Pierre Silvie
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Pierre Silvie has a total of 73 co-authors. Here are the top co-authors by number of shared publications:
- Pierre Martin
CIRAD, UPR AIDA
6 shared publications - Marianne Huchard
LIRMM Univ. Montpellier CNRS Montpellier France
4 shared publications - Priscilla Keip
UPR AÏDA
4 shared publications - Thierry Brévault
CIRAD, UPR AIDA, Montpellier, France
3 shared publications Alain Gutierrez
2 shared publications- Alain Gutierrez
LIRMM Univ. Montpellier CNRS Montpellier France
2 shared publications
Latest Publications
- Open Access
- 10 October 2026
- Published by Elsevier in SSRN Electronic Journal
Hybrid Neuro-Symbolic AI for Incremental Discovery and Decision-Making in Plant-based Crop Protection
Designing agroecological crop protection strategies based on local resources remains a major challenge in the Global South. The Knomana initiative compiles scientific knowledge on plant-based alternatives to synthetic pesticides, describing protection systems that link a plant, a crop, and a pest, together with contextual information such as location, plant part, and extract type. To incrementally explore this multi-relational dataset and derive actionable insights, this paper presents a method enabling to explore the dataset, starting from its core model that is progressively enriched adding data tables, by presenting underlying regularities and generalizations expressed in natural language. To do so, this method combines the multi-relational extension (RCA) of Formal Concept Analysis (FCA), to extract the Duquenne-Guigues base of implications for each grouping of tables of the dataset, with a Large Language Model (LLM), that rephrases implications into natural language. A case study on Brassicaceae protection evaluates LLM-generated outputs against the judgments of computer scientists and entomologists. Qualitative and quantitative results show that hybridizing RCA and LLMs reliably preserves the formal content of the implications for logical validation, while making them more accessible to domain experts -- an appropriation that is already encouraging, though intelligibility and usefulness still leave room for improvement. Building on this, the proposed incremental exploration helps experts grasp the complexity of the dataset step by step. This work paves the way for AI-based decision-making on agroecology grounded in trade-offs.
- 10 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessHybrid Neuro-Symbolic AI for Incremental Discovery and Decision-Making in Plant-based Crop Protection
Designing agroecological crop protection strategies based on local resources remains a major challenge in the Global South. The Knomana initiative compiles scientific knowledge on plant-based alternatives to synthetic pesticides, describing protection systems that link a plant, a crop, and a pest, together with contextual information such as location, plant part, and extract type. To incrementally explore this multi-relational dataset and derive actionable insights, this paper presents a method enabling to explore the dataset, starting from its core model that is progressively enriched adding data tables, by presenting underlying regularities and generalizations expressed in natural language. To do so, this method combines the multi-relational extension (RCA) of Formal Concept Analysis (FCA), to extract the Duquenne-Guigues base of implications for each grouping of tables of the dataset, with a Large Language Model (LLM), that rephrases implications into natural language. A case study on Brassicaceae protection evaluates LLM-generated outputs against the judgments of computer scientists and entomologists. Qualitative and quantitative results show that hybridizing RCA and LLMs reliably preserves the formal content of the implications for logical validation, while making them more accessible to domain experts -- an appropriation that is already encouraging, though intelligibility and usefulness still leave room for improvement. Building on this, the proposed incremental exploration helps experts grasp the complexity of the dataset step by step. This work paves the way for AI-based decision-making on agroecology grounded in trade-offs.
