NileTMRG at SemEval-2017 Task 8: Determining Rumour and Veracity Support for Rumours on Twitter.

Proceedings of the 11th International Workshop on Semantic Evaluations (SemEval-2017), pages 470–474, Vancouver, Canada, August 3 - 4, 2017. c⃝2017 Association for Computational Linguistics NileTMRG at SemEval-2017 Task 8: Determining Rumour and Veracity Support for Rumours on Twitter Omar Enayet and Samhaa R. El-Beltagy Center for informatics sciences Nile University Giza, Egypt omar.enayet@gmail.com, samhaa@computer.org Abstract This paper presents the results and conclu- sions of our participation in SemEval-2017 task 8: Determining rumour veracity and support for rumours. We have participated in 2 subtasks: SDQC (Subtask A) which deals with tracking how tweets orient to the accuracy of a rumourous story, and Ve- racity Prediction (Subtask B) which deals with the goal of predicting the veracity of a given rumour. Our participation was in the closed task variant, in which the pre- diction is made solely from the tweet it- self. For subtask A, linear support vector classification was applied to a model of bag of words, and the help of a naïve Bayes classifier was used for semantic fea- ture extraction. For subtask B, a similar approach was used. Many features were used during the experimentation process but only a few proved to be useful with the data set provided. Our system achieved 71% accuracy and ranked 5th among 8 systems for subtask A and achieved 53% accuracy with the lowest RMSE value of 0.672 ranking at the first place among 5 systems for subtask B. 1 Introduction Over the past 15 years, and in a gradual man- ner, social media has started to become a main source of news. However, social media has also become a ripe ground for rumours, spreading them in a matter of a few minutes. A rumour is de- fined as a claim that could be true or false. False rumours may greatly affect the social, economic and political stability of any society around the world, hence the need for tools to help people, es- pecially journalists, analyze the spread of rumours and their effect on the society as well as determine their veracity. Twitter is a famous social media platform ca- pable of spreading breaking news, thus most of rumour related research uses Twitter feed as a ba- sis for research. SemEval (Semantic Evaluation) is an ongoing series of evaluations of computational semantic analysis systems. Task 8 (RumourEval) (Derczyn- ski, et al. (2017)) is one of 12 tasks presented in SemEval 2017. This paper describes the system that we have used to participate in this task. The task consists of 2 subtasks: SDQC (Subtask A) which has the objective of tracking how other tweets orient to the accuracy of a rumourous story, and Veracity Prediction (Subtask B) for which has the goal to predict the veracity of a given rumour. Task B has two variants: an open variant and a closed one. We have only participated in the closed variant, in which the prediction should be made solely from the tweet itself. Scientific literature related to rumours on social media has started to emerge …

Content truncated to 500 words