Team Name:
404NotFounders
Team Members:
404NotFounders
In the heart of Australia, a story of two peoples has long been told — a story of shared land, but of a persistent divide. For generations, we have talked about "the Gap," a painful space that separates Indigenous and non-Indigenous Australians in health, education, and opportunity. Decades of effort and countless reports have been dedicated to this challenge, yet for all the good intentions, the gap remains. Some of the most critical indicators — incarceration rates, child removal from families, suicide rates — are not just stagnant. They are moving in the wrong direction.
But what if we could change the story? What if instead of just measuring the distance, we could build a bridge to a unified future?
This is the purpose of GapCloser. It is not just another data platform, it is a living, breathing tool for empowerment and connection. It is built on a foundation of trust, guided by the principle that true change comes from the ground up, led by the voices of those who know their communities best.
We deliver on this vision through our core features, which are designed to CLOSE the gap, one insight at a time.
Our project understands that the challenges are not isolated, they are deeply interconnected. By leveraging machine learning methods, we delve into vast datasets to reveal the five most important features for each "Closing the Gap" target. This feature acts as a powerful analytical engine that not only provides users with clear visual insights but also offers actionable recommendations to reduce the gap for a targeted feature. For instance, an analysis might reveal that in a specific community, a rise in early childhood literacy is the most crucial lever for improving employment rate. By finding these multiplier effects, we can focus our efforts where they will have the most profound impact, turning a single action into a ripple of positive change.
The future of a community is not a matter of chance; it is a matter of informed vision. The Lookout feature allows us to see beyond the horizon. It turns data into foresight, helping stakeholders model potential scenarios and predict the outcomes of different interventions. This isn't about guessing, it's about using evidence to forge a clearer path forward, to anticipate challenges before they arise, and to build a future based on deliberate, respectful strategy.
This is the heart of the project. For too long, the story has been told through numbers alone, often without the lived experience behind them. The Our Voice feature gives power back to the communities themselves. It is a self-evaluation tool that allows Indigenous Australians to share their experiences and perspectives directly, providing a vital layer of qualitative data that complements the numbers. It is the crucial step toward genuine partnership and shared decision-making, ensuring that every strategy is born from understanding and respect.
Australia is a continent of diverse lands and peoples. Our platform brings this reality to life on a map. Spatial Insights visualizes the data across every community, highlighting where progress is thriving and where challenges remain. By grounding the numbers in a specific place, it allows government agencies, community leaders, and the public to see a tangible picture of progress, enabling targeted, localized action that truly honors the unique identity of each region.
In a world filled with noise, every voice must have an opportunity to be heard. Echo is a powerful communication assistant, a chatbot finely tuned to understand and respond in a variety of local languages. It ensures that language is no longer a barrier to accessing information or sharing a story. This feature is about amplifying the voices of First Nations peoples, creating a dialogue that resonates across the country and a platform that truly speaks their language.
To present GapCloser’s objectives within a limited timeframe, we selected the employment rate gap between Indigenous and non-Indigenous peoples as our research focus. The main outputs are presented through the Correlation Analyser and Lookout modules, with related visualisations available in GapCloser.
For each gap feature, we provide visualisations that allow users to quickly and clearly understand trends. This supports data analysis and the development of effective policies.
We built a machine learning model to explore ways of narrowing the employment rate gap between Indigenous and non-Indigenous peoples. Our findings are as follows:
- youth_engagement (the percentage of people aged 15–24 years fully engaged in employment, education, or training) had the highest feature importance.
- This was followed by youth_education (the percentage of people aged 20–24 years who had attained Year 12 or equivalent).
- Both features were positively correlated with the target variable (y).
The model performed strongly, with an explainability score of 98.6%.
Based on these findings, GapCloser generates insights for users, providing them with the most effective recommendations for narrowing gaps.
To enhance the model’s usability and accessibility, our system can identify the most highly correlated features for any selected gap feature. This provides decision-makers with valuable insights, supports data-driven decision-making, and further contributes to reducing disparities for Indigenous peoples.
Note: The above example is implemented on the GapCloser website. The code can be found in the Python files on Google Drive.
Go to Challenge | 12 teams have entered this challenge.
