Chennai floods: How social media and crowdsourcing helps people on ground
Bengaluru, Dec 5: Floods in Chennai have spelt doom for the people there and in times like this people from all across are trying their best to contribute.
Whether it is creating online campaigns to save the people in Chennai or doing mobile recharge for a complete stranger, people from across the country and even abroad are helping as much as they can.
Social media platforms have played a huge role in such campaigns. One such example is that of the team headed by Professor Amit Sheth. Professor Sheth's team is no stranger to natural calamities.
They have worked extensively during the Jammu and Kashmir floods in Sept 2014, where they used social media which helped relief operations immensely.
Dr Shonali Krishnaswamy, Department head at Singapore's Institute for Infocomm Research was worried about her parents who were stuck in Chennai. She contacted Prof Sheth and sought his help in the situation. Prof Sheth and and his team then asked for the location and using crowdsourcing techonolgy via social media were able to tell Krishnaswamy that there was no flooding in the area where her parents lived.
Prof. Sheth and his team at the Ohio Center of Excellence in Knowledge-enabled Computing (Kno.e.sis, Wright State University, USA) are carrying out a new NSF-funded project Social and Physical Sensing Enabled Decision Support for Disaster Management and Response. They have mobilised to use their technology to monitor and analyze social media and crowdsourcing to support better situational awareness for the Chennai floods.
How social media helps in such situations
In case of the Chennai floods, images showing flood water provide critical information that can be helpful in understanding the situation on the ground and planning for rescue. Therefore, Prof Sheth and his team, using Twitris and new Photo Mapping tools, have created an initial crisis map where pictures of flooded areas are pinned (Read more)
A key challenge is that most of the pictures (or corresponding tweets) do not come with geolocation. So the team pursued crowdsourcing to identify location in terms of relevant neighborhood or crossroads (such as the location terms Krishnaswamy provided in her query).
Kushal Shah, a volunteer from Noida went through available information online as well as reached back to those who had shared the photos and was able to tag over 60% of photos. The figure provides a current snapshot (see image 2) while the Kno.e.sis team is still in the process of adding more capabilities to automatically identify locations through text processing and/or through direct interactions with image posters.
This allows anyone to find recent images from the areas of their interest.
People concerned about their friends and family may be interested in getting the latest information from social media regarding the situation where his family or friends live.
Additionally, this can also help the rescue teams, who can also review the level of water and other crowdsourced information, including requests from the affected area while planning a rescue operation.