Meet the winners of the Hack the Continent Open Buildings Challenge
Meet the winners · 27 Oct 2022, 08:12 · 5 mins read ·
10

Meet the winners of the Hack the Continent Open Buildings Challenge, which ran earlier this year on Zindi. The challenge was to propose a solution that uses data science to extract useful insights from Google’s Open Buildings dataset - a dataset of building footprints across Africa and Asia designed to support applications for social good.

The competition attracted 230 participants and 4 submissions from 49 countries. All vying for a $3000 prize pool. Winners Emmanuel Kipngetich (Kenya), Graham Webber (South Africa), and Julien Yise Peniel Adounkpe (Benin) recount their experiences and share what they learned below.

1st place: Emmanuel Kipngetich, Kenya

📷

Please introduce yourself.

I am Emmanuel Kipngetich (EMMANUEL_KIPNGETICH) from Nairobi, Kenya. I am currently a third-year student pursuing a Bachelor of Science in Geospatial Information Science and Remote Sensing, at the Institute of Geomatics, Geospatial Information Systems and Remote Sensing at the Dedan Kimathi University of Technology.

Tell us a bit about your solution.

I integrated Google’s Open Building Geospatial Dataset with Earth Observation Systems and developed a model that determines the number of human settlements affected by floods, and pinpoints their location for preparedness, response and relief in addressing the United Nations Sustainable Development Goal (SDG) 11 (this goal aims to significantly reduce the number of deaths, number of people affected and direct economic losses caused by disasters).

What set your winning solution apart from others?

The use of data enrichment to address a problem to its satisfaction

What are the biggest areas of opportunity you see in AI in Africa over the next few years?

I see a great opportunity in AI for Earth Observation. For decades, Africa’s land surface and coastline images have been continually captured by satellites, recording a wide range of information about land and water resources. These images are a rich source of information, but they are difficult to acquire, scale up or down, and compute and analyse – their high quality means managing many petabytes of data.

Data science and AI has a great opportunity in processing and analysing this data to generate insights for sustainable development.

What are you looking forward to most about the Zindi community?

I am looking forward to learning and exposing myself to global challenges that give me a chance to grow in the field of data science and AI.

Click here to view his solution.

2nd place: Graham Webber, South Africa

📷

Please introduce yourself.

I am Graham Webber(grahamw), living in Johannesburg, South Africa and working as a senior consultant for BSG. I believe data science is more than mathematics: it is the study of meeting human needs through scientific and innovative endeavours. With lateral thinking, research and hard work, I believe that I can bring valuable input to any situation through problem-solving and creative thinking. I have a keen interest in improving the lives of others through science and technology.

Tell us a bit about your solution. What set your winning solution apart from others?

This is a study on the use of Google’s Open Buildings data to extrapolate Gauteng's quality of life data to the rest of South Africa. It also uses the relative wealth index produced from Meta’s available data. Google used image recognition on satellite images to produce the Open Buildings data set. I investigated how the average building size and Google’s trust in the data correlated to different fields in the quality of life data set. In the learning and forecasting model, I also included a relative wealth index that was calculated previously for the entire country. The quality of life survey only covers Gauteng, the model is thus trained using Gauteng data and then used to predict values for the rest of South Africa.

What are you looking forward to most about the Zindi community?

Interacting with talented and creative data scientists from all around Africa.

What is one thing that the Zindi platform does well?

I like that the challenges are African-based.

Click here to view his solution.

3rd Place: Julien Yise Peniel Adounkpe, Benin

📷

Please introduce yourself.

I am Julien Yise Peniel Adounkpe (pyajk) from Cotonou, Benin and currently working at the International Water Management Institute (IWMI). I am a hydrologist at root, a cartographer by passion and a climate data scientist by a commitment to providing solutions to one of our greatest generational challenges: climate change. My goal is to apply new technologies like GIS and AI to understand the problems of climate change and water-related challenges and propose practical solutions.

Tell us a bit about your solution.

The goal of my solution was to identify the latest locations particularly exposed to drought, using CHIRPS precipitation and Google's Open Buildings data. To achieve this goal I did the following:

1. I clustered the buildings

2. Calculated the one-month Standardised Precipitation Index (SPI)

3. Found the drought exposure locations by intercepting the clustered buildings and the SPI drought locations

What set your winning solution apart from others?

I think this project has two great added perspectives - adding socioeconomic data for drought risk mapping, and adding predicted and projected precipitation to anticipate future droughts.

I hope my solution would help anticipate drought disasters by identifying drought occurrences in near-real-time (later months). This could help better planning of humanitarian responses and disaster risk reduction programs.

What are the biggest areas of opportunity you see in AI in Africa over the next few years?

As a climate data scientist, I foresee that AI would highly impact climate-related research in Africa in the next decades. This would give rise to better climate prediction models that would help protect more people from the adversities of climate change.

What are you looking forward to most about the Zindi community?

I look forward to more challenging competitions to reveal more African talents!

Click here to view his solution.

Back to top
If you enjoyed this content upvote this article to show your support
Discussion 0 answers