Fouille de textes et intelligence artificielle pour l'agriculture numérique

Bio : Mathieu Roche is a Senior Research Scientist (PhD, HDR) at CIRAD - TETIS research unit (France). Between 2005 and 2013, he was an Associate Professor (Maître de Conférences) at the University Montpellier 2, France. Mathieu Roche obtained a PhD in Computer Science from University Paris 11 (Orsay) in 2004. He defended his HDR (Habilitation à Diriger des Recherches - Accreditation to supervise research) in 2011. Mathieu Roche led several academic and industrial projects in text-mining. Currently he is involved in 2 European projects: H2020 MOOD – 2020-24 (Executive Board member) and Horizon CEA-First 2023-27 dealing with data science applied to epidemiology surveillance and food security. He has supervised 21 PhD students since 2006.

Description intervention : The ability to identify information in textual data is challenging for applications in the agriculture domain (e.g., agroecology, food security, epidemiology surveillance for animal and plant health). In this context, event-based surveillance (EBS) systems integrate text-mining and artificial intelligence methods (i.e. language models) to deal with huge amounts of textual data. This talk focuses on the use artificial intelligence and multidisciplinary approaches in order to mine heterogenous data dealing with the agriculture domain. The proposed methods associated with labeled textual datasets are integrated in the main steps of EBS systems: data acquisition, information retrieval (i.e., identification of relevant texts and topics), information extraction in textual content (i.e., locations, dates, themes), information to communicate to end-users.