Cross-Lingual Keyword Extraction for Pesticide Terminology in Brazilian Portuguese and English
Journal
Journal of the Brazilian Computer Society
ISSN-Digital
1678-4804
Type
journal article
Date Issued
2025-10-09
Author(s)
De Souza, José Victor
;
Amamou, Hazem
;
Chen, Rubing
;
Salari, Elmira
;
;
;
Serpa, Talita
;
Lima, Marcela Marques de Freitas
;
Pinto, Paula Tavares
;
Kshirsagar, Shruti
;
Davoust, Alan
;
;
Avila, Anderson Raymundo
Abstract
Agriculture plays a crucial role in Brazil's economy. As the country intensifies its activities in the sector, the use of pesticides also increases. Hence, the risks associated with pesticide-laden food consumption have become a concern for chemistry researchers. An issue affecting regulatory standardization of pesticides in Brazil is the difficulty in translating pesticide names, particularly from English. For example, the word malathion can be translated from English to Portuguese as malatiom or malatião, resulting in inconsistent labeling. This issue extends to the broader problem of translating highly technical terms between languages, in particular for low-resource languages. In this work, we investigate terminological variation in the chemistry of organophosphorus pesticides. Our goal is to study strategies for domain-specific multilingual keyword extraction. To that end, two corpora were built based on pesticide-related scientific documents in Brazilian Portuguese and English, which led to a total of 84 and 210 texts, respectively, representing the low- and high-resource languages in this study. We then assessed 6 methods for keyword extraction: Simple Maths, TF-IDF, YAKE, TextRank, MultipartiteRank, and KeyBERT. We relied on a multilingual contextual BERT embedding to retrieve corresponding pesticide names in the target language. Fine-tuning was also explored to improve the multilingual representation further. Moreover, we evaluated the use of large language models (LLMs) combined with the recent retrieval-augmented generation (RAG) framework. As a result, we found that the contextual approach, combined with fine-tuning, provided the best results, contributing to enhancing Pesticide Terminology Extraction in a multilingual scenario.
Language
English
Keywords
Multilingual keyword extraction
word aligment
BERT embeddings
pesticides
HSG Classification
contribution to scientific community
Refereed
Yes
Publisher
Sociedade Brasileira de Computacao - SB
Volume
31
Number
1
Start page
972
End page
989
Subject(s)
File(s)![Thumbnail Image]()
Name
5815-Article Text-32536-1-10-20251009.pdf
Type
Main Article
Size
1.05 MB
Format
Adobe PDF
Checksum (MD5)
91ea6a32f523b00ef92e54cb4c4b5a73