Steve Martin at SemEval-2019 Task 4: Ensemble Learning Model for Detecting Hyperpartisan News

Proceedings of the 13th International Workshop on Semantic Evaluation (SemEval-2019), pages 990–994 Minneapolis, Minnesota, USA, June 6–7, 2019. ©2019 Association for Computational Linguistics 990 Steve Martin at SemEval-2019 Task 4: Ensemble Learning Model for Detecting Hyperpartisan News Youngjun Joo Department of Computer Engineering Yonsei University, Seoul, Korea yj.joo@yonsei.ac.kr Inchon Hwang Department of Computer Engineering Yonsei University, Seoul, Korea ich0103@yonsei.ac.kr Abstract This paper describes our submission to task 4 in SemEval 2019, i.e., hyperpartisan news detection. Our model aims at detecting hy- perpartisan news by incorporating the style- based features and the content-based features. We extract a broad number of feature sets and use as our learning algorithms the GBDT and the n-gram CNN model. Finally, we apply the weighted average for effective learning be- tween the two models. Our model achieves an accuracy of 0.745 on the test set in subtask A. 1 Introduction The proliferation of misleading information in the media has made it challenging to identify trust- worthy news sources, thus increasing the need for fake news detection tools able to provide in- sight into the reliability of news contents. Since the spread of fake news is causing irreversible re- sults, near-real-time fake news detection is cru- cial. However, knowledge-based and context- based approaches to fake news detection can only be applied after publication; they may not be fast enough (Potthast et al., 2017). As a practical alternative, style-based ap- proaches try to detect fake news by capturing the manipulators in the writing style of news con- tent. This approach captures style signals that can indicate a decreased objectivity of news content and thus the potential to mislead consumers, such as hyperpartisan style. Hyperpartisan style repre- sents extreme behavior in favor of a particular po- litical party, which often correlates with a strong motivation to create fake news. Linguistic-based features can be applied to detect hyperpartisan ar- ticles (Potthast et al., 2017). Deep network mod- els, such as convolution neural networks (CNN), applied to classify fake news detection (Wang, 2017). In this paper, we employ the stylometry- based approach and N-gram CNN model for de- tecting hyperpartisan news. 2 System Overview For this task, we extract a broad number of fea- tures from the training data and then apply the classifier model to make predictions. Our system employs a gradient boosting decision tree (GBDT) model and N-gram CNN model. In subsequent sections, we describe data preprocessing, feature engineering and learning algorithms. 2.1 Data Preprocessing Before applying the models, we need to do some transforming tasks of the article texts (i.e., xml parsing, text tokenizing, stemming, lemmatization, and removing stopwords) and extracting tasks of the internal and external links for each article. Apart from these tasks, we construct the bias do- main dictionary from the mediabiasfactcheck site 1 to check the bias on the external linked domain in the article. For this ends, we crawled the top-level domain information from the sites corresponding to the five categories associated with hyperparti- san (e.g., Left, Center, Least Biased, Right-center Bias, and …

Content truncated to 500 words