My name is Samuel Oliveira Alves de Freitas. I am a 25-year-old Brazilian data scientist looking for an entry-level job. I earned a bachelor's degree in Accounting in 2021 but decided to use my analytical skills to build data solutions.

The main objective of this data science personal project portfolio is to demonstrate my skills in solving business challenges through my knowledge and tools of Data Science, please check out my Data Science Portfolio.

I have already developed solutions for important business problems such as ranking customers, regression for sales forecasting.

The details of each solution are described in the projects below.

Contacts:

• Data Collecting: SQL, DBeaver, PostgreSQL, SQLite, MongoDB, AWS S3.
• Data Processing and Analysis: Python, Pandas, Numpy, Seaborn, Scikit-learn, Boruta, BeautifulSoup, Selenium, Kedro, DBT, Airflow.
• Development: Git, Github, Gitlab, WSL.
• Data Visualization: Matplotlib, Seaborn, Plotly, Folium.
• Machine Learning Modeling: Regression, Classification, Rank-to-learn, Clustering, Time Series, Ensemble Methods, NA Value Imputation.
• A/B Test Modeling: A/B Testing, Bayesian Inference, Multi-Armed Bandit.
• Deployment and Clouds: Streamlit, Heroku, Fly.io, Flask, Google Sheets, Telegram bot, AWS.

Project with the objective of understanding the relationship between product demand and price changes for an e-commerce company.

Ranking new users for their first reservation given a number of available countries, using techniques for multi-class and unbalancing.

Project focused on creating a loyalty program with the most valuable customers of a UK ecommerce company. finding the most valuable customers for the company, who will be eligible for the Insiders loyalty program, as well as providing a detailed profile description of these customers.

Machine Learning with Rank-to-learn to detect the best customers to cross-sell insurance. In this project I develop an Machine Learning Model that classifies customers to those most likely to purchase car insurance for a company that already has Health insurance sales. The implementation of the project can deliver a financial gain from 10.34M to 27.54M, an increase of 2.66x from the value.

Machine Learning with regression and time series for sales forecasting. In this project I developed an algorithm that predicts the sales of a company for the next 6 weeks with a maximum error of 9%. I also found a possible insight that could increase the annual revenue by at least 8%.

Project for selecting the best performing page on a website using MAB.

Worked on improving the conversion rate of iSketch's email capture page.

Conducted an A/B/n test to increase the click-through rate of a category on the University of Montana's library webpage.

Worked on a project for Electronic House to increase the conversion rate of a product's sales page.

Analyzed the oil and gas market and applied buying and selling strategies for portfolio composition.

Analyzed real estate prices and identified potential buying and selling prices through data analysis.

Collected data from a Jeans company's website using web scraping and saved it to a local database to generate insights about the market.

Developed a machine learning model to predict hotel reservation cancellations.

Developed a credit card limit increase evaluation model for Billion Bank in a team-based competition.

Developed a project to forecast the value of used cars with the objective of reducing Mobility Cars' stock.