Pascal Neveu is a recognised leader in the field of managing massive and diverse data sets for agriculture and, like the conductor of a digital orchestra, he is able to guide his teams in new and innovative directions.
After completing his studies in applied mathematics, Neveu turned his attention to statistics. In 1983, he joined INRA as a consultant statistician at the Jouy-en-Josas Centre. He developed effective statistical methods to analyse data from his colleagues. Looking at everything from pigs to horses, and meat to milk, Neveu’s work and the people he worked with were very diverse. Quite quickly, Neveu became aware of the issues underlying the data being produced through research.
Neveu began to focus on computer science. He moved to Montpellier to work on artificial neural networks — digital computing systems inspired by the neural network of the human brain that are used to manage and drive data collection. In doing so, he became interested in artificial intelligence (AI).
Making sense of data
As systems to store, search and analyse digital information continue to evolve, “Big Data” has taken a significant place in the scientific community. Neveu has made it his mission to look at how massive data sets are acquired, managed, analysed and used, and the processes and tools necessary to treat, share and reuse this data.
“It’s a real issue for INRA. Increasingly collaborative research is giving rise to more, and more complex, data,” says Neveu. “The Internet also makes it possible to access sources of data that are faraway.”
At the start of the 2000s, Neveu took on the challenge of data. He sought ways to find and assemble data in order to add value to the information and, ultimately, answer major questions for science and society. “There was a lot of ground to cover as it was an emerging field that few people were working in,” he says.
Towards more intelligent systems
Starting in 2004, the MISTEA Unit began working with a number of information systems, including OpenSILEX, a unique, open-source digital tool for collecting, managing and mining agricultural and environmental data. The aim is to have data that can be shared and reused in future research and in other projects. To do so, it is necessary to describe, contextualise, annotate and link the data, and do so in a standardised way using what are known as FAIR principles for findability, accessibility, interoperability and reusability.
The people who made it possible, that's what's important
A number of information systems followed, including ALFIS, used in the study of alcoholic fermentation, PHIS, for plant phenotyping, and SILEX-LBE, for data management and online monitoring of organic processes to break down liquid effluent and solid waste. OpenSILEX became internationally recognised, and Neveu began offering tailored and turnkey services for people to use the tool. Demand for training and access to the tool grew as well, initiating a variety of partnerships and projects.
Neveu became deeply involved in teaching and training. He also became director of the MISTEA Unit, a role he took on with pleasure as he works to ensure his teams always grow and develop. But being director is not always easy. “Before becoming Director, I used to play guitar,” Neveu said jovially.
Over the years, Neveu and his work became more widely recognised. He is happy to receive the 2019 INRA Innovation Award, but dedicates the award to his colleagues, without whom, none of his achievements would have been possible.
61 years old
Since 2014 Director of the Mathematics, Informatics, Statistics for Environment and Agronomy Joint Research Unit (MISTEA) (INRA, Montpellier SupAgro) at the INRA Occitanie-Montpellier Research Centre
2009–2013 Associate director of the MISTEA Unit
2009–2014 Leader of the Management, Analysis and Modelling for Agronomy Data (GAMMA) team in the MISTEA Joint Research Unit
2012–2018 Manager at the Centre for Automation and Data Processing–Knowledge and Experimental Data (CATI CODEX)
Since 2006 Adjunct professor in the fields of algorithms, data management and scientific computing for universities and engineering schools
Since 2005 Research engineer, Mathematics and Applied Informatics from Genome to Environment Research Unit (MAIAGE) at INRA Jouy-en-Josas, then later at the MISTEA Joint Research Unit
1982 Associate’s degree in statistics, economic studies and quantitative management techniques
With his love of organisation, Neveu has spent a lot of time thinking about the future. In the medium term, Neveu has, amongst others, three goals. First is to make OpenSILEX more accessible to the research community and to private organisations. Next is to find logic in research data. And lastly, is to find ways to manage data in more analytical and context-specific ways. In partnership with the French Digital Agriculture Convergence Lab (#DigitAg), Neveu has been working with his teams towards these goals to meet the social and economic challenges of precision agriculture.
Neveu P. et al. 2018.Dealing with multi-source and multi-scale information in plant phenomics: the PHIS open-source Information. System. New Phytologist 221: 588.
Aceves-Lara C.A. et al. 2018. The virtual food system: Innovative models and experiential feedback in technologies for winemaking, the cereals chain, food packaging and eco-designed starter production. Innovative Food Science & Emerging Technologies 46: 54.
Muljarto A.R. et al. 2017. A generic ontological network for Agri-food experiment integration – Application to viticulture and winemaking. Computers and Electronics in Agriculture 140: 433.
Symeonidou D. et al. Key Discovery for Numerical Data: Application to Oenological Practices · pp. 222-236. In Graph-Based Representation and Reasoning - 23rd International Conference on Conceptual Structures, ICCS 2018, Edinburgh, UK, June 20-22, 2018, Proceedings. Chapman P., Endres D. & Pernelle N. (Eds.). Springer International Publishing AG, part of Springer Nature.
Pradal C. et al. 2017. InfraPhenoGrid: A scientific workflow infrastructure for plant phenomics on the Grid. Future Generation Computer System 67: 341.