Percorrer por autor "Silva, Lucas Ribeiro"
A mostrar 1 - 2 de 2
Resultados por página
Opções de ordenação
- Development of an intelligent agent for knowledge extraction in the pathogens in foods (PIF) database with machine learningPublication . Silva, Lucas Ribeiro; Alves, Paulo; Cadavez, VascoScientific databases like the Pathogens in Foods (PIF) Database hold valuable public health data but are often inaccessible to experts lacking programming skills. This research addresses this gap by developing and evaluating a novel Visual Natural Language Interface (V-NLI) for the PIF database. The resulting PIF Intelligent Agent empowers users to perform complex queries, conduct meta-analyses, and generate dynamic reports using natural language. The agent uses a hybrid, dual-mode architecture separating language interpretation from statistical computation. An "Open Chat Mode" offers a flexible exploratory interface via a tool-calling Small Language Model (SLM) with Retrieval-Augmented Generation (RAG). A "Guided Meta-Analysis Mode" provides a structured workflow for generating reproducible scientific reports through a dedicated Rserver backend. A comprehensive evaluation benchmarked five SLMs: Phi-4 Mini (3.8B), MFDoom/deepseek-r1-tool-calling (14B), Cogito (14B), Qwen 3 (8B), and Gemini 2.5 Pro. While all models achieved flawless functional accuracy, their effectiveness was determined by interpretive quality. The ability to generate concise, factually coherent text was the key differentiator, with smaller, instruction-tuned models showing performance comparable or superior in conciseness to larger models. The end-to-end system proved highly reliable, validating the architecture and establishing interpretive fidelity as a critical benchmark for domain-specific agents.
- Pathogens-in-Foods (PIF): An open-access European database of occurrence data of biological hazards in foodsPublication . Gonzales-Barron, Ursula; Faria, Ana Sofia; Thebault, Anne; Guillier, Laurent; Mendes, Lucas Ribeiro; Silva, Lucas Ribeiro; Messens, Winy; Kooh, Pauline; Cadavez, VascoThe collection of occurrence data of foodborne pathogens in foods faces the hindrances of dispersion of information, lack of standardisation and harmonisation, and ultimately, high expenditure in time and resources. The Pathogens-in-Foods (PIF) database was conceived as a solution to centralise published data on prevalence and concentration of pathogenic bacteria, viruses and parasites occurring in foods, obtained through systematic review (SR), and categorised in harmonised data structures under controlled terminologies. The present article outlines how PIF was constructed to adhere to the FAIR (findability, accessibility, interoperability and reusability) principles for scientific data management; and proceeds with a description of the PIF concept, which entails two phases: the SR process and the population of PIF. The protocolled SR process is supported by a welldefined search strategy, inclusion criteria, and rules for internal validation assessment; whereas the population of PIF with new data relies in data extraction, validation and release. The article then introduces a novel data quality approach, named as the CCC approach (data consistency, conformity and completeness), which ensures proper interpretation of data, richness of data, and flawless transcription of data. After a brief explanation of the three PIF components – database, back-end and front-end – the article proceeds with the exposition of the data model, as well as the capabilities of the front-end, including data search, insertion and curation. The future of PIF lies in expanding its capabilities, addressing emerging challenges, and leveraging technological advancements to maintain its relevance and utility in the evolving landscape of food safety.