Computer Science student at Federal University of Pernambuco, passionate about Data Science and Machine Learning.
Links:
-
End-to-end data science project with ENEM 2019 data: https://bit.ly/3oE0dIW
Project that goes through several stages of data science: understanding of the problem, sampling, manipulation and preparation of data, visualization and exploration of data, hypothesis tests, AutoML with PyCaret, CatBoost algorithm, evaluation metrics for regression, interpretability with SHAP. Here, a hypothetical asked by a CEO question has been defined to guide development:
"Hey, I was traveling on the web and find a kind of "brazilian SAT", called ENEM. After read more about it, I was wondering if we can take some insights of the data provided by the test organizers and and perhaps create some solution that will help the students to plan which colleges they want to apply to before they even have the official score. What do you think?"
-
Pneumonia detection in x-ray images with Tensorflow: https://bit.ly/3ltPj6X
Detecting pneumonia in advance is important, because when you have pneumonia, it is possible for your lungs to fill with fluid. If this happens, they will not be able to transfer enough oxygen into the blood or get rid of carbon dioxide from the blood. It is a serious condition because your organs need oxygen to function. In this sense, in this project we use of deep learning techniques to detect the disease in x-ray chest images.
-
Leads recommendation with Nearest Neighbors: https://bit.ly/30QIncZ
Some companies would like to know who are the other companies in a given market that are most likely to become their next customers. Therefore, this project is about the development of a product to provide an automated service that recommends leads to a user, given his current list of customers (portfolio).
-
Credit card fraud detection with Random Forest: https://bit.ly/2Da3kWt
In the age of technology, the concern with information security is increasingly latent and one of the sectors that seek to improve fraud prevention is credit cards. Here, in this project, we will predict a credit card fraud using the dataset creditcardfraud, wich has been collected and analysed during a research collaboration of Worldline and the Machine Learning Group (http:https://mlg.ulb.ac.be) of ULB (Université Libre de Bruxelles) on big data mining and fraud detection
- [PT-BR] A matemática na computação: derivadas e gradiente descendente: https://bit.ly/3fa2ves
- [PT-BR] O futuro da inteligência artificial: human-centered AI: https://bit.ly/3haMMgd