mappshow - medical appointment show up prediction

Data

The dataset analyzed in this project contains 110527 records of medical appointments in Brazil and is focused on the question of whether or not patients show up for their appointment.

The dataset is available on Kaggle.

The dataset is preprocessed using dataset_prep_eda.ipynb notebook and stored in the data folder.

Problem Statement

The aim of this project is to prepare a model that predicts whether or not a patient will show up for their scheduled appointment. The model decision should be explainable and should be able to predict the probability of a patient showing up for their appointment.

Machine Learning and Explainable AI

The machine learning and XAI part of the project is implemented in notebooks:

• ml_classifiers/classifiers_xai.ipynb
• ml_classifiers/basic_classifiers.ipynb
• dl_classifiers/tabnet.ipynb
• dl_classifiers/tabnetpfn.ipynb
• glass_box_classifiers/glass_box_classifiers.ipynb

Environment Setup

Using conda environment:

conda env create -f env.yml
conda activate lmappshow