Filtering out fake news about disasters on social media platforms such as WhatsApp, Twitter and Facebook is going to get simpler.
The Indian Institute of Technology Kharagpur has come up with a solution that uses artificial intelligence to extract critical information from social media platforms that is difficult to obtain manually. This information can be used to determine the authenticity of posts and also pass on data to aid rescue and relief operations.
“Our solution can detect fake news and can even alert users in the time of disasters through deep machine learning algorithms,” said Saptarshi Ghosh of the Department of Computer Science & Engineering at IIT Kharagpur.
This solution is limited only to disasters in terms of identifying rumours or fake news. Ghosh said the solution is capable of at least 90% accuracy in distinguishing fake news from real.
In addition, the solution helps to glean information that can help victims of disasters. It is estimated that only 2% of tweets have relevant information on disasters while the remaining content is largely conversations, mostly sympathy for victims.
This is how it works: While one tweets about the situation of victims hit by an earthquake in English or Hindi, a computer programme can read through the empathetic post and send the relevant information to relief operators nearby.
“We are developing AI methods to extract specific types of critical and actionable information from social media content posted during disasters such as resource needs, resource availability and information about trapped people,” said Ghosh.
The institute along with its collaborators at Qatar Computing Research Institute (QCRI) has submitted a proposal to Microsoft Research India. “If the proposal gets accepted, we will collaborate with Microsoft Research and QCRI to develop the systems for aiding post-disaster relief operations,” said Ghosh.
A person sitting in the control room can get live updates about what resources are needed and where they are available and coordinate relief operations accordingly.
Or a relief worker can be updated via smartphone about trapped people so that they can be rescued.
IIT Kharagpur, along with the Centre for Urban Science & Engineering at IIT Bombay, has also submitted a research project proposal to the Department of Science and Technology.
“As part of this project, we wish to focus on emergency situations in urban areas, e.g., the recent floods in Chennai and Mumbai,” Ghosh said.