Mathieu Roche

Content [Research]

Contact

UMR TETIS

Keywords

• Text Mining

• Natural Language Processing (NLP)

• Information Retrieval (IR)

• Information System (IS)

• AgroNLP

AgroNLP: Text Mining and Agriculture (see description of projects)

• Animal Disease Surveillance

• Food security

• Terminology Extraction for Document Matching and Open Data in Agrictural Domain

• Information Extraction from Experimental Data of Agricultural Domain

Work focusing on geospatial information extraction

• HERELLES project (2020-2024) [project supported by ANR - WP co-leader]: Image analysis based on textual information

• MOOD project (2020-2024) [project supported by European Union (H2020) - EB member]: Heterogeneous Data Science based on 3 dimensions: spatial, thematic, and temporal

• SONGES project (2016-2020) [project supported by Occitanie and European Union (FEDER) - leader]: Heterogeneous Data Science based on 3 dimensions: spatial, thematic, and temporal

• QDoSSI project (2016-2017) [project supported by CNRS (Mastodons)]: Quality of geospatial data dealing with international migration topic

• Senterritoire project (2012-2014) [project supported by MSH-M - leader]: Sentiment analysis and geospatial information

• ANIMITEX project (2013-2014) [project supported by CNRS (Mastodons) - leader]: Image analysis based on textual information

(see our survey written with B. Drury)

Other current work (see publications)

Sentiment Analysis
• Opinion Mining and classification (since 2007)
• Sentiment Analysis and Geospatial Information (since 2012)

Document Classification
• Classification of social Web data (blogs, tweets) (since 2007)
• Classification of poor and heterogeneous textual data (since 2007)

Named Entities
• Named Entity identification (since 2010)

Softwares

• KEOPS: Knowledge ExtractOr Pipeline System (since 2019)

KEOPS software:
- KEOPS (Web)
- Reference: RCIS'2021
Contact: Thierry Helmer (DSI), Mathieu Roche (TETIS), Pierre Martin (AIDA)

• Platform for Automated extraction of Disease Information from the web (since 2014)

PADI-Web software:
- PADI-Web (Web)
- Reference: Plos One
Contact: Mathieu Roche (TETIS) or Renaud Lancelot (ASTRE)

• Terminology extraction adapted to biomedical domain (since 2012)

BioTex software:
- BioTex (Web)
- BioTex (JAR - Linux, Mac, Windows)
- Reference: Information Retrieval Journal
Contact: Juan Antonio Lossio Ventura (National Institutes of Health (NIH) - USA)

• Terminology extraction in a text-mining process (since 2005)

EXIT software:
- download
- software manual
Contact: Thomas Heitz

• Other Softwares (since 2018)
- WEIR-P (Reference)
- Epid-News (Reference)
- Epid-Vis (Reference)
- Readitopics (Reference)
- Gemedoc (Reference)

Past work (see publications)

Information Retrieval

• Classification of job applications (2008-2010)
with R. Kessler (LIA - PhD student), J.M. Torres-Moreno (LIA), M. El-Bèze (LIA), N. Béchet (LIRMM - PhD student)

• Visualization of textual data in biomedicine (2009-2011)
with S. Bringay (LIRMM), M. Teisseire (UMR Tetis), A. Sallaberry (LABRI & PIKKO - PhD student)

Terminology Extraction and Disambiguation

• Acronym/expansion disambiguation (2007-2012)
with V. Prince (LIRMM), I. Mougenot (LIRMM)

• Terminology extraction from Old French data (2006-2007)
with C. Serp (Univ. Montpellier 3 - PhD student), E. Cazal (CNAM internship), M. Teisseire (UMR Tetis), A. Laurent (LIRMM)

• Conceptual class enrichment (2006-2009)
with N. Béchet (LIRMM - PhD student), J. Chauché (LIRMM)

• Automatic extraction of glosses (2006-2012)
with A. Mela (Univ. Montpellier 3), A. Steuckardt (LPL, Univ. of Provence)

Information Extraction

• Information extraction from log files (2008-2012)
with H. Saneifar (LIRMM/Satin-Technologies - PhD student), S. Bonniol (Satin-Technologies company), P. Poncelet (LIRMM)

String processing

• Schema matching (2005-2008)
with F. Duchateau (LIRMM - PhD student), Z. Bellahsene (LIRMM), F. Pinet (Cemagref)

• Trend detection (2007-2010)
with A. Laurent (LIRMM), B. Laurent (Namae Concept), S. Jaillet (CrysaLEAD)

Datawarehouse and Documents

• Extraction and aggregation of features for textual datawarehouses (since 2009)
with P. Poncelet (LIRMM), M. Teisseire (UMR TETIS), S. Bringay (LIRMM), F. Bouillot (CNAM internship), N. Béchet (INRIA post-doc)