Mining Evidences for Concept Stock Recommendation

Proceedings of NAACL-HLT 2018, pages 2103–2112 New Orleans, Louisiana, June 1 - 6, 2018. c⃝2018 Association for Computational Linguistics Mining Evidences for Concept Stock Recommendation Qi Liu and Yue Zhang Singapore University of Technology and Design, 8 Somapah Road, Singapore 487372 {qi liu, yue zhang}@sutd.edu.sg Abstract We investigate the task of mining relevant stocks given a topic of concern on emerg- ing capital markets, for which there is lack of structural understanding. Deep learning is leveraged to mine evidences from large scale textual data, which contain valuable market in- formation. In particular, distributed word sim- ilarities trained over large scale raw texts are taken as a basis of relevance measuring, and deep reinforcement learning is leveraged to learn a strategy of topic expansion, given a small amount of manually labeled data from fi- nancial analysts. Results on two Chinese stock market datasets show that our method outper- forms a strong baseline using information re- trieval techniques. 1 Introduction Stock prices are affected by events. For exam- ple, recent announcement of a state plan to build a new economic region, Xiong’an near Beijing, by the Chinese government has led to the rise of hundreds of stocks, which can directly or indi- rectly benefit from the plan. As a second example, the winning of a lawsuit against IP (Intellectual Property) breach can strengthen investors’ confi- dence on technological and entertainment com- panies. We refer to the topics or themes of such events (e.g. Xiong’an and IP) as concepts and their relevant stocks as concept stocks. Given a news event, it can be highly useful for investors to find a list of relevant concept stocks for making invest- ment decisions. For popular concepts, lists of relevant concept stocks can be found from analyst reports from financial websites. On the other hand, concepts are dynamic and flexible. In addition, insights can be relatively scarce for emerging capital markets, such as the Chinese market, which had been closed to foreign investments before 2015. It is therefore 雄安 Xiong’an 白洋淀 Baiyangdian 安新县 Anxin county 容城 Rongcheng 保定市 Baoding3city 河北省 Hebei3province 冀东 East Hebei 京津冀 Beijing-Tianjin- Hebei 河北省政府 Hebei3government 石家庄市 Shijiazhuang3city 京津冀一体化 Coordinated3development3 of3Beijing-Tianjin-Hebei 经济带 Economic district 滨海新区 Binhai new3area Figure 1: Concept relatedness. A ⇒ B indicates that B is one of the 3 most related concepts to A. a challenging research question how to automat- ically find out potentially relevant stocks given a topic of interest, from a large market of multi- thousand equities. Intuitively, evidences between concepts and stocks exist in text documents over the Internet. For example, news articles report events and com- panies involved. In addition, company filings such as annual/quarter reports contain factual knowl- edge about stocks, which can also be useful back- ground information. For example, knowing that a company invests heavily on research is useful for correlating the company with IP-protection laws. Such evidence-mining process can involve mul- tiple steps. As shown in Figure 1, starting from concept, Xiong’an, one might learn that the new economical region is located …

Content truncated to 500 words