Fake News Detection Models and Performances
Pramod Sekharan Nair1, Kefyalew Aragaw Hailekidan2, Vinitha Gangadharan Nair3
1Pramod Sekharan Nair*, Computer Science and Engineering, Medi-Caps University, Indore, India.
2Kefyalew Aragaw Hailekidan, Computing Technology, Aksum University, Axum, Ethiopia.
3Vinitha Gangadharan Nair, Management Studies, Medi-Caps University, Indore, India.
Manuscript received on November 20, 2019. | Revised Manuscript received on December 15, 2019. | Manuscript published on December 30, 2019. | PP: 3754-3757 | Volume-9 Issue-2, December, 2019. | Retrieval Number: B2928129219/2019©BEIESP | DOI: 10.35940/ijeat.B2928.129219
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Fake News detection is a hard problem for decades after the advent of social media. As misinformation, so called fake news continues to be rapidly distributing on internet, the reality has becoming increasingly shaped by false information. Time after time we have consumed or being exposed to inaccurate information. The last few years have been talking about guarding against misinformation and not progressed much in this direction. The social media is one of the medium where the fake news spreads so rapidly and impact many in a lesser span of time. Machine Learning and Natural Language processing are the core techniques to detect the fake news and stopping from spreading on social media. Many researchers putting their effort in this new challenge to curb down. This paper provides an insight on feature extraction techniques used for fake news detection on soft media. Text feature extraction works with extracting the document information which represent the whole document without loss of the sole information but words which are considered irrelevant were ignored for the purpose of improving the accuracy. Term Frequency Inverse Document Frequency (TF-IDF), BoW(Bag of Words) are some of the important techniques used in text feature extraction. These techniques are discussed with their significance in this paper. One of the important approach, Automated Readability Index is used to test the readability of the text to build the model also discussed in this paper. This paper will play a significant role for the researchers who are interested in the area of fake news Identification.
Keywords: Fake News, Social Media, Fake News Detection, Fake News Identification, TF-IDF Approach, Bag of Words.