Generating an unambiguous interpretation of ingested content to produce knowledge
A method includes identifying identigens for each word of a trusted phrase, where the identigens of a first word of the trusted phrase includes one or more different meanings of the first word. The method further includes identifying two adjacent words of the trusted phrase that are lacking an ident...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Patent |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Zusammenfassung: | A method includes identifying identigens for each word of a trusted phrase, where the identigens of a first word of the trusted phrase includes one or more different meanings of the first word. The method further includes identifying two adjacent words of the trusted phrase that are lacking an identigen pairing and determining whether the identigen pairing can be inferred for the two adjacent words based on known identigen pairings of other words of the trusted phrase. When the identigen pairing can be inferred for the two adjacent words, the method further includes adding to a knowledge database the identigen pairing of the two adjacent words. |
|---|