Childhood Pneumonia Data and Machine Learning Algorithms

Description: This repository contains all developed database and scripts:

i) a relevant data dictionary built alongside a domain specialist based on different factors correlated to pneumonia;

ii) machine learning algorithms for analyzing and predicting potential childhood pneumonia deaths with high accuracy;

iii) a consistent database as a baseline for training and testing the algorithms.

How to cite:

SOARES, Felipe A. L.; LOUSADA, Efrem E. O.; SILVEIRA, Tiago B.; MINI, Raquel A. F.; ZÁRATE, Luis E.; FREITAS, Henrique C.. Analysis and Prediction of Childhood Pneumonia Deaths using Machine Learning Algorithms. In: SYMPOSIUM ON KNOWLEDGE DISCOVERY, MINING AND LEARNING (KDMILE), 9. , 2021, Rio de Janeiro. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2021 . p. 16-23. ISSN 2763-8944. DOI: https://doi.org/10.5753/kdmile.2021.17456.

Configuring the Environment (how to run the experiments):

To configure the environment, Python 3.9 was used with the following libraries: pandas, numpy, tensorflow and scikit-learn. To install these libraries, use pip:

• pip install pandas
• pip install numpy
• pip install tensorflow
• pip install sklearn

Hardware used to run the experiments:

The environment used to run the experiments was an Intel® Core™ i5 9400f 2.90 GHz (6 cores), 16 GB, Geforce GTX 1660. The operating system was Linux Ubuntu 20.10 64 bits