The “fake news” phenomenon may have captured the imagination of Americans during the 2016 Presidential Campaign and an investigation then Russia’s efforts to swing the election to Donald Trump using fake news on Facebook among other schemes.
The truth is that fake or fake news has existed as a temporary tool and by many people to spread the theory of propaganda and conspiracy for many years before the 2016 election. The website including Infowars and Brietbart, among others, has spread false news that supports their agenda.
However, this has become a political and social problem since the election and a bad Facebook have become a poster child of the website that falls in love with the scheme.
Recently, social media companies have recognized their mistakes and have tried to fix them with their customers. Now marks fake news articles that will go to Facebook members through their news bait. This uses AI to achieve this.
The company uses AI to identify words or phrases that might indicate that the article is actually fake. Data for this task is based on articles marked by Facebook members individually as fake stories.
This technology currently uses four methods to find fake news. They include:
Web page score. First using this technique is Google. It uses facts to make scores for websites. Obviously, the website score is an ongoing action. However, as Google did, this technology has grown significantly.
Considering facts. This method uses natural language processing machines to review story problems. AI uses other models knowing whether other sites report the same facts.
Predict reputation. This technique is based on AI using predictive analytic and machine learning to estimate the reputation of the website by considering a number of features including domain names and web ranking Alexa.
Find sensational words. Fake news supporters have used sensational headlines to take potential audience interest. This technique finds and marks fake news canopy using keyword analytic.
The actual detection of this type of article by AI is a difficult effort. Of course, large data analysis is involved, but also concerns the truth of the data. Identifying it is actually involved in determining the truth of the data. This can be done using the method of weighing facts. What happens if fake news articles appear on hundreds of websites simultaneously? In this situation, using the technique weighing facts can cause AI to determine that the story is valid. Maybe using the reputation prediction method in relation to weighing facts can help, but there are still problems. For example, a reliable news source website that does not take the time to verify the news can take it with the assumption that it is true.
It is clear that using AI to identify these articles requires more development. A number of organizations are involved in improving AI’s abilities. One of the establishments involved is West Virginia University.
Reed College of Media in collaboration with Benjamin M. Statler College of Engineering and Mineral Resources of West Virginia University has created a course that focuses on the use of AI to identify fake news articles.
Senior students who take the Elective Course of Computer Science work in teams to develop and implement their own AI programs are also involved in this project.