Fake News detector

The fake news challenge is a global problem from which neither a state nor an individual is spared. It is from the destructive issues that defy Egypt nowadays, representing an obstacle in the country's growth in a peaceful manner which represents the fake news spread rate for 2020 (from January to April). Social media has become a part of our day to-day life and has become a significant source of information. Still, MIT University has released a study on the extent of the spread of fake news in the world of social media from 2006 to 2017. The results were as follows: 80% of the news on social media is fake news, in addition to that, the rate of spreading fake news reaches 70% compared to the real one, which leads to many conflicts and disputes. Moreover, fake recruitment advertisements spread on various websites on the Internet, which leads to many people losing their private data. Many banks lose their reputation and sometimes their partnerships with distinguished companies and clients due to untruthful news about them. Moving on to a lot of insincere medical news, the last of which is about the emerging coronavirus, as some people exploited this situation and spread news about harmful and fake preventive treatment in order to collect money. And finally, fake news about the sports community. Some solutions were presented, but they did not solve all defaults of fake news, but the most prior solution that attracted my attention was political news detection, as it contains political news from 2005 to 2018 and has some advantages:

1- Dataset is large.

2- It is characterized by high accuracy

that is almost 100%. However, it also has some important flaws:

1- It reveals only one type of news

2- Depending on the environment surrounding us, not many of the technology users are interested in this type of news. Therefore, it has not been widely and widely spread. The attempts were made to obtain the advantages of all the previous solutions we learned about and avoid their defects. AI model was created: to determine the news's reliability on social media, general websites, and all news in the internet world. After that, it has been linked to a website to make it easier for users to recognize the result. The most important benefit is to make it easier for users to handle one site to find out the reliability of all possible news types. Our project proved that it is much more productive than the previous solutions and achieved all the design requirements: capability of machine learning mechanism, high accuracy = 99.8%, and effectiveness through a survey to know the effect of the program on the users.