Analysing the potential of ChatGPT to support plant disease risk forecasting systems

This study explores the potential of two versions of the ChatGPT large language model (GPT-3.5 Turbo and GPT-4) in supporting plant disease risk forecasting through the translation of model-based predictions into advisory messages. A dataset of 3125 messages, each referred to an artificially generat...

Full description

Saved in:
Bibliographic Details
Main Authors: Roberta Calone, Elisabetta Raparelli, Sofia Bajocco, Eugenio Rossi, Lorenzo Crecco, Danilo Morelli, Chiara Bassi, Rocchina Tiso, Riccardo Bugiani, Fabio Pietrangeli, Giovanna Cattaneo, Camilla Nigro, Marco Secondo Gerardi, Simone Bussotti, Angela Sanchioni, Danilo Tognetti, Mariangela Sandra, Irene De Lillo, Paolo Framarin, Sandra Di Ferdinando, Simone Bregaglio
Format: Article
Language:English
Published: Elsevier 2025-03-01
Series:Smart Agricultural Technology
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2772375525000589
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850054101304344576
author Roberta Calone
Elisabetta Raparelli
Sofia Bajocco
Eugenio Rossi
Lorenzo Crecco
Danilo Morelli
Chiara Bassi
Rocchina Tiso
Riccardo Bugiani
Fabio Pietrangeli
Giovanna Cattaneo
Camilla Nigro
Marco Secondo Gerardi
Simone Bussotti
Angela Sanchioni
Danilo Tognetti
Mariangela Sandra
Irene De Lillo
Paolo Framarin
Sandra Di Ferdinando
Simone Bregaglio
author_facet Roberta Calone
Elisabetta Raparelli
Sofia Bajocco
Eugenio Rossi
Lorenzo Crecco
Danilo Morelli
Chiara Bassi
Rocchina Tiso
Riccardo Bugiani
Fabio Pietrangeli
Giovanna Cattaneo
Camilla Nigro
Marco Secondo Gerardi
Simone Bussotti
Angela Sanchioni
Danilo Tognetti
Mariangela Sandra
Irene De Lillo
Paolo Framarin
Sandra Di Ferdinando
Simone Bregaglio
author_sort Roberta Calone
collection DOAJ
description This study explores the potential of two versions of the ChatGPT large language model (GPT-3.5 Turbo and GPT-4) in supporting plant disease risk forecasting through the translation of model-based predictions into advisory messages. A dataset of 3125 messages, each referred to an artificially generated five-day disease risk scenario, was inspected using lexical, consistency, and sentiment analyses. A participatory approach was adopted involving officers and technicians from eleven Italian regional phytosanitary services in the message evaluation. Lexical analysis indicated that GPT-4 produced more detailed advises, leading to diversified responses across disease pressure scenarios. In contrast, GPT-3.5 generated concise and straightforward messages, making it well-suited for routine tasks requiring clarity and brevity. Sentiment analysis revealed the greater adaptability of GPT-4 in shifting from a reassurance to an urgency tone as the risk level increased, while GPT-3.5 maintained a more neutral stance across disease pressure scenarios. Consistency analysis demonstrated greater stability in GPT-3.5 messages, whereas GPT-4 exhibited more expressivity and creativity. Expert evaluations highlighted promising potential of both models for operational use, with GPT-4 noted for its precision in communicating disease risk and supporting technical bulletins drafting. However, both GPT versions were criticized for producing overly generic advices, highlighting the importance of domain-specific training to integrate information on best management practices, locally authorized substances, and historical treatment schedules. Such implementation is crucial to align large language models with Integrated Pest Management principles and improve their precision towards their operational use.
