Photo

Hermenegildo Fabregat

Ph.D Student @ UNED NLP Group

Madrid, Spain

gildo.fabregat (at) lsi.uned.es

Curriculum Vitæ

Github


Fields of study

Machine Learning

Natural language processing

Computer Vision

Big Data

Algorithmics

Languages

Spanish

English



Hermenegildo Fabregat is currently a student at computer science Ph.D at the Deparment of "Lenguajes y Sistemas Informáticos (LSI)" at the National Distance Education University (UNED). He has been working as a researcher at LSI deparment at UNED since 2016.
In 2017, he obtained the MSc in Computer Science from the Complutense University of Madrid and in 2015, the degree in Computer Science from the Polytechnic University of Valencia.
His main research interest include Machine Learning and Text Mining applied to Biomedical datasets.

Education

National Distance Education University , Madrid, Spain
Current Expected 2020

Ph.D in Natural language processing


Complutense university of Madrid, Madrid, Spain
2015 - 2017

Master in Computer science

  • Thesis Topic: Improving and Classification of biomedical images using Automatic HDL Code Generation tool
  • Advisor: Guillermo Botella Juan
  • Advisor: Alberto del Barrio


Universitat Politècnica de València, Valencia, Spain
2011 - 2015

Bachelor’s Degree in Computer Engineering

  • Thesis Topic: Una aplicación de minería de datos para el análisis de la propiedad de terminación de SRTs
  • Advisor: Maria Jose Ramírez Quintana
  • Advisor: Javier Piris Ruano

Projects

EXTRECM Extracting Relations among Medical Concepts from Heteregenous Information Sources.

(Ministerio de Economía y Competitividad, TIN2013-46616-C2-2-R)


MAMTRA-MED Modelado, AutoMatización de exTracción de Relaciones cAtegorización de informes MEDicos para la recomendación de códigos CIE-10.

(Ministerio de Economía y Competitividad, TIN2016-77820-C3-2-R)


Publications

  • 2018, CMPB Deep neural models for extracting entities and relationships in the new RDD corpus relating disabilities and rare diseases.

    Abstract not available.

    • Hermenegildo Fabregat. Department of Computer Science, Universidad Nacional de Educación a Distancia (UNED)
    • Lourdes Araujo. Department of Computer Science, Universidad Nacional de Educación a Distancia (UNED)
    • Juan Martinez-Romo. Department of Computer Science, Universidad Nacional de Educación a Distancia (UNED)

Conferences

  • 2017, Summer Sim Simulation and implementation of a low-cost platform to improve the quality of biological images.

    The development of fluorescence microscopy techniques has helped advance in the understanding of molecular mechanisms present at biological systems, as the case of cell migration (Castillo-Lluva et al. 2010, Castillo-Lluva et al. 2013). These techniques allow studying the location, distribution as well as the movement of molecules that have been fluorescently tagged at a high resolution, even dealing with live cells (Rieckher 2017). In this paper a filtering and pattern recognition flow is simulated and implemented on a low-cost hardware platform in order to automate the detection of circular and ellipsoidal cells within the microscopic images. Results show that images keep a high quality after filtering and that 78.4% of the objects are properly recognized.

    • Guillermo Botella. Department of Computer Architecture and Automation, Complutense University of Madrid.
    • H. Fabregat. Complutense University of Madrid.
    • Alberto A. Del Barrio. Department of Computer Architecture and Automation, Complutense University of Madrid.

  • 2015, PROLE Analysing the Termination of Term Rewriting Systems using Data Mining.

    During the last decades, researchers in the field of Term Rewriting System (TRS) have devoted a lot of effort in order to develop techniques and methods able to demonstrate the termination property of a TRS. As a consequence, some of the proposed techniques have been implemented and several termination tools have been developed in order to automatize the termination proofs. From 2004, the annual Termination Competition is the forum in which research groups compare their tools trying to provide termination proofs of as many TRS as possible. This event generates a large amount of information (results obtained by the different tools, time spent on each proof, …) that is recorded in databases. In this paper, we propose an alternative approach to study the termination of TRS: to use data mining techniques that, based on the historical information collected in the competition, generate models to explore the termination of a TRS. The goal of our study is not to develop a termination tool but to show, for the first time, what machine learning techniques can offer to the analysis of TRS termination.

    • J. Piris. DSIC, Universitat Politecnica de Valéncia
    • H. Fabregat. DSIC, Universitat Politecnica de Valéncia
    • M.J. Ramírez-Quintana. DSIC, Universitat Politecnica de Valéncia

Workshops

  • 2018, DIANN Overview of the DIANN Task: Disability Annotation Task

    Abstract not available.

    • Hermenegildo Fabregat. Department of Computer Science, Universidad Nacional de Educación a Distancia (UNED)
    • Lourdes Araujo. Department of Computer Science, Universidad Nacional de Educación a Distancia (UNED)
    • Juan Martinez-Romo. Department of Computer Science, Universidad Nacional de Educación a Distancia (UNED)

  • 2018, DIANN UNED at DIANN 2018: Unsupervised System for Automatic Disabilities Labeling in Medical Scientific Documents

    Abstract not available.

    • Hermenegildo Fabregat. Department of Computer Science, Universidad Nacional de Educación a Distancia (UNED)
    • Lourdes Araujo. Department of Computer Science, Universidad Nacional de Educación a Distancia (UNED)
    • Juan Martinez-Romo. Department of Computer Science, Universidad Nacional de Educación a Distancia (UNED)