Pavel Soriano Morales

           

I am a Computer Science PhD specialized on Natural Language Processing. I am currently working at Fanvoice, were we extract insights from crowdsource generated data. Specifically, we understand what people is talking about with ad-hoc topic modelling techniques.

My interests lie in the area of Natural Language Processing (NLP) and Text Mining. I like creating and using large networks extracted from text and using their linguistic features to create single, enriched representation spaces. Then, I try to come up with solutions to NLP tasks by leveraging the structure of my networks as well as combining the features within.

Also, during my PhD and industrial experience, I have become very interested in scientific programming and its related topics: parallel/distributed programming, testing, experimentation workflow, handling parameters, etc. I also like running sporadically and making bread.

Publications

2017

Edmundo-Pavel Soriano-Morales, Julien Ah-Pine, Sabine Loudcher
Fusion Techniques for Named Entity Recognition and Word Sense Induction and Disambiguation
Discovery Science 2017

2016

Edmundo-Pavel Soriano-Morales, Julien Ah-Pine, Sabine Loudcher
Hypergraph Modelization of a Syntactically Annotated English Wikipedia Dump
LREC 2016

Edmundo-Pavel Soriano-Morales, Julien Ah-Pine, Sabine Loudcher
Using a Heterogeneous Linguistic Network for Word Sense Induction and Disambiguation
CICLING 2016 (Computación y Sistemas 20)

Julien Ah-Pine, Edmundo-Pavel Soriano-Morales
A Study of Synthetic Oversampling for Twitter Imbalanced Sentiment Analysis
DMNLP 2016

Adrien Guille, Edmundo-Pavel Soriano-Morales
Topic modeling and hypergraph mining to analyze the EGC conference history
EGC 2016 (EGC Challenge winner)

Adrien Guille, Edmundo-Pavel Soriano-Morales
TOM: A library for topic modeling and browsing
EGC 2016

2015

Sabine Loudcher, Wararat Jakawat, Edmundo-Pavel Soriano-Morales, Cécile Favre
Combining OLAP and information networks for bibliographic data analysis: a survey
Scientometrics 103 2015

Master Thesis

Edmundo-Pavel Soriano Morales
Combining rule-based and statistical-based techniques in Sentiment Analysis

Teaching

I have been a teaching assistant as part of my activities as a PhD student and as a research and teaching assistant. Some of the courses I have created and teached are:

2016-2017

  • DUT Qualité Logistique Industrielle et Organisation (QLIO1)
    • Conception des systèmes d’informations
      CM1 CM2 CM3
  • Licence Coordinateur de Projets en Systèmes d’information (CPSI)
    • Les systèmes d’information décisionnels
      CM

2015-2016

  • European Master in Data Mining and Knowledge Management (M1)
  • DUT Statistique et Informatique Décisionnelle (STID2)
  • Licence d’Informatique Décisionnelle et d’Économétrie Appliquée (L1)
    • Logiciels Scientifiques
      TD

2014 - 2015

Workshops

Contact

ERIC Laboratory
Université Lyon 2
Bâtiment K
5 avenue Pierre Mendès France
69676 Bron, France

@:pavelsoriano[at]fastmail.fr