Welcome to my Data Science Portifolio

This portfolio is a compilation of data science projects which I created for exploration of machine learning algorithms, also for data analysis, presented in the form of iPython Notebooks.

Classification problems

Credit Card fraud detection:

• Dataset provided by Kaggle.
• Visualization and preparation of the data itself.
• Model using SMOTE and Random Forest.

Hand Writting Digits recognition:

• Evaluate the performance and predictive power of a model that has been trained and tested on 28 x 28 pixel images.
• Visualization e exploration of the MNIST data set.
• Comparison of different algorithms performance.
• Usage of ROC and PR Curve.

Support or Contact

If you liked what you saw, want to have a chat with me about the portfolio, work opportunities, or collaboration, shoot an email at pedroaugusto.eng@gmail.com.