format Article
id doaj-art-d31494b8b0c94e7ba75bef9993df95bf
institution DOAJ
issn 2772-3755
language English
publishDate 2025-03-01
publisher Elsevier
record_format Article
series Smart Agricultural Technology
spelling doaj-art-d31494b8b0c94e7ba75bef9993df95bf2025-08-20T02:52:21ZengElsevierSmart Agricultural Technology2772-37552025-03-011010082410.1016/j.atech.2025.100824Analysing the potential of ChatGPT to support plant disease risk forecasting systemsRoberta Calone0Elisabetta Raparelli1Sofia Bajocco2Eugenio Rossi3Lorenzo Crecco4Danilo Morelli5Chiara Bassi6Rocchina Tiso7Riccardo Bugiani8Fabio Pietrangeli9Giovanna Cattaneo10Camilla Nigro11Marco Secondo Gerardi12Simone Bussotti13Angela Sanchioni14Danilo Tognetti15Mariangela Sandra16Irene De Lillo17Paolo Framarin18Sandra Di Ferdinando19Simone Bregaglio20CREA - Council for Agricultural Research and Economics, Research Centre for Agriculture and Environment, Bologna BO I-40128, Rome I-00184 , Italy; Corresponding author.CREA - Council for Agricultural Research and Economics, Research Centre for Agriculture and Environment, Bologna BO I-40128, Rome I-00184 , ItalyCREA - Council for Agricultural Research and Economics, Research Centre for Agriculture and Environment, Bologna BO I-40128, Rome I-00184 , ItalyCREA - Council for Agricultural Research and Economics, Research Centre for Agriculture and Environment, Bologna BO I-40128, Rome I-00184 , ItalyCREA - Council for Agricultural Research and Economics, Research Centre for Agriculture and Environment, Bologna BO I-40128, Rome I-00184 , ItalyCREA - Council for Agricultural Research and Economics, Research Centre for Agriculture and Environment, Bologna BO I-40128, Rome I-00184 , ItalyServizi Alle Imprese Agricole e Florovivaismo, CAAR (Centro Agrometeorologia Applicata Regionale), Laboratori Regionali Analisi Terreni-Produzioni Vegetali e Fitopatologico, Liguria Region, Sarzana SP I-19038, ItalyPlant Protection Service, Emilia-Romagna Region, Via A. da Formigine 3, Bologna BO I-40129, ItalyPlant Protection Service, Emilia-Romagna Region, Via A. da Formigine 3, Bologna BO I-40129, ItalyServizio Supporto Specialistico all'Agricoltura, Via Nazionale 37, Villanova di Cepagatti, Pescara PE I-65012, ItalyRegione Lombardia, Plant Protection Service, Milan MI I-20124, ItalyLucana Agency for Development and Innovation in Agriculture, Basilicata Region, Via Annunziatella 64, Matera MT I-75100, ItalyLAORE Sardegna, Regional Agency for Agriculture Development, Via Caprera 8, Cagliari CA I-09123, ItalyFondazione Agrion, Via Falicetto, 24, Manta CN I-12030, ItalyAMAP - Agenzia per l'innovazione nel settore agroalimentare e della pesca ''Marche Agricoltura Pesca'', Osimo AN I-60027, ItalyAMAP - Agenzia per l'innovazione nel settore agroalimentare e della pesca ''Marche Agricoltura Pesca'', Osimo AN I-60027, ItalyServizio fitosanitario e chimico, ricerca, sperimentazione e assistenza tecnica. Agenzia regionale per lo sviluppo rurale del Friuli-Venezia Giulia – ERSA, Via Sabbatini 5, Pozzuolo del Friuli UD I-33050, ItalyARPAV. Dipartimento Regionale per La Sicurezza Del Territorio. U.O.C. Meteorologia e Climatologia, Veneto Region, Via Marconi 55, Teolo PD I-35037, ItalyRegione Veneto - Unità Organizzativa Fitosanitario, Viale dell'Agricoltura 1/A, Buttapietra VR I-37060, ItalyARSIAL – Agenzia Regionale per lo Sviluppo e l'Innovazione dell'Agricoltura del Lazio, Area Qualità e Pianificazione Territoriale, Via Rodolfo Lanciani 38, Roma I–00162, ItalyCREA - Council for Agricultural Research and Economics, Research Centre for Agriculture and Environment, Bologna BO I-40128, Rome I-00184 , ItalyThis study explores the potential of two versions of the ChatGPT large language model (GPT-3.5 Turbo and GPT-4) in supporting plant disease risk forecasting through the translation of model-based predictions into advisory messages. A dataset of 3125 messages, each referred to an artificially generated five-day disease risk scenario, was inspected using lexical, consistency, and sentiment analyses. A participatory approach was adopted involving officers and technicians from eleven Italian regional phytosanitary services in the message evaluation. Lexical analysis indicated that GPT-4 produced more detailed advises, leading to diversified responses across disease