I enjoy working on open source and ultra learning projects. These are a few I am proud of.

Learning Projects

Solutions to Introduction to Algorithms

Algorithms, Math (Jun 2016 – Present)

My solutions for the problems from Introdution to Algorithms, Third Edition.

Read the PDF. The source code is also available on GitHub.

Solutions to Elements of Programming Interviews

Algorithms (Feb 2017 – Present)

My solutions for the problems from Elements of Programming Interviews, Second Edition.

The source code is available on GitHub.

Research Projects

Crowd Detection with BinBoost Descriptors

Computer Vision, Machine Learning (Mar 2014 – Jun 2014)

Method to detect crowds of people on images. Achieved an accuracy of over 94%, based on a dataset of 660 images collected from Google Image Search.

Read the paper.

Low False Positive Learning with Support Vector Machines

Machine Learning (Dec 2012 – Feb 2014)

An extension of the traditional support vector machine classifier that incorporates the ability to control the false positive rate to a user-specified maximum. Method effectiveness showed trough comparisons to other solutions in 33 datasets.

Risk Area SVM

Read the paper. The source code is also available on GitHub.

Feature Based Product Ranking

Machine Learning (Aug 2012 – Dec 2012)

A product ranking algorithm that takes into account the cost-benefit ratio. Agreement of 64% according to a test conducted with 79 users.

ItemRank

The source code is available on GitHub.

Applications

Invoices Web Client

Web Application (Apr 2017 – Present)

A React web client for managing invoices.

The source code is available on GitHub.

Website and Submission System of the IEEE IFC 2013

Web Application, CLI Application (May 2013 – Present)

The project provided students and researchers the opportunity to showcase and pit their skills in a rigorous competition in the application of digital image processing.

The website and the submission system are available online.

Flight Scan

Web Crawler, CLI Application (Oct 2013 – Mar 2014)

Search for flight tickets directly from the terminal.

$ python2.7 search_date.py -a avianca -o GRU -d GIG -e 2017-05-30 -r 2017-06-29
> SEARCH SUMMARY
. GRU to GIG at 30/05/2017
. GIG to GRU at 29/06/2017

> COLLECTING TICKETS
. loading headers of avianca airline ... [ 0.000s ]
. collecting tickets ... [ 1.719s ]

> GRU to GIG
. 08:15 - 09:15 (BRL 375.0) {0 stop(s)}
. 12:05 - 13:10 (BRL 375.0) {0 stop(s)}
. 18:45 - 19:55 (BRL 375.0) {0 stop(s)}
. 22:35 - 23:45 (BRL 199.0) {0 stop(s)}

> GIG to GRU
. 06:15 - 07:30 (BRL 115.0) {0 stop(s)}
. 09:00 - 10:15 (BRL 145.0) {0 stop(s)}
. 14:45 - 16:00 (BRL 115.0) {0 stop(s)}
. 18:00 - 19:10 (BRL 115.0) {0 stop(s)}

> RECOMMENDED TRIP
. GRU to GIG: 22:35 - 23:45 (BRL 199.0)
. GIG to GRU: 06:15 - 07:30 (BRL 115.0)
. BRL 199.00 + BRL 115.00 = BRL 314.00 (plus taxes)

Save BRL 10.00 leaving 1 day before.

The source code is available on GitHub.

DecMac

Web Application (Jun 2013)

A decision machine to decide better and faster.

DecMac

Demo. Read the blog post. The source code is available on GitHub.

Spider for Google Shopping and Buscapé

Web Crawler, Web Application (May 2012 – Aug 2012)

An HTML spider to collect laptops data in real-time from Google Shopping and Buscapé. Collected and refined data of 495,000 offers from 34,000 laptops during 7 months.

Read the paper (brazilian portuguese). The source code is also available on GitHub.