Jorge Luis Guevara Diaz  Jorge Luis Guevara Diaz photo         

contact information

Research staff member
Brazil Research Lab, Brazil



I am research scientist at IBM-research within the the Natural Resources Analytics group.I got a PhD degree in computer science from the Institute of Mathematics and Statistics- University of Sao Paulo Brazil. I was oriented by professor Roberto Hirata Jr. Also, I did a  PhD internship at LITIS laboratory-INSA Rouen, France working with Stephane Canu. I am interested in fuzzy mathematics, fuzzy machine learning approaches, machine learning on sets of probability distrbutions, speech recognition and applied machine learning. My main PhD contributions were the developement of similarity measures between fuzzy sets with geometrical interpretation in Hilbert spaces using the kernel Hilbert embedding approach. I applied those similarity measures on supervised classification tasks on dislexy, sport and  noisy datasets; and kernel hypothesis testing on data with linguistic attributes. I also contributed on the developement of a machine learning models for the group anomaly detection task: Detecting anomaly distributions from sets of distributions. I applied those models to detect anomalous clusters of galaxyes on the Sky Loan Digital Survey.

Nowadays, I am working on applied machine learning for problems related to the Oil and Gas Industry. In the past I worked as a full professor in Peru.