EY 2024 Better Working World Data Challenge: If you are a university student, young professional with less than two years of experience or an EY employee looking to learn more about data science and are interested in building a sustainable future – join us as we ask: How do you preserve biodiversity with the click of a button?

Life in all its various forms is biodiversity. This variety of life is fundamental to the function of ecosystems, the health of forests – and even our prosperity.
Help us build computational models to locate biodiversity, specifically frogs. Frogs are a go-to for scientists wanting to study the health of a particular ecosystem.
The winning outputs will help scientists, policymakers and governments to protect and predict the richness of biodiversity in a specific area.

Table of Contents

EY 2024 Better Working World Data Challenge.

The 2024 Better Working World Data Challenge asked university students and young professionals around the world to build an automated fire edge detection and prediction model. The winning solutions are currently being field tested by the Country Fire Authority in Victoria, Australia, and they have been supplied to other fire authorities around the world.

Join us to collectively help save the plants, animals and microorganisms that are critical to a healthy society and a thriving economy.

Choose from three available options, from beginner to advanced – all are welcome. Curated learning programs from EY, Data Camp and Microsoft to will help you prepare.

also read:
WAEC Timetable for School Candidates 2024/2025 Is Out | PDF

EY Better Working World Data Challenge Levels.

Level 1: Local Frog Discovery Tool

Closes in 49 days, 6 hours and 19 minutes – Jun 2, 2024, 11:00:00 PM

Overview

This challenge is to predict the occurrence of a single species of frog for a single location using a single data source at a coarse spatial resolution.
The output will be a species distribution model of one species of frog. Species distribution models are one of the most widely used ecological tools, a cornerstone in many countries worldwide of environmental regulation and conservation.
Why frogs? Frogs are an indicator species. This means they are a go-to for scientists wanting to find out more about the environmental health of a particular ecosystem.
Because they have permeable skin, frogs are very sensitive to pollutants, and because they can live on both land and in the water, they are a good indicator of the health of these two different environments.
Frogs are poorly served by existing species distribution models. They have very localized distributions, more restricted than suggested by a potentially suitable habitat, and therefore existing models struggle to represent their range accurately.
As indicators of ecological health and proxies for biological diversity, the disappearance of frogs is of great concern. Where frogs occur, we see healthy, thriving, resilient ecosystems. Where frogs have disappeared, we see ecosystems in poor health. All the 2030 Sustainable Development Goals (SDGs) are underpinned by healthy ecosystems. This means we won’t reach our goals if we don’t prevent and reverse the loss of healthy ecosystems.
By joining this challenge, you are part of an important community who have decided to engage in activism using space tech and AI to monitor biodiversity at scale. Monitoring is key to prioritizing actions intended to protect and restore biodiversity.
The estimated time effort this level requires is approximately 16 hours.

Skills

Participants in this challenge will benefit from a basic understanding of math and statistics, as well as some experience in coding, but there are no pre-requisites for taking part.
Participating in this challenge will improve your skills in Python for data science, machine learning and managing large volumes of data.
An example Jupyter notebook has been built, which has a preliminary F1 score of 0.61. The notebook connects to public facing data sources on Microsoft’s Planetary Computer. The output of the notebook is a .csv file that can be uploaded to the challenge platform to give a relatively low score on the ranking board that can be improved over the course of the challenge.
Opportunities for improvement include but are not limited to: modifying the way weather data is sampled over both space and time (i.e. spatiotemporal sampling), modifying the biomes that are sampled (e.g. exclude non-forest biomes), modifying the spatiotemporal sampling of frog occurrence data (e.g. limit to 2024-2025), standardizing the variables, selecting a different set of weather variables (there are 18 to choose from), engineering new features, hyperparameter tuning.
This challenge is considered low complexity because participants are expected to build models that are optimized to work across a single country (Australia) using a single source of data (weather). The course spatial resolution (4 kms) of the test set will also make computation requirements lower and reduce complexities associated with model building at very fine spatial scales.
This challenge has been co-created with remote sensing expertise from NASA (Dr. Brian Killough) and Swinburne University (Dr. Jack White), and biodiversity expertise from the Australian Museum (Dr. Jodi Rowley) and the CSIRO (Dr. Cam Slatyer).
Head to the 2024 GitHub repository for instructions on how to setup your own data and analytics environment in Microsoft Azure using a ‘magic link’.

