Features
Age, height, weight, activity, vegetables, water, and family history become model inputs.
Change body and lifestyle data, run the simulation, then inspect the score, class, and contributing factors. The model API can replace the dummy engine later without changing the interface.
Enter data and run the simulation.
Important limitation: this is an educational prototype, not a health diagnosis, clinical calculator, or personalized dietary recommendation.
The application exposes the workflow: features enter, values are normalized, a score is calculated, and the result is explained. The API will later replace only the dummy calculation step.
Age, height, weight, activity, vegetables, water, and family history become model inputs.
The model maps the feature combination to a risk class and probabilities.
Factor contributions explain the output without turning correlations into a medical diagnosis.
The first version uses deterministic browser rules to imitate the response shape of a classification model. BMI is calculated from height and weight, then combined with activity, vegetable intake, water intake, age, and family history. These rules exist to test the interface before the API is available, not to replace a trained model. Once the backend is connected, the same form will send features to the prediction endpoint and display the returned class, probabilities, and explanation. The user experience can therefore stay consistent while the calculation engine is upgraded.