- 20 March 2023
- Published by Wiley in Journal of Applied Entomology
- Vol. 147 (5) , 299-306
Effect of maize seed treatment on oviposition preference, larval performance and foliar damage of the fall armyworm
The fall armyworm (FAW), Spodoptera frugiperda (J. E. Smith; Lepidoptera: Noctuidae), is an invasive pest that has recently become a major worldwide threat for maize production. Greatest crop yield reductions occur following damage by FAW larvae at early vegetative plant stages, between the first and second week after plant emergence. In this study, field and semi‐field experiments were carried out to assess the effect of cyantraniliprole‐based seed treatment on the protection of early stages of maize plants against FAW and the associated mechanisms. In the field, maize seed treatment using cyantraniliprole (Fortenza®) significantly reduced the percentage of infested seedlings (77%) and the percentage of seedlings with severe damage (86%) up to 3 weeks after emergence. In a choice situation, semi‐field experiments showed that gravid moths preferred to lay eggs on plants from untreated seeds (68%) rather than on plants from treated seeds. To the best of our knowledge, this is the first evidence that insecticide residues present in plant tissues deter oviposition of Noctuid moths. Seed treatment with cyantraniliprole also reduced plant infestation (65%) 10 days after artificial inoculation of neonate larvae at 3 weeks after plant emergence, reducing the severity of leaf damage (by 67%). Insecticide seed treatment is a relevant tool for the integrated pest management of the FAW in high‐value crops such as sweet corn, especially for reducing foliar applications of insecticides during the most susceptible stages of the crop, as far as the risk of impacts on non‐target organisms is low and no alternative biocontrol option is available.
- 20 March 2023
- Published by Wiley in Journal of Applied Entomology
- Vol. 147 (5) , 299-306
Effect of maize seed treatment on oviposition preference, larval performance and foliar damage of the fall armyworm
The fall armyworm (FAW), Spodoptera frugiperda (J. E. Smith; Lepidoptera: Noctuidae), is an invasive pest that has recently become a major worldwide threat for maize production. Greatest crop yield reductions occur following damage by FAW larvae at early vegetative plant stages, between the first and second week after plant emergence. In this study, field and semi‐field experiments were carried out to assess the effect of cyantraniliprole‐based seed treatment on the protection of early stages of maize plants against FAW and the associated mechanisms. In the field, maize seed treatment using cyantraniliprole (Fortenza®) significantly reduced the percentage of infested seedlings (77%) and the percentage of seedlings with severe damage (86%) up to 3 weeks after emergence. In a choice situation, semi‐field experiments showed that gravid moths preferred to lay eggs on plants from untreated seeds (68%) rather than on plants from treated seeds. To the best of our knowledge, this is the first evidence that insecticide residues present in plant tissues deter oviposition of Noctuid moths. Seed treatment with cyantraniliprole also reduced plant infestation (65%) 10 days after artificial inoculation of neonate larvae at 3 weeks after plant emergence, reducing the severity of leaf damage (by 67%). Insecticide seed treatment is a relevant tool for the integrated pest management of the FAW in high‐value crops such as sweet corn, especially for reducing foliar applications of insecticides during the most susceptible stages of the crop, as far as the risk of impacts on non‐target organisms is low and no alternative biocontrol option is available.
- Open Access
- 1 January 2023
- Published by FapUNIFESP (SciELO) in Brazilian Archives of Biology and Technology
- Vol. 66
Cut off Behavior of Spodoptera frugiperda (Smith, 1797) (Lepidoptera: Noctuidae) in Soybean (Glycine max (L.) Merrill) Seedlings
- 1 January 2023
- Published by FapUNIFESP (SciELO) in Brazilian Archives of Biology and Technology
- Vol. 66
Open AccessCut off Behavior of Spodoptera frugiperda (Smith, 1797) (Lepidoptera: Noctuidae) in Soybean (Glycine max (L.) Merrill) Seedlings
- 14 September 2022
- Published by Taylor & Francis in Botany Letters
- Vol. 170 (1) , 28-41
Plant-based extracts for cotton pest management in Sub-Saharan Africa: a review
Cotton crops are hampered by a complex of arthropod species, insects and mites that attack all plant organs. In sub-Saharan Africa, crop pest control mainly involves synthetic pesticide applications based on a treatment calendar consisting of 5–6 applications at 14-day intervals. Organic cotton production chains have however been set up to supply a niche market, with botanical extracts used for crop protection. This review presents published results of experiments conducted in sub-Saharan African countries. A total of 37 plant species were laboratory tested in research stations (24 trials) and village fields in Central and West African countries. Neem (Azadirachta indica) was the plant most commonly used in the form of seed-pressed oil. In Mali, this compound was combined with Carapa procera oil in organic cotton cropping conditions. Cassia nigricans is another species reported to be used in Mali, Burkina Faso and Benin. Aromatic species such as Ocimum gratissimum, Cymbopogon citratus, Citrus sp. were also tested. Further research could focus on species such as Anacardium occidentale and Hyptis suaveolens, which seem to be locally available. The use of essential oils does not appear to take priority over aqueous extracts, which could be tested in crop protection programmes involving partial substitution of synthetic insecticides.