pressure scenarios. In contrast, GPT-3.5 generated concise and straightforward messages, making it well-suited for routine tasks requiring clarity and brevity. Sentiment analysis revealed the greater adaptability of GPT-4 in shifting from a reassurance to an urgency tone as the risk level increased, while GPT-3.5 maintained a more neutral stance across disease pressure scenarios. Consistency analysis demonstrated greater stability in GPT-3.5 messages, whereas GPT-4 exhibited more expressivity and creativity. Expert evaluations highlighted promising potential of both models for operational use, with GPT-4 noted for its precision in communicating disease risk and supporting technical bulletins drafting. However, both GPT versions were criticized for producing overly generic advices, highlighting the importance of domain-specific training to integrate information on best management practices, locally authorized substances, and historical treatment schedules. Such implementation is crucial to align large language models with Integrated Pest Management principles and improve their precision towards their operational use.http://www.sciencedirect.com/science/article/pii/S2772375525000589Large Language ModelsIntegrated Pest ManagementAdvisory MessagesStakeholders EngagementLexical AnalysisSentiment Analysis
spellingShingle Roberta Calone
Elisabetta Raparelli
Sofia Bajocco
Eugenio Rossi
Lorenzo Crecco
Danilo Morelli
Chiara Bassi
Rocchina Tiso
Riccardo Bugiani
Fabio Pietrangeli
Giovanna Cattaneo
Camilla Nigro
Marco Secondo Gerardi
Simone Bussotti
Angela Sanchioni
Danilo Tognetti
Mariangela Sandra
Irene De Lillo
Paolo Framarin
Sandra Di Ferdinando
Simone Bregaglio
Analysing the potential of ChatGPT to support plant disease risk forecasting systems
Smart Agricultural Technology
Large Language Models
Integrated Pest Management
Advisory Messages
Stakeholders Engagement
Lexical Analysis
Sentiment Analysis
title Analysing the potential of ChatGPT to support plant disease risk forecasting systems
title_full Analysing the potential of ChatGPT to support plant disease risk forecasting systems
title_fullStr Analysing the potential of ChatGPT to support plant disease risk forecasting systems
title_full_unstemmed Analysing the potential of ChatGPT to support plant disease risk forecasting systems
title_short Analysing the potential of ChatGPT to support plant disease risk forecasting systems
title_sort analysing the potential of chatgpt to support plant disease risk forecasting systems
topic Large Language Models
Integrated Pest Management
Advisory Messages
Stakeholders Engagement
Lexical Analysis
Sentiment Analysis
url http://www.sciencedirect.com/science/article/pii/S2772375525000589
work_keys_str_mv AT robertacalone analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT elisabettaraparelli analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT sofiabajocco analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT eugeniorossi analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT lorenzocrecco analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT danilomorelli analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT chiarabassi analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT rocchinatiso analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT riccardobugiani analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT fabiopietrangeli analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT giovannacattaneo analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT camillanigro analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT marcosecondogerardi analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT simonebussotti analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT angelasanchioni analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT danilotognetti analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT mariangelasandra analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT irenedelillo analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT paoloframarin analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT sandradiferdinando analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems
AT simonebregaglio analysingthepotentialofchatgpttosupportplantdiseaseriskforecastingsystems