Enhancing LexSynonym Features in the Lexical Tools

Concept mapping is vital to natural language processing (NLP) for bioinformatics. Query expansion using synonyms for subterm substitutions is an effective technique to increase recall when no direct concept mapping can be found through normalization. For example, no concept can be found by direct mapping through normalization if the source vocabulary is “calcaneal fracture”. By substituting the subterm “calcaneal” for its synonym, “heel bone”, the matched UMLS concept [C0281926, fracture of calcaneus] is found. In this example, “calcaneal” and “heel bone” are element synonyms while “heel bone fracture” is the expanded term. The effectiveness of this technique relies on the completeness and quality of both the element synonyms and the UMLS synonym thesaurus in the query expansion pipeline of UMLS concept mapping (Figure 1).