Winning Topic: Topic 3: Automated Emergency Event Analysis Across 3 or More Data Stream Categories, and Preferably Across All Data Categories - $30,000 prize
Team Summary: Neuratrace, a subset of IKM Technosys, will be one of the first products to cater to problems with multiple data categories and provide end to end solutions for emergency responders. The team is backed by students and researchers from renowned institutions like Indiana University and Purdue University. Research & Development is one of our key competitive advantages with Ph.D. students under advising faculties from NYU & IUPUI. Founders and members with extensive experience in the public safety domain give us a strong foundation in empathizing with emergency responders. Working professionals working at Data-intensive companies bring in real-world experience with data. With a diverse team of data scientists, computer scientists, researchers, and business professionals, we plan to build end to end solutions targeting emergency responders.
Team members: Madhukar Karmacharya, Ananth BhimiReddy, Sahitesh Reddypelly, Kumar Apurv, Archith Krishnan Rammohan, Aamir Khan, SaiManoharReddy Peddireddy (Indiana University/Indiana Purdue University)
Madhukar Karmacharya is a member of a winning Contest 1 team. He joined the ASAPS team via Zoom to answer some questions about his team’s winning solution and give some insight about the ASAPS Challenge.
To begin with, can you give us an idea of what your mission was and how you achieved it in Contest 1?
Madhukar: Our mission is to empower emergency responders and public safety officials with automatic real-time data assets as soon as possible. This contest is the first to pose the exploring data challenge at a massive level in public safety. Our vision is to solve, collaborate and flawlessly empower emergency responders with real-time data.
Could you share the top two or three questions that you have for your public safety partner? What is it that you are hoping you can learn from the public safety partner?
Madhukar: The first question is why emergency responders would even trust an actionable interface like this? The second question is how are we going to design an actionable interface like this? – which will be personalized and relevant to their actions because we are trying to design something which is for a very critical situation and we are not sure about why and how we can build that trust. That is a major question from my side, from our team side, how can we build this out that the system is so robust and fast that we cannot expect any latency and we can trust the information which is coming here.
What sort of things would be helpful in the data or specific to computing resources that would help contestants in your situation, and maybe contestants in general, do a good job responding to the goals of the ASAPS challenge?
Madhukar: ASAPS is a very first of a kind contest, which is dealing with such a massive data explosion challenge. I believe that with the intensity and the multitude of data, the details of the data, which comes out to be at a massive city level like how you have mentioned before, like eight hours of city data. We are really excited about that kind of data. If it is varied enough in each data stream category, suppose if we have video and it has some different events which covers a day-to-day basis all the events and emergency you want, then we can train the model if we have videos and recordings from different incidents. The more diversified data we have in all these incidents, then at least one or two times we will repeat it to the model. The model will be very mature and then we can build a trust with the emergency responders that way. They can trust its accuracy and its detection. And in terms of community resources, I believe we would need to align more and more with the emergency responders, because we really do not know about their day-to-day functioning. We do not know about their actions. We do not know about how actually they perform in these critical situations, which is very pressurized, and what mental state they go through. We really do not know anything about it. We try to research more and more on the back end, but if we get support from the community, then we will definitely get closer to them.