Techniques for suggesting correct identifiers
First Claim
1. One or more tangible non-transitory computer-readable storage media for storing computer-executable instructions executable by processing logic, the media storing one or more instructions for:
- creating expected identifiers in a computing environment;
calculating frequencies of occurrence of the expected identifiers in the computing environment;
detecting one or more unrecognized identifiers received by the computing environment;
calculating first numerical scores using a string matching algorithm and a keystroke penalty matrix, where the first numerical scores indicate that one or more unrecognized identifiers were provided to the computing environment in place of one or more expected identifiers;
calculating second numerical scores using the first numerical scores, the frequencies of occurrence of the one or more expected identifiers, and Bayes theorem, where the second numerical scores indicate that one or more unrecognized identifiers were intended to include one or more expected identifiers; and
selecting one or more expected identifiers based on the second numerical scores.
1 Assignment
0 Petitions
Accused Products
Abstract
In an illustrative embodiment, an apparatus, computer-readable medium, or method may be configured to suggest correct identifiers. Expected identifiers may be created and their frequencies of occurrence may be calculated. Unrecognized identifiers may be detected. First numerical scores indicating that the unrecognized identifiers were provided in place of one or more expected identifiers may be calculated. Second numerical scores indicating that the unrecognized identifiers were intended to include expected identifiers may also be calculated. The second numerical scores may be calculated using the first numerical scores and the frequencies of occurrence of the expected identifiers. The system may select one or more expected identifiers based on the second set of numerical scores.
16 Citations
26 Claims
-
1. One or more tangible non-transitory computer-readable storage media for storing computer-executable instructions executable by processing logic, the media storing one or more instructions for:
-
creating expected identifiers in a computing environment; calculating frequencies of occurrence of the expected identifiers in the computing environment; detecting one or more unrecognized identifiers received by the computing environment; calculating first numerical scores using a string matching algorithm and a keystroke penalty matrix, where the first numerical scores indicate that one or more unrecognized identifiers were provided to the computing environment in place of one or more expected identifiers; calculating second numerical scores using the first numerical scores, the frequencies of occurrence of the one or more expected identifiers, and Bayes theorem, where the second numerical scores indicate that one or more unrecognized identifiers were intended to include one or more expected identifiers; and selecting one or more expected identifiers based on the second numerical scores. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. One or more tangible non-transitory computer-readable storage media for storing computer-executable instructions executable by processing logic, the media storing one or more instructions for:
-
creating a keystroke penalty matrix by performing a first operation on keystroke mistype probabilities in a computing environment; receiving textual input in the computing environment; calculating initial scores for expected identifiers in the computing environment, the calculating the initial scores comprising; performing a first operation on the initial scores, where the first operation includes; comparing the textual input and the expected identifiers to the keystroke penalty matrix; and calculating a final score for the expected identifiers, the calculating a final score comprising; performing a second operation on the initial scores in the computing environment, the second operation using Bayes theorem. - View Dependent Claims (18, 19, 20)
-
-
21. A computer-implemented method comprising:
-
creating expected identifiers in a computing environment by one or more processors; calculating frequencies of occurrence of the expected identifiers in the computing environment by the one or more processors; identifying one or more unrecognized identifiers in the computing environment by the one or more processors; calculating first numerical values in the computing environment by the one or more processors, where the first numerical values; indicate probabilities of inputting one or more unrecognized identifiers when intending to input one or more expected identifiers, and are calculated using; a string matching algorithm, and a penalty matrix; calculating second numerical values, where the second numerical values; indicate probabilities that one or more unrecognized identifiers were intended to be one or more expected identifiers, and are calculated using; a first theorem including Bays theorem, the frequencies of occurrence of the one or more identifiers, and the entries in the first plurality of numerical values; and selecting one or more expected identifiers in the computing environment by the one or more processors based on the second numerical values. - View Dependent Claims (22, 23, 24)
-
-
25. A system comprising:
-
a storage device; an input device; and a processor configured to; identify expected identifiers from the storage device; calculate frequencies of occurrence for the expected identifiers; receive a command from the input device; identify the command as including one or more unrecognized identifiers; compute first numerical values using a string matching algorithm and a keystroke penalty matrix, where the first numerical values; indicate probabilities of inputting one or more unrecognized identifiers when one or more expected identifiers were mistyped; compute second numerical values, where the second numerical values; indicate probabilities that one or more unrecognized identifiers were one or more mistyped expected identifiers, and are calculated using; the first numerical values, the frequencies of occurrence for the one or more expected identifiers, and a second algorithm using Bayes theorem; and identify one or more expected identifiers as the one or more unrecognized identifiers based on the second numerical values in the computing environment. - View Dependent Claims (26)
-
Specification