Loading…

A review on fake news and fake user profile detection

We used to rely on reputable newspapers and mainstream media sources for our daily news since they are held to the highest ethical standards. Online news has become more significant to netizens who are looking for information. The quick rise in popularity of social media has changed how people consu...

Full description

Saved in:
Bibliographic Details
Main Authors: Rampurkar, Vyankatesh, Thirupurasundari, Dasarada
Format: Conference Proceeding
Language:English
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We used to rely on reputable newspapers and mainstream media sources for our daily news since they are held to the highest ethical standards. Online news has become more significant to netizens who are looking for information. The quick rise in popularity of social media has changed how people consume news. The internet has created a new channel for the publication and consumption of news pieces with lax editorial standards. But the great majority of these reports are incorrect and misleading; they often use sensationalism, satire, propaganda, and misleading headlines. People could find it challenging to distinguish between fake news pieces and legitimate news reports because of how similar they seem. The main source of bogus news on social media is fraudulent profiles. The goal of fake user and fake user profile detection is to identify and distinguish real people from fraudulent or misleading entities on online platforms. Before being used as an input for training the models, the web-based dataset collection undergoes preprocessing. The dataset is divided into training and testing portions after cleaning. After the dataset has been split into training and testing sets, TF-IDF is applied. To get the best accuracy for the given dataset, a variety of machine learning approaches, including Naive Bayes, Decision trees, Passive-aggressive, and Logistic Regression, are continuously taught. To choose the most accurate and appropriate model, the accuracy levels of all the machine learning models are compared. The project’s output establishes the foundation for machine learning-based social network fraud detection of fake news and profiles. Naive Bayes is utilized for many tasks due to its adaptability, effectiveness, and simplicity in a variety of scenarios. Decision trees are another kind of machine learning technique that is used for the identification of fake news because of their simplicity, interpretability, and effectiveness when processing text data. To maintain trust in online news sources, it’s imperative to recognize phoney user accounts and news. We can lessen the negative effects of false information and user accounts that are misleading by utilizing sophisticated techniques based in social network analysis and artificial intelligence. This will help to create a more dependable and trustworthy online environment.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0218812