JISE


  [ 1 ] [ 2 ] [ 3 ] [ 4 ] [ 5 ] [ 6 ] [ 7 ] [ 8 ] [ 9 ] [ 10 ] [ 11 ] [ 12 ] [ 13 ] [ 14 ] [ 15 ] [ 16 ]


Journal of Information Science and Engineering, Vol. 37 No. 5, pp. 1177-1186


Hyperpartisan News Classification with ELMo and Bias Feature


GERALD KI WEI HUANG AND JUN CHOI LEE
Faculty of Computer Science and Information Technology
Universiti Malaysia Sarawak
Kota Samarahan, 94300 Malaysia
E-mail: 19020037@siswa.unimas.my; jclee@unimas.my


Hyperpartisan news is a kind of news riddled with twisted, untruthful, and often extremely one-sided. This kind of news can spread more successfully than the others. One of the obvious traits of hyperpartisan news content is that it can mimic regular news articles. Most are favour fake news detection algorithms, and there is less research conducted for hyperpartisan news. This research aims to perform classification on the hyperpartisan news using ELMo and bias features. ELMo was used to develop a classification model to perform classification on the BuzzFeed Webis News Corpus dataset. The model uses ELMo embedding with bias word score generated from bias lexicon to train a deep learning model using Tensorflow and Keras. We had compared the final result with two proposed baseline models that utilized ELMo from other research. The discussion section further investigated the contribution of ELMo and bias feature in the hyperpartisan task.


Keywords: natural language processing, classification, hyperpartisan, ELMo, bias detection

  Retrieve PDF document (JISE_202105_13.pdf)