This is a simple project with a simple goal: to document the daily small good things that happen so that we keep perspective and get a boost to get on with our day.

Snippetmate let's you organize the problems you've faced as a developer and their solutions, all in one place!

An effort to standardize ML experiments and to move from script-like code that saves results to file to a more structured framework that can save results in a database in such a way so that the experiments can be repeated and/or resumed with the same parameters and in the same way and the results can be compared with a simple query.

This repository is for the Roadrunner "Efficient and Scalable Processing of Big Data" project. The result of this project was published in SENTIRE 2016.

I am the 1/2 of the Productive Engine team. Snippetmate is one of our projects.

My Thesis project, "Sentiment Analysis on Figurative Data: Figurative speech on Twitter". This was my first attempt on ML/ Data Science and Python. It started on 2013 with a different goal (ABSA), evolved into this project in late 2014 in order to participate in SemEval 2015 Task 11 (DsUniPi team), where it got the 10th place among 15, and finally, it was finished, finalized and delivered in December of 2015.

Analyzing short, noisy data like class names and attributes, tweets and hashtags means that there is a need for some kind of normalization. This component attempts to combine several methods to normalize the aforementioned kind of data. It focuses on word-level analysis (not a whole phrase that can be split in words by spaces but e.g. "SomeClient", "FistName", "lastName" etc) and uses regex, spell-check, nlp and dynamic programming to try analyze the data and return a list of normalized words.