Species Name Identification for Essential Oils from Biomedical Abstracts Using Text Mining and Natural Language Processing

Abstract

In recent years, the publication of scientific papers related to essential oils has achieved exponential growth due to the popularity of aromatherapy, although no studies using natural language processing and text mining methods to extract information from scientific articles related to essential oils are currently found. Accordingly, this study is the first to use natural language processing and text mining methods to identify species names appearing in abstracts related to essential oils. We obtained 34,637 abstracts using keywords, “essential oil” to quarry PubMed on 2024/03/15. The 1,081,005 species names of plants and fungi were obtained from Taxonomy FTP on the same day. The nouns from titles of articles related to essential oils were obtained via identification of parts-of-speech and from titles and abstracts extracted within italicized labels. These nouns were used to identify 10,445 plant and fungal species names downloaded from FTP appearing in abstracts related to essential oils with these identification terms being used to detect whether abstracts related to essential oils revealed the species names. 156,371 records contained links between PMID and Taxonomy ID. To the best of our knowledge, our study shows this method can efficiently identify the names of species from abstracts related to essential oil.