Not every message in your WhatsApp groups or on your Twitter and Facebook timelines is true.
Far from the truth, the news reports, images and videos may well be a figment of someone’s imagination, and have the ability to trigger tensions.
With the fake news menace virtually going out of control, researchers at the International Institute of Information Technology (IIIT-H) have developed a Fake-O-Meter — a web engine that quickly analyses voluminous data to assess the veracity of the report in question.
“The solution is based on Machine Learning, Artificial Intelligence and Natural Language Processing technologies. You copy the URL (web address) of a report that you wanted to verify and paste it in the query box.
“You will get an answer,” Vasudeva Varma, Professer and Dean (Research) of IIIT (H), told BusinessLine .
After testing the solution in labs and demonstrating it at a few seminars, the institute plans to soon release the tool for public use. The solution gets better by the day as it takes questions, solves them and learns, resulting in better results. “Fake news is affecting our lives. It is even resulting in killings. The existing methods have limitations. Checking them manually takes a lot of time,” said Varma.
A 10-member team, led by Vasudeva and J Ramachandran, the mentor of data visualisation firm Gramener, developed the Fake-O-Meter, which rates the reports in green, amber and red after checking them in the background.
Sitting on Artificial Intelligence and Machine Learning and NLP technologies, the engine learns by the day when used more and more.
The solution does content analysis, network (identifying who is spreading the news) analysis and domain analysis.
Currently available in English, Chinese and Spanish, the meter will be available in Hindi, Tamil and Telugu at a later stage.
Tailormade versions
While it will be offered for free to the public, tailormade versions will be offered to media houses for a fee.
“We are doing the analysis in specific verticals such as politics, entertainment and technology,” he said.
“For media organisations, we can deploy on their servers and they can check the reports against the data they provide,” he said.
The problems created by technology can be solved by technology itself, he asserted.