AI is the key to spotting fake news


The proliferation of yellow journalism or fake news and the way it spreads, especially on social media, has become of great concern due to its devastating effects. While the information we need is just a click away, there is also a lot of misinformation out there about products, religion, communities, etc. on the internet that is spreading more through social media, print media, and news outlets. information channels.

There are almost no restrictions by social media platforms on the content that is posted. Most people do not verify the source of the information they view online before sharing it, which leads to the rapid spread of fake news, even viral.

In addition, it is very difficult to identify the source of this information, which makes it more difficult to assess its accuracy. . Social media has become a dominant source of news and information and has radically reshaped the media industry.

However, fake news existed long before social media arrived. It became a buzzword after the 2016 US presidential elections. The internet has given impetus to fake news, whether it is spurious reports or rumors.

The good news is that in the near future, artificial intelligence or, to be more precise, models based on machine learning will help a user to verify if the news is true or false. Although research in this area is ongoing, there is still a lot to do. A particular area of ​​research known as natural language processing (NLP) is receiving a lot of attention from academics and academics around the world.

Automated detection

With the growing number of online media users, automated detection seems to be the only way to fight fake news. So far, there are text-based approaches to detecting fake news, but they haven’t given the desired results. Almost all machine learning models use handcrafted functionality extracted from input textual content.

In the future, we will see a contextual approach to detecting fake news. In 2016, some researchers found that almost half of the information on Facebook is fake and hyper-partisan. And news agencies depend on Facebook for 20% of their traffic.

Fake news also spread on Twitter. Recently. thanks to a new approach, it was found that fake news was being tweeted during the Covid-19 pandemic to mislead the target population. Machine learning and highly sophisticated deep learning models are continually used by researchers and industry to develop automated models based on the detection of false information.

Many of these models detect particular types of information such as political and religious. Some research reviews reveal that these models have characteristics for specific data sets that match their topic of interest. Such approaches can suffer from a bias in the data sets and perform poorly on news from another topic.

Much more advanced

Models based on deep learning are bringing about a revolution in almost every area of ​​life. Recent developments in natural language processing show promise in fake news detection. With the advent of Keras (a deep learning API) and Tensorflow (end-to-end open source platform for machine learning), coding and implementing these smart models has become much more. easy compared to ten years ago.

In the future, deep learning models will be able to classify fake news and legitimate news. Decades of experiments in deception detection show that humans are not good at identifying lies in text.

The spread of fake news harms public behavior, beliefs and attitudes, which in turn can endanger democratic processes. The early detection of this false information and the control of its propagation is the main challenge for researchers today.

The author is associate professor, Great Lakes Institute of Management, Gurgaon

Previous The real challenge of reconciliation between Christians and Jews
Next Mary K. Johnson | Obituary

No Comment

Leave a reply

Your email address will not be published. Required fields are marked *