LexMapr is a command line program for translating short biomedical specimen descriptions into ontology terms. Short texts often have very limited features available to build robust machine learning systems as compared to longer texts. LexMapr applies a two step transformation process, first cleaning up and normalizing text, and second, looking for direct matches of n-words in ontology labels and synonyms, or matches achieved by n-gram combinations of tokens that are likely to match ontology term labels and synonyms.
More information is available at https://github.com/Public-Health-Bioinformatics/LexMapr . As well, a slideshow examining LexMapr functionality is at: https://tinyurl.com/y3k5msqy.
A less-robust version of LexMapr is also available as a web application, at https://watson.bccdc.med.ubc.ca/lexmapr.