Fake news is an issue we all face on social platforms. Sometimes, it’s hard to tell the truth from the false stories we see as we scroll through our feed. What if there was a technology that can decipher news articles and inform you if what you’re reading is true? This new algorithm will save you time, and the trouble of possibly sharing a fake story. 

Machine Learning

AI is an aspect of tech we should be on the lookout for in the coming years. Engineers have already made large leaps in advancing what this system can achieve. Now, they finally have figured out a way for artificial intelligence to detect fake news. However, it’s not an easy process. 

The way AI detects language is through an algorithm. The engineers give rules, laws, and boundaries to this system in which it acts upon its given task. You may imagine that it would be hard for a computer to understand fake news vs sarcasm because the human touch on phrases is something computers will never fully understand. 

With not enough data to train their algorithms to understand fake news complete, they decided to load the system with a bunch of fake news articles. This will allow the AI to compare newer articles to those existing in their system to see which ones they can detect to be fake. 

Researchers from the University of Michigan and the University of Amsterdam are implementing natural language process to search for linguistic patterns or cues that are common in fake news articles. 

It’s not perfect, it gets things wrong about 24 percent of the time, but the humans messed things up 30 percent of the time, so the point goes to the machines.

https://news.umich.edu/fake-news-detector-algorithm-works-better-than-a-human/

The Problem

The problem with wanting an algorithm to detect fake news is that there just isn’t enough data. This tech will take ample time to grasp news articles and the formats we see. 

As mentioned before, AI needs rules and examples to compute any sort of data learning. Without proper examples of fake news, this intelligence could potentially mistake satire for false articles. 

Solution

Developers decided to make a corpus of 9 requirements for an article to be deemed “fake news”. – Read more here 

They decided to feed the system real and fake news for it to decide on its own how to decipher what is truth and what is not. It’s not perfect, but its learning fast. 

Keep a lookout on future endeavors when it comes to Artificial Intelligence. 

Blue Archer can assist you with blog writing techniques and how to develop a story for your clients.