Challenge Objective

Build a Species Distribution Model for the frog species – “Litoria Fallax” across Australia using TerraClimate variables.

Dataset Used

For Target variable – Frog occurrence datasets (Australia)
For Predictor variables – TerraClimate dataset (Climatic variables) available on Microsoft Planetary Computer portal (You will find additional information for the data to be used in the Data Description tab).

Evaluation

During the in progress phase:

A score out of 1.0 will be generated based on the F1 performance metric.
Participants will be evaluated on the extent to which they improve the accuracy of an existing model.
Once phase 2 closes June 3rd, 2024 at 23:59 (CET), you won’t be able to submit your .csv files in the submission tab.

During the final assessment phase: please check the shortlisting process section in the terms and conditions of the challenge

The organizer will publicly announce the global semi-finalists and will ask those teams to submit a “content package” to support their submissions that will be used to judge whether they are chosen as finalists.
Additional details on how these deliverables are evaluated are detailed in the terms and conditions of the challenge.

RULES:

Participants will be working in teams, create and evolve models/algorithms and submit the output in this platform on an ongoing basis during the duration and in accordance to the timeline of the challenge.
The submissions are scored based on their accuracy relative to standard data sets.
Teams will be required to submit their code for review, as well as any other content used to produce their results. Participants must make a genuine attempt to solve the challenge using the resources available, rather than guessing or cheating using private satellite data for example.

DATA DESCRIPTION:

The datasets for this level are available on the below links:

1) Frog Dataset – Australia (Post downloading the zip folder, Participants can use the “occurence.txt” file for the frog presence data)

2) Getting started notebook on Python – Installing python , Getting started with Jupyter notebook

3) Reference Model building notebook – Sample model notebook

4) Submission Template for participants to predict the frog count on given lat-lon

To Do Task for Participants:

Build a Species Distribution (Binary Classification) Model for the frog species – “Litoria Fallax” across Australia using weather data from the TerraClimate dataset available on the Microsoft Planetary Portal. A sample reference python notebook is provided, built using the sample Climatic variables as predictors and the target variable from the frog dataset having frog presence details.

ALSO READ:
Digital Skills Training for Unemployed and Underemployed Young Females in SA

Level 2: Global Frog Discovery Tool

Closes in 49 days, 6 hours and 14 minutes – Jun 2, 2024, 11:00:00 PM

Overview

Participants who choose to undertake Challenge 2 will develop a species distribution model (SDM) for nine selected frog species across Australia, Costa Rica, and South Africa using a variety of open-source geospatial datasets.
This challenge is for those with intermediate to advanced data skills.
The output will be a multi-species distribution model. Species distribution models are one of the most widely used ecological tools, a cornerstone in many countries worldwide of environmental regulation and conservation.
Why frogs? Frogs are an indicator species. This means they are a go-to for scientists wanting to find out more about the environmental health of a particular ecosystem.
As indicators of ecological health and proxies for biological diversity, the disappearance of frogs is of great concern. Where frogs occur, we see healthy, thriving, resilient ecosystems. Where frogs have disappeared, we see ecosystems in poor health. All the 2030 Sustainable Development Goals (SDGs) are underpinned by healthy ecosystems. This means we won’t reach our goals if we don’t prevent and reverse the loss of healthy ecosystems.
By joining this challenge, you are part of an important community who have decided to engage in activism using space tech and AI to monitor biodiversity at scale. Monitoring is key to prioritizing actions intended to protect and restore biodiversity.
The estimated time effort this level requires is approximately 60 hours.

