Identifying and Tracking Sentiments and Topics from Social Media Texts during Natural Disasters

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 527–533 Copenhagen, Denmark, September 7–11, 2017. c⃝2017 Association for Computational Linguistics Identifying and Tracking Sentiments and Topics from Social Media Texts during Natural Disasters Min Yang1,4, Jincheng Mei2, Heng Ji3, Wei Zhao4, Zhou Zhao5, Xiaojun Chen1∗ 1 Shenzhen University 2 University of Alberta 3 Rensselaer Polytechnic Institute 4 Tencent AI Lab 5 Zhejiang University Abstract We study the problem of identifying the topics and sentiments and tracking their shifts from social media texts in dif- ferent geographical regions during emer- gencies and disasters. We propose a location-based dynamic sentiment-topic model (LDST) which can jointly model topic, sentiment, time and Geolocation information. The experimental results demonstrate that LDST performs very well at discovering topics and sentiments from social media and tracking their shifts in different geographical regions during emergencies and disasters1. 1 Introduction Social media has become pervasive in our daily life, and it is a great way to spread important in- formation efficiently. Using social media (e.g., Twitter, Facebook, Pinterest), people can conve- niently inform others and express support during emergencies and disasters. Nowadays, the social media is keeping produc- ing a huge amount of information. Unlike before, people are not only interested in identifying the static topics and sentiments from given texts, but also, perhaps more concerned with tracking the evolution of topics and sentiments among differ- ent geographical regions. On the one hand, this new requirement can be very helpful, especially in the case of emergencies, such as pre-disaster preparation and post-disaster relief in local natural disasters (Beigi et al., 2016). For example, peo- ple’s sentiments on the topics related to medical and rescue may guide the management and distri- bution of emergency supplies. On the other hand, ∗ Corresponding author (xjchen@szu.edu.cn) 1Data and codes are available at https://goo.gl/uee3QK the existing models do not take the temporal evo- lution and the impact of location over topics and sentiments into consideration, which makes them unable to fulfill the new requirement of tracking the evolution of topics and sentiments in different geographical places. In this paper, we aim to identify the topics and sentiments and track their shifts in different ge- ographical regions during emergencies and dis- asters. We are inspired by several observations. First, people are interested in not only the overall sentiment or topic distribution of the documents but also the sentiments towards specific topics. For example, a person may be happy with that the dis- aster passed away, but at the meanwhile he/she may be unsatisfied with the post-disaster relief. Second, most existing sentiment-topic models ig- nore the temporal evolution of topics and senti- ment in a time-variant data corpus such as the Twitter stream. There are strong evidences which indicate that people’s attitudes toward a disaster will gradually change over time with the distri- bution of emergency supplies (Beigi et al., 2016; Caragea et al., 2014; Mandel et al., 2012). Third, people in different places tend to have different opinions towards particular topics. …

Content truncated to 500 words