Table 3 NOBLE matching options with examples
From: NOBLE – Flexible concept recognition for large-scale biomedical natural language processing
Matching feature | Explanation | Example input text | Example input vocabulary | Example Output based on ParameterSetting (T, F) | |
|---|---|---|---|---|---|
TRUE | FALSE | ||||
Subsumption | Only more comprehensive concepts are coded | “Deep margin” | Deep | Deep margin | Deep |
Margin | Margin | ||||
Deep margin | Deep margin | ||||
Overlap | A mapped concept may be fully or partially within the boundaries of another concept | “Deep lateral margin” | Deep | Deep | Deep |
Lateral margin | Lateral margin | Lateral margin | |||
Deep margin | Deep margin | ||||
Contiguity | Words in text that map to concept term must be continuous (within word gap) | “Deep lateral margin” | Deep | Deep | Deep |
Margin | Margin | Margin | |||
Deep margin | Deep margin | ||||
Order | Order of words in text must be the same as in the concept term | “Margin, deep” | Deep | Deep | Deep |
Margin | Margin | Margin | |||
Deep margin | Deep margin | ||||
Partial | Input text that only partially matches the concept term is coded | “Margin” | Deep margin | Deep margin | No concepts coded |