Skills

Participants in this challenge will benefit from an intermediate-advanced understanding of math and statistics, as well as some experience in coding, but there are no pre-requisites for taking part.
Participating in this challenge will improve your skills in computer vision, machine learning, deep learning and management of large volumes of data.
An example Jupyter notebook has been built, which has a preliminary F1 score of 0.18. The notebook connects to public facing data sources on Microsoft’s Planetary Computer. The output of the notebook is a csv file which can be uploaded to the challenge platform to give a relatively low score on the ranking board which can be improved over the course of the challenge. Participants can build one or more models but must submit a single results file.
Hints are scattered throughout the notebook to help participants make progress. Opportunities for improvement include but are not limited to: modifying the way weather data is sampled over both space and time (i.e. spatiotemporal sampling), modifying the biomes that are sampled (e.g. exclude non-forest biomes), modifying the spatiotemporal sampling of frog occurrence data (e.g. limit to 2024-2025), standardizing the variables, selecting a different set of predictor variables (there are dozens to choose from), engineering new features, hyperparameter tuning.
This challenge is considered high complexity because participants are expected to build models that are optimized to work across multiple countries (Australia, Costa Rica and South Africa) using multiple sources of data (weather, water extent, spectral bands, elevation, and land cover data). The very fine spatial resolution (10-50 meters) of the test set will also make computation requirements higher and increase complexities associated with model building at very fine spatial scales.
Head to the 2024 GitHub repository for instructions on how to setup your own data and analytics environment in Microsoft Azure using a ‘magic link’.

Challenge Objective

To develop a Species Distribution Model for predicting the frog occurrence for nine frog species across Australia, Costa Rica, and South Africa using a variety of open-source geospatial datasets available on Microsoft Planetary Portal (TerraClimate, Sentinel L-2, Copernicus Elevation, Esri 10m Land Cover and JRC Water Surface).

Dataset Used

For Target variable – Frog occurrence datasets (Australia, South Africa and Costa Rica)
For Predictor variables – TerraClimate, Sentinel L-2, Copernicus Elevation, Esri 10m Land Cover and JRC Water Surface available on the Microsoft Planetary Computer portal (You will find additional information for the data to be used in the challenge on the Data Description tab).

Evaluation

During the in progress phase:

A score out of 1.0 will be generated based on the F1 performance metric.
Participants will be evaluated on the extent to which they improve the accuracy of an existing model.
Once phase 2 closes on 3 June 2024 at 23:59 (CET), you won’t be able to submit your .csv files in the submission tab.

During the final assessment phase: please check the shortlisting process section in the terms and conditions of the challenge

The organizer will publicly announce the global semi-finalists and will ask those teams to submit a “content package” to support their submissions that will be used to judge whether they are chosen as finalists.
Additional details on how these deliverables are evaluated are detailed also in the terms and conditions of the challenge.

RULES:

Participants, will be working in teams, create and evolve models/algorithms and submit the output in this platform on an ongoing basis during the duration and in accordance to the timeline of the challenge.
The submissions are scored based on their accuracy relative to standard data sets.
Teams will be required to submit their code for review, as well as any other content used to produce their results. Participants must make a genuine attempt to solve the challenge using the resources available, rather than guessing or cheating using private satellite data for example.

DATA DESCRIPTION:

The datasets for this level are available on the below links:

1) Frog Dataset – Australia, South Africa & Costa Rica (Post downloading the zip folder, Participants can use the “occurence.txt” file for the frog presence data)

2) Getting started notebook on Python – Installing python , Getting started with Jupyter notebook

3) Reference Model building notebook – Sample model notebook

Also you can find additional help for challenge completion in the following notebooks:

4) Submission Template for participants to predict the frog species

To Do Task for Participants:

To develop a Species Distribution Model for predicting the frog occurrence for 9 frog species across Australia, Costa Rica, and South Africa using a variety of open-source geospatial datasets available on Microsoft Planetary Portal (TerraClimate, Sentinel L-2, Copernicus Elevation, Esri 10m Land Cover and JRC Water Surface).

