Deep Learning based Real and Fake news Detector

Project Title

Deep Learning based Real and Fake news Detector

Project Area of Specialization Artificial IntelligenceProject Summary

In the Era of social media and the internet, everyone worldwide is connected. Any happening in one corner of the world could be seen by people in another corner. News shared on social media spread in a matter of seconds which indeed has made this world a global village. We are aware of progress, technology academia, and scientific advancements in no time.

despite the fact that people use social media and the internet for learning, running their virtual businesses around the globe, and freelancing. fake news shared on social media spreads hustle and misguides people to a great extent. most of the time people don't look at the source of the news and believe everything they watch on social media.

many rumors and accidents are been reported only because fake news was spread/shared on social media. This project proposed to develop a web/mobile-based application. which will fetch news from all of your selected sources and will differentiate between real and fake news using Deep Learning Convolutional Neural Network. we will design a system and any video/textual news would be evaluated for authenticity.

Project Objectives

This study aims to facilitate social media users with only authentic news and reduce fake news spreading, which causes misunderstanding and chaos. Internet is now accessible to everyone despite education level, class, or intellectual level.  people tend to believe what is shared.

What our system proposed:

1. Identification of fake or real news
2. Identifying valid and suspicious origins for news
3. providing accurate/ facts about the romers.

Project Implementation Method

This project is divided mainly into three phases

Phase 1: Natural Language processing (English & Urdu Text)

In phase 1, through Natural language processing, We will collect data on fake and real news (mainly for Urdu and English language), annotate/ prepare data set for training machine learning classifiers, and then perform supervised learning to train & test the accuracy of our classifier's accuracy.

Phase 2: DeepLearning for Video based news

In this phase, we will use deep learning for identifying real or fake news shared in the form of videos/images. for this purpose, we will use deep learning architecture and will train our neural network. In this process, we will gather data set of real and fake news videos and will train our model on this dataset. testing would be performed later for the evaluation of data.

Phase 3: Mobile/web application development

In this phase, we will combine both classifiers so that users can use the application and can check news authentication for both textual and video/image-based news.

Benefits of the Project

Nationa that advance timlely can sustain and survive. Today, world has adopted automation and technology in every domain of life. In Pakistan, use of smartphones and social media is common and any news spread from any source spread faster than fire. Identifying fake/real news can save us from several mishappenings and a chaos could be timely stopped. 

Otherthan this deeplearning/Artificial intelligence is helping in every aspect from identifying disease to business insights.

Mainy we offer following benefits:

1. Natural Language processing for Urdu (Throughout the world language mostly worked on are English, Hindi and Arabic. Urdu is still needed to be explored and processed using Machine learning/Natural Language processing)
2. Identification of fake news will help society and the community to stop any rumor from spreading.
3. deep Learning and use of advance technology will upgrade problem solving

Technical Details of Final Deliverable

After completion of the project, we will be liable to provide:

Software Requirement specification Document:

All details about the projects, requirements(from the user's perspective), Dataset acquired and used and classifiers trained.

Mobile/Web Application

The final product of this project would be a platform independent application that will provide users to identify real of fake news based on two options:

• either they can provide a link to a textual news
• or they can provide a link to a video/image of the news

Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther Industries Media Core Technology Artificial Intelligence(AI)Other Technologies Big DataSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	34000
Sapphire Pulse RX 570 4GB GDDR5 Graphics Card (Used)	Equipment	1	34000	34000