Topic Knowledge Acquisition and Utilization for Machine Reading Comprehension in Social Media Domain

Topic Knowledge Acquisition and Utilization for Machine Reading Comprehension in Social Media Domain Zhixing Tian1,2, Yuanzhe Zhang1, Kang Liu1,2, Jun Zhao1,2, 1 National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China 2 University of Chinese Academy of Sciences, Beijing, 100049, China {zhixing.tian, yzzhang, kliu, jzhao}@nlpr.ia.ac.cn Abstract In this paper, we focus on machine reading comprehension in social media. In this domain, one normally posts a message on the assumption that the readers have specific background knowl- edge. Therefore, those messages are usually short and lacking in background information, which is different from the text in the other domain. Thus, it is difficult for a machine to understand the messages comprehensively. Fortunately, a key nature of social media is clustering. A group of people tend to express their opinion or report news around one topic. Having realized this, we propose a novel method that utilizes the topic knowledge implied by the clustered messages to aid in the comprehension of those short messages. The experiments on TweetQA datasets demonstrate the effectiveness of our method. 1 Introduction As an increasing number of people share and obtain information from social media, social media is now becoming an important real-time information source. The unprecedented volume, variety of user- generated content, and the user interaction network constitute new opportunities for understanding social behavior and building socially intelligent systems. It is important and challenging to teach a machine to automatically understand the content presented in social media. Although considerable progress has been made in the task of Machine Reading Comprehension (MRC), most of the previous works only focus on the comprehension of the other domains, such as news (Hermann et al., 2015; Chen et al., 2016), story (Sachan et al., 2015; Narasimhan and Barzilay, 2015) and Wikipedia (Seo et al., 2016; Yang et al., 2018), and there are very few works addressing the problem of social media MRC. Table 1 shows an example of social media comprehension. Different from the other domains, in social media domain, one normally posts a message on the assumption that the readers have specific background knowledge. Those messages are generally short and contain lim- ited contextual information as shown in the table. Thus, it is difficult for a machine to understand them thoroughly based only on the text itself. In this example, if only look at the message, without background knowledge, a machine reader would be puzzled about its topic and could not answer the question. Message: in case you missed it, here’s your first look at jamie bell’s the thing: # fantasticfour empire magazine (@ empiremagazine) april 9, 2015 Question: what movie is being shared here? Answer: fantastic four Table 1: An example of machine reading comprehension in social media domain. To obtain background knowledge for a machine reader, one feasible method is to introduce external knowledge from the knowledge base like ConceptNet (Liu and Singh, 2004) and WordNet (Fellbaum, 1998), as the previous works do in other domains (Lin et al., 2017; Wang et …

Content truncated to 500 words