Wrapping up BuildwithAI Hackathon 2020
Every year Hackmakers organizes the #BuildwithAI Global Hackathon, a bi-annual global hackathon centered around the application of best-practice data science and artificial intelligence processes to address global challenges.
COVID-19 has exposed some of society's greatest challenges in mobilizing health information, global supply chain shocks and disease monitoring. Due to both the sudden impact and black swan status of this phenomena, a lot of data has been generated. However, this is often scattered and conflicting, with unreliable insights generated. Many poorer nations do not have access to strong health systems and communications capabilities, meaning that automated systems are needed with even more urgency.
The BuildwithAI Hackathon was born out of a desire to help society emerge stronger from a post COVID-19 world, using AI and Data Science. Each challenge required understanding and analysis of various data science and AI concepts, including natural language processing, developing predictive algorithms or effectively visualizing data.
This year we partnered with Oracle, Amazon Web Services, Google Cloud and IBM as our major sponsors and DATAcated Academy, Packt, Black Nova Group and Antlers as our community partners.
The hackathon revolved around the following topics:
The July 2020 challenge was themed around the global COVID-19 pandemic, with over 4,000+ participants, 250+ mentors from 70+ countries and 20+ timezones working together to produce meaningful solutions.
The hackathon was supported by globally celebrated panelists, speakers, judges and mentors including:
Dr Cassie Kozyrkov- Chief Decision Scientist, Google, USA
Cherie Ryan- Vice President and Regional MD, Oracle ANZ, Australia
Dr. Ian Opperman - Chief Data Scientist, Government of New South Wales, Australia
Matthew Brown - Founder, Black Nova Group, Australia
Steve Nouri - Founder and CEO, AI4Diversity
Lillian Pierson - Founder Coach, Data-mania LLC
Kate Strachnyi - CEO Founder, Datacated Academy
The Promo video for Global AI Hackathon #BuildwithAI 2020: Emergence was Premiered on 9 Jul 2020, on :
After weeks of judging hundreds of innovative and creative solutions, our judges have finally come up with the best solutions presented in the hackathon. And the winners of the hackathons are:
Global Showcase Competition
Top Finalist: AI assisted social distancing monitoring
Members: Mohamed Mossad
Solution: This project is unique as it delivers a real time analytics dashboard, analyzing how people are keeping social distancing and actually where social distancing is violated
Winner: Team COVID Doctor
Members: Subham Goel
Solution: COVID Doctor offers the following services:
Real-time Pulse Detection - Pulse detection plays a vital role in the detection of COVID. Respiratory rate is the number of breaths (inhales and exhales) you take in a minute and ranges from 12-20 breaths in a healthy person. Resting heart rate is the number of times your heart beats per minute and ranges between 60-90 in a healthy person, and heart rate variability is the time in between each heartbeat. So the change in pulse rate will help in early detection of COVID 19 and this can be tracked using just a camera and a browser.
COVID X-Ray Classification - With the help of this feature we can get an immediate result of our X-Ray instead of going to the hospital and waiting for the outcome, which can lead to the spread of the deadly virus. So, using Deep Learning technology we have built an X-ray Classifier which classifies whether the patient is COVID Positive or Negative. This will save time and stop the spreading of the virus. Live COVID 19 Stats - People can monitor the live stats of Coronavirus Confirmed, Active, Recovered cases, etc. in one place.
Real-Time Self Assessment Chatbot - People can take their self-assessment checkup by answering a few questions and based on those answers, the user will get the advice by the chatbot such as quarantine or visit the hospital or precautions if the user is healthy. This will help people in ensuring their safety.
Data Visualization Challenge
Top Finalist: Team SaferWherever
Members: Tugrul Guner, Dhruv Sheth, Ignacio Amat, Parijat Khan, Ramona Rajagopalan, Deepak Vohra, Arijit Das.
Solution: They have built a detection model that utilizes real-time CCTV footage to determine the pop. density within a small area (1 camera per area) and classify hot spots, where the density is very high. The model also detects violations in the use of face masks and in social distancing within a camera’s viewpoint. Apart from this, they've created a temp. sensor model to detect temp. as people enter the mall. The outputs are sent to a cloud-based service and presented as a bubble map on the dashboard.
Winner: Team Maverick
Members: Siddhartha Ranjan Bhattacharya, Sherry Aggarwal, Debasis Syasmal, Biplab Das Chowdhury, Nithya Subbaraman, Mahavir Gupta, Stuti Singh , Rithika Venkatesan
Solution: They provide a view of the overall employment trend starting with 2019 and covering the pandemic period for each state wise. Post that there is employment based on job profiles and industry sector. High impacted & low impacted areas are clearly called out. Post that they have provided a risk profile for each job type based on current demand and job postings.
Natural Language Processing Challenge
Top Finalist: Team Cognitive Camp
Members: Senthil Sundaram, Keerthi Prasad, Serena Raju, Hemen Boro, Ramya Mamidipaka
Solution: The decision making tool will give the pointers to authorities to show Behavioral patterns to,
1. Increase in consumer demand will happen only when there is confidence among people to start improving their lifestyle
2. Businesses should feel that there is demand as a result of people trying to increase their lifestyle.
3. To address people’s fear, by helping governments to target campaigns on awareness by segmenting groups that will result in bringing people out of fear.
Winner: Team Infodemic Warriors
Members: Supriya Medapati, Sarika Kondra, Naga Pakalapati, Ripon Kumar Saha, Jesus Rodriguez, Aayush Patni
Solution: They attempted to understand public perception and the influence of misinformation on public sentiment to help policymakers with data-driven decision making to introduce checks and balances to curb the spread of misinformation, inspire media and influential personalities to be cognizant of their influence and share the right news and wake up general public to be more conscious about their news consumption.
Predictive Algorithm Challenge
Top Finalist: Team Make Unicorns Great Again
Members: Raquel Pérez Arnal, Manuel Gijón Agudo, Ferran Parés, Dmitry Gnatyshak Pont, Dídac Fernández Cadenas,Idril Geer,Leo Lamsdorff.
Solution: They used the SEIRS model to predict the number of cases of covid, we estimated the parameters of SEIRS using genetic algorithms.
Winner: Team T-Cube
Members: Taarak Rapolu, Brahmani Nutakki, Raakhal Rapolu, Brian H. Hough, Ruben Correoso Campillo, Subhash Sarangi
Solution: They propose a data-driven model based on epidemiology that forecasts the number of confirmed and deceased cases over the next three weeks. The forecasts are set up on an interactive dashboard with multiple features. Their potential users are the Governments and other systems adopting the solution to assess the demand and supply for resources and the forecasts would help in alarming the people in Hotspot areas. Another use case is in distribution chains to manage their deliveries and supply chain
The Awards Ceremony was broadcasted live on 18th August 2020 on:
A huge thanks to all the participants, mentors, ambassadors, sponsors and partners for being a crucial part of this hackathon, for providing both positive and constructive feedback and making this event a reality!