Not common or prevalent a few years ago, the term ‘fake news’ earned its repute as 2017’s word of the year. And rightly so, especially as there is no dearth of content on the Internet technically labelled as disinformation.
To differentiate between authentic information vs the fake, a team from IIIT Hyderabad led by Professor Vasudeva Verma, who is also CEO of CIE IITH, in collaboration with J. Ramachandran, an industry veteran and Founder-CEO of Gramener, a data-visualisation company, has created a tool known as a “fake-o-metre” to determine authenticity of content.
Explaining the collaboration, Ramachandran, who came back to India in 2010 after working in the US for several years, says, “We have been working on this industry-institute collaboration for the last four months. The students of the institute were involved in the process, discussion and building stages. I visit the Information Retrieval and Extraction Lab (IREL) a couple of times in a week to work with the team and share my expertise.”
So what made them zero in on a “fake-o-metre”? “There’s a lot of content with easy accessibility on the Internet. The percentage of information that is wrong is rising due to the intentionally wrong motives of some people. So we created the “fake-o-metre” — a web engine where one can copy the tweet or headline of a news story and submit it to get a colour-coded response in percentage terms on authenticity. Lower percentages are coded green, indicating the news has low probability of being fake while higher percentages are coded with red, indicating a high degree of fakeness,” explains Professor Verma.
The team uses a combination of machine learning algorithms, and sophisticated natural language processing to get the results. “The tool looks at two things — the actual content itself and the manner in which the content is being spread. In terms of dissemination, fake news spreads far more quickly and farther than true stories,” says Professor Verma.
The tool works for many languages such as English, Hindi, Spanish and Chinese and includes various forms of online content and genres like political (Indian and US) news, bizarre news, hate news, click-bait celebrity news, technology updates and so on. Presently, the “fake-o-metre” is at the industrial demo stage, but the team is working on a strategy to make it available to every individual domain. Fake-o-metre is one example of how Industry-Institute collaboration can work for the benefit of the society, adds Ramachandran.