ALSO READ:
CARN ASPIRE Grant Program 2024 | $5,000 for African Graduate Students

Level 3: Frog Counting Tool

Closes in 49 days, 6 hours and 15 minutes – Jun 2, 2024, 11:00:00 PM

Overview

This challenge asks you to build a computational model that can predict the count of frogs for a specific location using multiple data sets, and to validate your model on two additional locations. This challenge is for those with expert-level data skills.
Why frogs? Frogs are an indicator species. This means they are a go-to for scientists wanting to find out more about the environmental health of a particular ecosystem.
Because they have permeable skin, frogs are very sensitive to pollutants, and because they can live on both land and in the water, they are a good indicator of the health of these two different environments.
Frogs are poorly served by existing species distribution models. They have very localized distributions, more restricted than suggested by a potentially suitable habitat, and therefore existing models struggle to represent their range accurately.
The estimated time effort this level requires is approximately 80 hours.

Skills

This level:

Requires prior experience in Python for data science and machine learning/AI knowledge skills. It is also helpful to have prior experience in neural networks and computer vision.
This challenge will improve your skills in neural networks, computer vision and knowledge of handing large volumes of data.

Challenge Objective

Build a combined Regression model to predict the density of frog population for regions covering Australia , South Africa and Costa Rica.

Dataset Used

For Target variable – Frog occurrence datasets (Australia, South Africa & Costa Rica)
For Predictor variables – TerraClimate dataset: Climatic data – available on Microsoft Planetary Computer portal. Participants are free to incorporate other data catalogs available on the Microsoft Planetary portal while building the model for frog density prediction.

FAQ Section

Participants can review the FAQ section for this Level for addressing queries if any.

Evaluation

During the in progress phase:

A score out of 1.0 will be generated based on the F1 performance metric.
Participants will be evaluated on the extent to which they improve the accuracy of an existing model.
Once phase 2 closes on 3 June 2024 at 23:59 (CET), you won’t be able to submit your .csv files in the submission tab.

During the final assessment phase: please check the shortlisting process section in the terms and conditions of the challenge

The organizer will publicly announce the global semi-finalists and will ask those teams to submit a “content package” to support their submissions that will be used to judge whether they are chosen as finalists.
Additional details on how these deliverables are evaluated are detailed also in the terms and conditions of the challenge.

RULES:

Participants, will be working in teams, create and evolve models/algorithms and submit the output in this platform on an ongoing basis during the duration and in accordance to the timeline of the challenge.
The submissions are scored based on their accuracy relative to standard data sets. Participants will only see the public score during the challenge.
Teams will be required to submit their code for review, as well as any other content used to produce their results. Participants must make a genuine attempt to solve the challenge using the resources available, rather than guessing or cheating using private satellite data for example.

DATA DESCRIPTION:

The datasets for this challenge are available on the below links:

1) FrogID Dataset – Australia , South Africa & Costa Rica

2) Getting started notebook on Python – Installing python , Getting started with Jupyter notebook

3) Reference Model building notebook – Sample model notebook

4) Test Data.csv for participants to predict the frog count on given lat-lon

To Do Task:

Build a combined regression model to predict the density of frog population for regions covering Australia , South Africa and Costa Rica. A sample reference python notebook is provided, built on the entire Australia region by using the Climatic variables as predictors and the target variable from the FrogID dataset having frog presence details.

Participants must include frog presence datasets for the time period Nov’2017 to Nov’2024 from Australia (frogID dataset), South Africa (iNaturalist Research-grade observations) and Costa Rica (iNaturalist Research-grade observations) to build the combined prediction model excluding Australia (State Province = “South Australia), South Africa (State Province = “Western Cape”) and Costa Rica (State Province = “Puntarenas”) from the respective datasets.

There are multiple data Catalogs available on Microsoft Planetary Computer portal which can be used while building the frog density model i.e. Sentinel-2 Level 2A, JRC Global Surface Water, TerraClimate etc. Participants are free to explore these Catalogs further. In the Appendix section of the reference notebook, there are two examples of data extraction code shown for Sentinel L2 Catalog and JRC Water Surface Catalog available on Microsoft Planetary Portal respectively.

For model accuracy, the test.csv contains the list of coordinates (lat-lon), on which the participants must predict the frog density on the above excluded regions (South Australia, Western Cape & Puntarenas).