies via denoising auto- encoders, and learn user representations from the representations of their browsed news via a gated recurrent unit (GRU) network. Wang et al. (2018) proposed to learn news representations from news titles via a knowledge-aware convolutional neu- ral network (CNN), and learn user representations from news representations using the similarity be- tween candidate news and browsed news. How- ever, these methods do not take the topic informa- tion of news into consideration. Our work is motivated by the …

Content truncated to 500 words