| Name | Kelwin Fernandes |
| Date of birth | 1990/02/22 |
| Address | Porto, Portugal |
| Languages | Spanish, English, Portuguese |
| Website | www.inesctec.pt/~kafc |
| kafc at inesctec dot pt |
My name is Kelwin Fernandes and I am a PhD student in
Informatics at
University of Porto
and at INESC TEC in Porto, Portugal.
I obtained my Bsc degree in Computer Engineering with
Summa Cum Laude distinction at
Universidad Simón Bolívar in
Caracas, Venezuela.
My main research insterest include: Machine Learning,
Computer Vision and Artificial Intelligence.
2014 - 2018 (expected) |
Universidade do Porto PhD Student in Informatics
|
2007 - 2012 |
Universidad Simón Bolívar Computer Engineering (Summa Cum Laude)
|
Learning to Rank |
Abstract: We study the problem of learning lexicographic preferences on multiattribute domains, and propose Rankdom Forests as a compact way to express preferences in learning to rank scenarios. We start generalizing Conditional Lexicographic Preference Trees by introducing multiple kernels in order to handle non-categorical attributes. Then, we define a learning strategy for inferring lexicographic rankers from partial pairwise comparisons between options. Finally, a Lexicographic Ensemble is introduced to handle multiple weak partial rankers, being Rankdom Forests one of these ensembles. We tested the performance of the proposed method using several datasets and obtained competitive results when compared with other lexicographic rankers. (code, data, results and paper) |
Imbalanced Data |
Abstract In classification, when there is a disproportion in the number of observations in each class, the data is said to be class imbalance. Class imbalance is pervasive in real world applications of data classification and has been the focus of much research. The minority class contributes too little to the decision boundary because the learning process learns from each observation in isolation. We discuss the application of learning pairwise rankers as a solution to class imbalance. We compare ranking models to alternatives from the literature. (code, data, results and paper) |
Directional Data |
Abstract: In different areas of knowledge, phenomena are represented by directional -angular or periodic- data; from wind direction and geographical coordinates to time references like days of the week or months of the calendar. These values are usually represented in a linear scale, and restricted to a given range (e.g. [0, 2π)), hiding the real nature of this information. Therefore, dealing with directional data requires special methods. So far, the design of classifiers for periodic variables adopts a generative approach based on the usage of the von Mises distribution or variants. Since for nonperiodic variables state of the art approaches are based on nongenerative methods, it is pertinent to investigate the suitability of other approaches for periodic variables. We propose a discriminative Directional Logistic Regression model able to deal with angular data, which does not make any assumption on the data distribution. Also, we study the expressiveness of this model for any number of features. Finally, we validate our model against the previously proposed directional naïve Bayes approach and against a Support Vector Machine with a directional Radial Basis Function kernel with synthetic and real data obtaining competitive results. |
2016 |
|
2015 |
|
2014 |
|
2012 |
|
2015 |
|
2014 |
|
2013 |
|
2012 |
|
2011 |
|
2008 |
|
Universidade do Porto |
|
Universidad Simón Bolívar |
|