Different Absorption from the Same Sharing: Sifted Multi-task Learning for Fake News Detection

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 4644–4653, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 4644 Different Absorption from the Same Sharing: Sifted Multi-task Learning for Fake News Detection Lianwei Wu, Yuan Rao, Haolin Jin, Ambreen Nazir, Ling Sun Lab of Social Intelligence and Complexity Data Processing School of Software Engineering, Xi’an Jiaotong University, Xi’an, 710049, China {stayhungry,jinhaolin,ambreen.nazir,sunling}@stu.xjtu.edu.cn raoyuan@mail.xjtu.edu.cn Abstract Recently, neural networks based on multi- task learning have achieved promising perfor- mance on fake news detection, which focus on learning shared features among tasks as com- plementary features to serve different tasks. However, in most of the existing approaches, the shared features are completely assigned to different tasks without selection, which may lead to some useless and even adverse fea- tures integrated into specific tasks. In this paper, we design a sifted multi-task learning method with a selected sharing layer for fake news detection. The selected sharing layer adopts gate mechanism and attention mecha- nism to filter and select shared feature flows between tasks. Experiments on two public and widely used competition datasets, i.e. Ru- mourEval and PHEME, demonstrate that our proposed method achieves the state-of-the-art performance and boosts the F1-score by more than 0.87%, 1.31%, respectively. 1 Introduction In recent years, the proliferation of fake news with various content, high-speed spreading, and exten- sive influence has become an increasingly alarming issue. A concrete instance1 was cited by Time Magazine in 2013 when a false announcement of Barack Obama’s injury in a White House explosion “wiped off 130 Billion US Dollars in stock value in a matter of seconds”. Other examples, an analysis of the US Presidential Election in 2016 (Allcott and Gentzkow, 2017) revealed that fake news was widely shared during the three months prior to the election with 30 million total Facebook shares of 115 known pro-Trump fake stories and 7.6 million of 41 known pro-Clinton fake stories. Therefore, automatically detecting fake news has attracted sig- 1http://business.time.com/2013/04/24/how-does-one- fake-tweet-cause-a-stock-market-crash/ ������� ��������������������������������� ������� ������� ������� ���������������������������� ������� ������� Figure 1: Two schemes for sharing features among tasks. Red circles and blue boxes represent the task- specific features, while the red and blue triangles mean shared features that benefit Task A and Task B, respec- tively. nificant research attention in both industries and academia. Most existing methods devise deep neural net- works to capture credibility features for fake news detection. Some methods provide in-depth analy- sis of text features, e.g., linguistic (Conroy et al., 2015), semantic (Yang et al., 2012), emotional (Wang et al., 2015), stylistic (Potthast et al., 2017), etc. On this basis, some work additionally ex- tracts social context features (a.k.a. meta-data features) as credibility features, including source- based (Castillo et al., 2011), user-centered (Long et al., 2017), post-based (Wang, 2017) and network- based (Ruchansky et al., 2017), etc. These methods have attained a certain level of success. Addition- ally, recent researches (Thorne et al., 2017; Dungs et …

Content truncated to 500 words