DataBridge AI

About the Project

Inspiration

DataBridge AI was inspired by a simple but important question: Can data help us better understand and anticipate health outcomes across African countries?**

I built this project using .World Bank World Development Indicators (WDI) to explore life expectancy across 54 African countries from 2000 to 2024. Life expectancy is influenced by many interconnected factors, including healthcare access, economic conditions, education, sanitation, nutrition, and other social and development indicators.

Working with these datasets showed me that valuable insights are often hidden within large and complex collections of data. I wanted to build a solution that could bridge the gap between raw data and meaningful insights, making data easier to understand and potentially useful for decision-making.

This became DataBridge AI - a project focused on using data and machine learning to transform development indicators into understandable insights about life expectancy.

What I Learned

Through this project, I strengthened my skills in the complete machine learning workflow, including:

Data collection and preparation using World Bank WDI data
Exploratory data analysis and visualization
Feature selection and engineering
Training and evaluating machine learning models
Interpreting model performance
Building an interactive application with Streamlit
Deploying a machine learning application for users to interact with

I experimented with different approaches and developed a Gradient Boosting model that achieved an MAE of 1.742 years, RMSE of 2.280 years, and (R^2 = 0.865) on the evaluation data.

Beyond the model itself, I learned that building an AI solution is not only about achieving good predictive performance. The data needs to be carefully cleaned, the results need to be communicated clearly, and users need to understand what the predictions mean and, equally importantly, what they do not mean.

How I Built It

I started by collecting World Development Indicators covering African countries between 2000 and 2024. I cleaned and prepared the data, handled missing values, explored relationships between development indicators and life expectancy, and identified features that could contribute to prediction.

I then developed and evaluated machine learning models, using appropriate metrics to understand how accurately they predicted life expectancy.

The final solution uses Gradient Boosting to generate predictions based on development indicators. I then created an interactive Streamlit application so users can explore the model and interact with the results rather than working directly with a notebook or raw dataset.

The project brings together:

World Bank WDI → Data Preparation → Exploratory Analysis → Machine Learning → Prediction → Interactive Insights

This workflow is the foundation of DataBridge AI's goal of bridging complex data with understandable, actionable insights.

Challenges I Faced

One of the biggest challenges was working with real-world development data. The dataset contained missing values and differences in data availability across countries and years, which required careful preprocessing.

Another challenge was deciding which indicators could provide meaningful information for predicting life expectancy without introducing unnecessary complexity into the model.

I also encountered challenges when deploying the machine learning application. Resolving dependencies and ensuring that the trained model could be loaded correctly in the deployment environment required debugging and iteration.

These challenges taught me that successful AI projects involve much more than training a model. Data quality, reproducibility, deployment, usability, and communication are equally important.

Why DataBridge AI?

DataBridge AI represents my belief that AI and data science can help transform complex development data into insights that are easier for people to explore and understand.

By connecting data, machine learning, and human-readable insights, the project demonstrates how AI can be applied to real-world challenges affecting communities across Africa.

The project is also a starting point. Future versions could incorporate additional datasets, explainable AI techniques, real-time indicators, and more advanced models to provide deeper insights while clearly communicating uncertainty and limitations.