Text mining tools for extracting information about microbial biodiversity in food

Area of Science:

• Food microbiology
• Bioinformatics
• Text mining

Background:

• Vast amounts of scientific literature on food microbial diversity are difficult to search.
• Researchers require efficient tools for bibliographic searches in large collections.
• Text mining and knowledge engineering offer solutions for information retrieval in Life Science.

Purpose of the Study:

• To apply the Alvis text mining platform to a large collection of PubMed abstracts in food microbiology.
• To extract specific information on microorganisms, their habitats, and phenotypes.
• To develop an online semantic search engine for food microbiology data.

Main Methods:

• Utilized the Alvis text mining platform on PubMed abstracts.
• Employed NCBI taxonomy and OntoBiotope ontology for information extraction.
• Indexed the text mining results for semantic search.

Main Results:

• Developed the AlvisIR Food online semantic search engine.
• Successfully extracted data on microorganisms, habitats, and phenotypes.
• Demonstrated the tool's potential through illustrative examples.

Conclusions:

• The AlvisIR Food tool significantly enhances the ability to search and analyze food microbiology literature.
• This semantic search engine aids in ecological diversity and microbial origin studies.
• Automated information extraction provides valuable insights into food microbial ecosystems.