| tech,0-2-J86-1002,ak | A method for <term> error correction </term> of <term> ill-formed input </term> is described that acquires <term> dialogue patterns </term> in typical usage and uses these <term> patterns </term> to predict new <term> inputs </term> . <term> Error correction </term> is done by strongly biasing <term> parsing </term> toward <term> expected meanings </term> unless clear evidence from the <term> input </term> shows the current <term> sentence </term> is not expected . | #16197 A method for error correction of ill-formed input is described that acquires dialogue patterns in typical usage and uses these patterns to predict new inputs. Error correction is done by strongly biasing parsing toward expected meanings unless clear evidence from the input shows the current sentence is not expected. |