- 14 September 2022
- Published by Taylor & Francis in Botany Letters
- Vol. 170 (1) , 28-41
Plant-based extracts for cotton pest management in Sub-Saharan Africa: a review
Cotton crops are hampered by a complex of arthropod species, insects and mites that attack all plant organs. In sub-Saharan Africa, crop pest control mainly involves synthetic pesticide applications based on a treatment calendar consisting of 5–6 applications at 14-day intervals. Organic cotton production chains have however been set up to supply a niche market, with botanical extracts used for crop protection. This review presents published results of experiments conducted in sub-Saharan African countries. A total of 37 plant species were laboratory tested in research stations (24 trials) and village fields in Central and West African countries. Neem (Azadirachta indica) was the plant most commonly used in the form of seed-pressed oil. In Mali, this compound was combined with Carapa procera oil in organic cotton cropping conditions. Cassia nigricans is another species reported to be used in Mali, Burkina Faso and Benin. Aromatic species such as Ocimum gratissimum, Cymbopogon citratus, Citrus sp. were also tested. Further research could focus on species such as Anacardium occidentale and Hyptis suaveolens, which seem to be locally available. The use of essential oils does not appear to take priority over aqueous extracts, which could be tested in crop protection programmes involving partial substitution of synthetic insecticides.
- 1 January 2022
- Published by OpenEdition
- p. 683-699
Chapitre 46. Le management des connaissances liées aux usages des plantes
- 1 January 2022
- Published by OpenEdition
- p. 683-699
Chapitre 46. Le management des connaissances liées aux usages des plantes
- 8 December 2021
- Published by Springer Nature
- p. 105-134
Dealing with Large Volumes of Complex Relational Data Using RCA
- 8 December 2021
- Published by Springer Nature
- p. 105-134
Dealing with Large Volumes of Complex Relational Data Using RCA
- Open Access
Geographic Monitoring of Insecticide Resistance Mutations in Native and Invasive Populations of the Fall Armyworm
Field evolved resistance to insecticides is one of the main challenges in pest control. The fall armyworm (FAW) is a lepidopteran pest species causing severe crop losses, especially corn. While native to the Americas, the presence of FAW was confirmed in West Africa in 2016. Since then, the FAW has been detected in over 70 countries covering sub-Saharan Africa, the Middle East, North Africa, South Asia, Southeast Asia, and Oceania. In this study, we tested whether this invasion was accompanied by the spread of resistance mutations from native to invasive areas. We observed that mutations causing Bt resistance at ABCC2 genes were observed only in native populations where the mutations were initially reported. Invasive populations were found to have higher gene numbers of cytochrome P450 genes than native populations and a higher proportion of multiple resistance mutations at acetylcholinesterase genes, supporting strong selective pressure for resistance against synthetic insecticides. This result explains the susceptibility to Bt insecticides and resistance to various synthetic insecticides in Chinese populations. These results highlight the necessity of regular and standardized monitoring of insecticide resistance in invasive populations using both genomic approaches and bioassay experiments.
Open AccessGeographic Monitoring of Insecticide Resistance Mutations in Native and Invasive Populations of the Fall Armyworm
Field evolved resistance to insecticides is one of the main challenges in pest control. The fall armyworm (FAW) is a lepidopteran pest species causing severe crop losses, especially corn. While native to the Americas, the presence of FAW was confirmed in West Africa in 2016. Since then, the FAW has been detected in over 70 countries covering sub-Saharan Africa, the Middle East, North Africa, South Asia, Southeast Asia, and Oceania. In this study, we tested whether this invasion was accompanied by the spread of resistance mutations from native to invasive areas. We observed that mutations causing Bt resistance at ABCC2 genes were observed only in native populations where the mutations were initially reported. Invasive populations were found to have higher gene numbers of cytochrome P450 genes than native populations and a higher proportion of multiple resistance mutations at acetylcholinesterase genes, supporting strong selective pressure for resistance against synthetic insecticides. This result explains the susceptibility to Bt insecticides and resistance to various synthetic insecticides in Chinese populations. These results highlight the necessity of regular and standardized monitoring of insecticide resistance in invasive populations using both genomic approaches and bioassay experiments.
