FAKE NEWS DETECTION ON SOCIAL MEDIA

Tracking #: 648-1628

Authors:
NameORCID
Masroor ShahORCID logo https://orcid.org/0000-0002-1208-3822


Submission Type: 

Position Paper

Abstract: 

The media is the plural form of the term “media”. Many of the things you read online, especially in social media feeds, seem to be true, but this is not the case. Fake news stories may look like trusted websites, or use similar names and URLs from reputable news organizations to deceive people. In existing research work, there is no proper solution proposed for fake news detection. It is also important to perform but supervised and unsupervised model for classification as well as real-time detection. Need a step by step method that sufficient for detection and classification. We need to propose a faster model that gives the best result from existing work by comparing the performance of it. In this paper used k-mean and artificial neural networks (ANN) machine learning model for fake news detection as well as classification. In results, the report shows precision (PR) is 87% and recall (RE) is 87% accuracy. The average accuracy (AC) of our model is 87% which is extremely awesome. The average accuracy represents the F1 87% score and receiver operating characteristic (ROC) is 94.7%. As a result, in evaluation Mean Absolute Error: 0.1429 and Root Mean Squared Error: 0.3108. with 87% average accuracy. By comparing to existing research work, we improved defect determination accuracy of existing research which was 82%.

Manuscript: 

Tags: 

  • Reviewed

Data repository URLs: 

None

Date of Submission: 

Friday, July 17, 2020

Date of Decision: 

Monday, July 27, 2020

Decision: 

Reject (Pre-Screening)