Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/rishabhmathur06/fake-news-detection


https://github.com/rishabhmathur06/fake-news-detection

bert-model machine-learning matplotlib-pyplot natural-language-processing neural-network numpy pandas python3 sklearn tensorflow transformer

Last synced: about 22 hours ago
JSON representation

Awesome Lists containing this project

README 

        
# Fake-News-Detection



Fake news is the intentional broadcasting of false or misleading claims as news, where the statements are purposely deceitful.



Newspapers, tabloids, and magazines have been supplanted by digital news platforms, blogs, social media feeds, and a plethora of mobile news applications. News organizations benefitted from the increased use of social media and mobile platforms by providing subscribers with up-to-the-minute information.



Consumers now have instant access to the latest news. These digital media platforms have increased in prominence due to their easy connectedness to the rest of the world and allow users to discuss and share ideas and debate topics such as democracy, education, health, research, and history. Fake news items on digital platforms are getting more popular and are used for profit, such as political and financial gain.



How Big is this Problem?

Because the Internet, social media, and digital platforms are widely used, anybody may propagate inaccurate and biased information. It is almost impossible to prevent the spread of fake news. There is a tremendous surge in the distribution of false news, which is not restricted to one sector such as politics but includes sports, health, history, entertainment, and science and research.



The Solution

It is vital to recognize and differentiate between false and accurate news. One method is to have an expert decide, and fact checks every piece of information, but this takes time and needs expertise that cannot be shared. Secondly, we can use machine learning and artificial intelligence tools to automate the identification of fake news.



Online news information includes various unstructured format data (such as documents, videos, and audio), but we will concentrate on text format news here. With the progress of machine learning and Natural language processing, we can now recognize the misleading and false character of an article or statement.



Several studies and experiments are being conducted to detect fake news across all mediums.



Our main goal for this tutorial is:

Ezoic



Explore and analyze the Fake News dataset.

Build a classifier that can distinguish Fake news with as much accuracy as possible.



Here is the table of content:



Introduction

How Big is this Problem?

The Solution

Data Exploration

Distribution of Classes

Data Cleaning for Analysis

Explorative Data Analysis

Single-word Cloud

Most Frequent Bigram (Two-word Combination)

Most Frequent Trigram (Three-word Combination)

Building a Classifier by Fine-tuning BERT

Data Preparation

Tokenizing the Dataset

Loading and Fine-tuning the Model

Model Evaluation

Appendix: Creating a Submission File for Kaggle

Conclusion