• 👋 Hi, I’m @hucodelab and my hobbies are: running, reading, acting, and studying.
• 👀 I’m interested in Data Engineering, Data Science, Business Intelligence, Logistics and Financial Markets
• 📫 How to reach me: [email protected]

Skills: Python, SQL, Cloud Computing, AI, Logistics, Quant.

• e-commerce binary classification with Starbucks Data

This project shows the ETL and Machine Learning model building process to predict whether a client will accept an e-commerce offer. It was also made a deploy of the app.

• Comparison of binary classification machine learning algorithms to predict US population income data (>85% accuracy)

This project shows the process of manipulating data and building a Machine Learning model to predict whether a US citizen has an income higher than $50,000 per year. The project compares the performance of 3 models: Random Forest, Random Forest with Grid Search (hyperparameter optimization), and logistic regression.

• LSTM model to predict Stock prices

This project shows the process of data manipulation and construction of a LSTM model to predict whether a stock it's going to increase its market price or not. It was built a feature selection model to select the best features for the LSTM model.

• Soybean yield predictions in Paraná - BR using an artificial neural networks (Multilayer Perceptron - MLP) regression model

This project shows the process of data manipulation and construction of a Deep Learning regression model trained with climatological data from the 20 largest soybean-producing municipalities in the state of Paraná - BR to make predictions of soybean productivity.

• Performance comparison of Street Number Recognition using neural networks models: Convolutional Neural Networks (CNN) and Multilayer Perceptron (MLP)

This project shows the process of data manipulation and Deep Learning classification model building trained to recognize Street numbers. The project compares the performance of two neural network models: Multilayer Perceptron (MLP) and Convolutional Neural Network (CNN).

• Brazilian elections' candidates Twitter data

This project shows metrics and indicators related to the twitter accounts of the main candidates of the Brazilian 2022 elections. The visualizations created in this project were deployed to a web application using DASH and REACT. This project was developed by Turing USP and I contributed to the data processing and visuals generation.

• StackOverflow 2020 Survey Developer Data Analysis

This project shows the process of extracting, manipulating, and analyzing data from the 2020 StackOverflow survey. The project contains visuals of the number of respondents by programming language, salaries, and developers' job satisfaction.

• Airbnb Brazil Web Scrapping

This project shows the scrapping (extraction) of Airbnb data from Airbnb Brazil. It was possible to make a comparative analysis between accommodation's prices three different brazilian cities as well. This software was developed using the python library: BeautifulSoup.

• Rotten Tomatoes Web Scrapping

This project shows the scrapping (extraction) of Rotten Tomatoes data and it was possible to build a dataset using the data. The project was developed by using the python library: Selenium.