- Open Access
Prototyping a Knowledge-Based System to Identify Botanical Extracts for Plant Health in Sub-Saharan Africa
Replacing synthetic pesticides and antimicrobials with plant-based extracts is a current alternative adopted by traditional and family farmers and many organic farming pioneers. A range of natural extracts are already being marketed for agricultural use, but many other plants are prepared and used empirically. A further range of plant species that could be effective in protecting different crops against pests and diseases in Africa could be culled from the large volume of knowledge available in the scientific literature. To meet this challenge, data on plant uses have been compiled in a knowledge base and a software prototype was developed to navigate this trove of information. The present paper introduces this so-called Knomana Knowledge-Based System, while providing outputs related to Spodoptera frugiperda and Tuta absoluta, two invasive insect species in Africa. In early October 2020, the knowledge base hosted data obtained from 342 documents. From these articles, 11,816 uses—experimental or applied by farmers—were identified in the plant health field. In total, 384 crop pest species are currently reported in the knowledge base, in addition to 1547 botanical species used for crop protection. Future prospects for applying this interdisciplinary output to applications under the One Health approach are presented.
Open AccessPrototyping a Knowledge-Based System to Identify Botanical Extracts for Plant Health in Sub-Saharan Africa
Replacing synthetic pesticides and antimicrobials with plant-based extracts is a current alternative adopted by traditional and family farmers and many organic farming pioneers. A range of natural extracts are already being marketed for agricultural use, but many other plants are prepared and used empirically. A further range of plant species that could be effective in protecting different crops against pests and diseases in Africa could be culled from the large volume of knowledge available in the scientific literature. To meet this challenge, data on plant uses have been compiled in a knowledge base and a software prototype was developed to navigate this trove of information. The present paper introduces this so-called Knomana Knowledge-Based System, while providing outputs related to Spodoptera frugiperda and Tuta absoluta, two invasive insect species in Africa. In early October 2020, the knowledge base hosted data obtained from 342 documents. From these articles, 11,816 uses—experimental or applied by farmers—were identified in the plant health field. In total, 384 crop pest species are currently reported in the knowledge base, in addition to 1547 botanical species used for crop protection. Future prospects for applying this interdisciplinary output to applications under the One Health approach are presented.
- Open Access
Cabbage Production in West Africa and IPM with a Focus on Plant-Based Extracts and a Complementary Worldwide Vision
In urban and peri-urban areas in West Africa, the cabbage Brassica oleracea L. (Brassicaceae) is protected using repeated high doses of synthetic insecticides. After a brief description of available IPM components, this paper presents a literature review focused on the botanical extracts that have been experimented with at the laboratory or in the field in West Africa against major cabbage pests. The literature reviewed mentions 19 plant species from 12 families used for cabbage protection in the subregion. The species most used are Azadirachta indica, Capsicum frutescens, Ocimum gratissimum and Ricinus communis. An overview of the world literature showed that a total of 13 plant species belonging to 8 families used to control cabbage pests are reported from the rest of Africa, and 140 plant species belonging to 43 families from the rest of the world. The most commonly used and tested plant species against insect pests in the three geographical areas considered is A. indica.
Open AccessCabbage Production in West Africa and IPM with a Focus on Plant-Based Extracts and a Complementary Worldwide Vision
In urban and peri-urban areas in West Africa, the cabbage Brassica oleracea L. (Brassicaceae) is protected using repeated high doses of synthetic insecticides. After a brief description of available IPM components, this paper presents a literature review focused on the botanical extracts that have been experimented with at the laboratory or in the field in West Africa against major cabbage pests. The literature reviewed mentions 19 plant species from 12 families used for cabbage protection in the subregion. The species most used are Azadirachta indica, Capsicum frutescens, Ocimum gratissimum and Ricinus communis. An overview of the world literature showed that a total of 13 plant species belonging to 8 families used to control cabbage pests are reported from the rest of Africa, and 140 plant species belonging to 43 families from the rest of the world. The most commonly used and tested plant species against insect pests in the three geographical areas considered is A. indica.