Kerala, After The Flood: IIT Hyderabad team uses data mining, verifies relief requests to help govt's efforts

While efforts continue to reach Kerala, on ground, there are numerous people who have been trying to do their bit to facilitate the rescue missions using information and communication technology. The community of students, faculty, and administrators at Indian Institute of Technology (IIT) Hyderabad has been working relentlessly towards ensuring that any call for help isn’t left unanswered. How are they helping? Nearly 100 volunteers, including students, faculty and administrators, have been trying to gather supplies and funds for the relief mission. In the past three days, the team has been able to raise nearly Rs 2.5 lakh and had 62 boxes of supplies dispatched from IIT-H. The team has tied up with an NGO, Anbodu Kochi, to transport and distribute the supplies. While the National Service Scheme (NSS) cell of IIT Hyderabad has been actively carrying out the collection drive, a group of students from Kerala have been providing back-end coordination for rescue and relief operations in tandem with the Kerala government’s website,  keralarescue.in. Keralarescue.in was launched by the government in the wake of the floods, as a platform to place requests for rescue, food, water, or medicines. Anurag, a software engineer at IIT Hyderabad-based startup in liaison with the IT mission cell, streamlined a process to the filter and verify the massive database of 30,000 requests. The IIT-H team has been verifying the real-time status of these requests by contacting the “requestees” and grading the level of emergency. Later, they teamed up with six similar groups of volunteers from across the country. The updated database then guided the rescue operations on ground. Apart from this, individual emergency rescue cases were also escalated with contacts in the armed forces and Indian Navy. [caption id="" align=“alignnone” width=“825” class=“noalign”]  In the past three days, the team at IIT Hyderabad dispatched 62 boxes of supplies for Kerala.[/caption] Data mining One of the volunteers, Adeeba Hakkim, identified a pattern in people reaching out to eminent personalities like film artists for help and rescue via social media platforms. She has been working towards manually gathering largely-unnoticed requests for rescue and supplies from comment threads. A team of dedicated volunteers then worked tirelessly to personally contact, verify and collate around 400 of these SOS messages. The team then attempted to resolve these cases through local contacts and openly available crowd-sourced data on rescue teams and volunteers. This untapped data was then synced with the larger governmental database, keralarescue.in. In a separate effort, Professor Sreejith, from the Computer Science department of IIT-Hyderabad, has been data mining from Twitter for food and water requests, to reach out to people who have been placing requests. Hotline Anurag, along with Vivek Mohan, a software engineer based at Bangalore, has created a temporary hotline to ensure that no calls for rescue or supplies go unanswered. The hotline runs as an IVR system and makes sure that the call is attended by one of the volunteers without having any wait time. The number is +91 9870513514. “The hotline is something we established after hours of manually following up on calls. Understanding the amount of call flow and the urgency of the situation, we have created an IVR (Interactive Voice Response), which makes sure that the call is transferred to people are ready to attend the call in that instant,” says Anurag. How can you help? Vimal Krishnan, another volunteer, points out to the fact that there are more request calls for consumables and less access to credible information regarding suppliers. Although there may be enough suppliers on the ground, the team is experiencing difficulty in accessing these contacts in real time. “What we currently need is the access to the data sources of such groups which are able to provide help on ground. We are running out of people who are able to meet to cater to the demands of the affected people. Access to crowd-sourced data from other institutes, groups or clusters, is the need of the hour.”  The team also calls out for help from any transportation establishments, trade unions, or communities from any place who can forward and offer their services. The team is now also engaging in post-relief work and looking out for help for rehabilitation of the affected people. The help could be in the form of requirements like back to home kits, house cleaning, finding missing peoples’ data from various camps and transportation facilities.

In a separate effort, Professor Srijith PK, from the Computer Science department of IIT Hyderabad, has been data mining from Twitter for food and water requests, to reach out to people who have been placing requests. The link is here. Should anyone with collated crowd-sourced data or information wish to share it with the team at IIT-H, please contact adeebahakkim@iith.ac.in.