Character input error correction

• US 10,394,344 B2
• Filed: 11/07/2017
• Issued: 08/27/2019
• Est. Priority Date: 11/07/2017
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: D, Manjunatha, Shanmugam, Dhandapani, Venkataramanachari, Yethish G., Adoni, Siddique M., Mitra, Suman
• Primary Examiner(s): Cheng, Joe H
• Application Number:
  US15/805,228
• Publication Number:
  US 20190138113A1
• Time in Patent Office: 658 Days
• CPC Class Codes: G06F 3/0238 Programmable keyboards key guide holders G06F3/0224
• CPC Class Codes: G06F 3/04886 by partitioning the display area of the touch-screen or the surface of the digitising tablet into independently controllable areas, e.g. virtual keyboards or menus
• CPC Class Codes: G06F 3/04895 Guidance during keyboard input operation, e.g. prompting
• CPC Class Codes: G06F 40/232 Orthographic correction, e.g. spell checking or vowelisation
• CPC Class Codes: G09B 21/00 Teaching, or communicating with, the blind, deaf or mute audible presentation of material to be studied G09B5/04; devices or methods for replacing direct visual or auditory perception by another kind of perception A61F9/08, A61F11/04; audible indication of meter readings or of colour G01D7/12; watches for blind persons G04B25/02; speech analysis, speech recognition G10L; sound-recording or reproducing, per se G11B

Abstract

Character input error correction is provided by receiving from a keyboard character input by a user. The character input includes character input errors by the user, the character input errors being input of incorrect characters by the user via the keyboard. The character input errors of the character input are automatically identified and classified into character error classification(s). Each character error classification corresponds to a respective visual transformation of input characters. It is determined that a frequency of the identified and classified character input errors exceeds a predefined threshold confidence level for triggering application of keyboard corrections for user input. Based on the determining, keyboard corrections are applied for user input. The applying keyboard corrections includes applying a correction to character(s) displayed for or input by the user. The applied correction is based on at least one of the classification(s) into which the character input errors are classified.

Claims

1. A computer-implemented method comprising:

receiving from a keyboard of a computer system character input by a user, the character input comprising character input errors by the user, the character input errors being input of incorrect characters by the user via the keyboard;

automatically identifying the character input errors of the character input and classifying the character input errors into one or more character error classifications, each character error classification of the one or more character error classifications corresponding to a respective visual transformation of input characters;

determining that a frequency of the identified and classified character input errors exceeds a predefined threshold confidence level for triggering application of keyboard corrections for user input; and

based on the determining, applying keyboard corrections for user input, the applying keyboard corrections comprising applying a correction to one or more characters displayed for or input by the user, the applied correction to the one or more characters being based on at least one of the one or more character error classifications into which the character input errors are classified, wherein the keyboard correction comprises, based on selection of a key of the keyboard, the key labeled with a character label representing a first character, providing a second character to the computer system different from the first character, the first and second characters being two characters that the user has interchanged when typing, as determined based on at least some of the identified character input errors.

2. The method of claim 1, wherein the keyboard comprises a dynamic keyboard and wherein the keyboard correction comprises changing character labeling of at least one key of the keyboard.

3. The method of claim 2, wherein a character provided to the computer system based on selection of a key of the at least one key remains the same prior to and after the changing, despite a character label of the key being changed to represent a different character.

4. The method of claim 2, wherein the changing the character labeling switches character labels of two or more keys of the keyboard.

5. The method of claim 4, wherein the changing the character labeling swaps character labels of two keys, the two keys labeled with two characters that the user has interchanged when typing, as determined based on at least some of the identified character input errors, and wherein a respective character supplied to the computer system based on selection of each key of the two keys remains the same prior to and after the swapping the character labels of the two keys.

6. The method of claim 1, wherein a visual transformation of input characters corresponding to a character error classification of the one or more character error classifications is selected from the group consisting of:

vertical transformation of input characters, horizontal transformation of input characters, and a visual similarity of input characters.

7. The method of claim 1, wherein the automatically identifying the character input errors comprises constructing and refining feature vectors for a machine learning model based on feature extraction, and applying the feature vectors to the character input by the user.

8. The method of claim 7, further comprising seeding the machine learning model with documents for analysis, the documents comprising literature specifying common character input errors and common character error classifications from observed individuals.

9. The method of claim 1, wherein the confidence level is selected from a plurality of different predefined confidence levels based on an identification of the user.

10. A computer system comprising:

a memory; and

a processor in communication with the memory, wherein the computer system is configured to perform a method comprising;

receiving from a keyboard of a computer system character input by a user, the character input comprising character input errors by the user, the character input errors being input of incorrect characters by the user via the keyboard;

automatically identifying the character input errors of the character input and classifying the character input errors into one or more character error classifications, each character error classification of the one or more character error classifications corresponding to a respective visual transformation of input characters;

determining that a frequency of the identified and classified character input errors exceeds a predefined threshold confidence level for triggering application of keyboard corrections for user input; and

based on the determining, applying keyboard corrections for user input, the applying keyboard corrections comprising applying a correction to one or more characters displayed for or input by the user, the applied correction to the one or more characters being based on at least one of the one or more character error classifications into which the character input errors are classified, wherein the keyboard correction comprises, based on section of a key of the keyboard, the key labeled with a character label representing a first character, provided a second character to the computer system different from the first character, the first and second characters being two characters that the user has interchanged when typing, as determined based on at least some of the identified character input errors.

11. The computer system of claim 10, wherein the keyboard comprises a dynamic keyboard and wherein the keyboard correction comprises changing character labeling of at least one key of the keyboard.

12. The computer system of claim 11, wherein a character provided to the computer system based on selection of a key of the at least one key remains the same prior to and after the changing, despite a character label of the key being changed to represent a different character.

13. The computer system of claim 10, wherein the automatically identifying the character input errors comprises constructing and refining feature vectors for a machine learning model based on feature extraction, and applying the feature vectors to the character input by the user, and wherein the method further comprises seeding the machine learning model with documents for analysis, the documents comprising literature specifying common character input errors and common character error classifications from observed individuals.

14. A computer program product comprising:

a non-transitory computer readable storage medium readable by at least one processor and storing instructions for execution by the at least one processor for performing a method comprising;

receiving from a keyboard of a computer system character input by a user, the character input comprising character input errors by the user, the character input errors being input of incorrect characters by the user via the keyboard;

automatically identifying the character input errors of the character input and classifying the character input errors into one or more character error classifications, each character error classification of the one or more character error classifications corresponding to a respective visual transformation of input characters;

determining that a frequency of the identified and classified character input errors exceeds a predefined threshold confidence level for triggering application of keyboard corrections for user input; and

based on the determining, applying keyboard corrections for user input, the applying keyboard corrections comprising applying a correction to one or more characters displayed for or input by the user, the applied correction to the one or more characters being based on at least one of the one or more character error classifications into which the character input errors are classified, wherein the keyboard correction comprises, based on selection of a key of the keyboard, the key labeled with a character label representing a first character, providing a second character to the computer system different from the first character, the first and second characters being two characters that the user has interchanged when typing, as determined based on at least some of the identified character input errors.

15. The computer program product of claim 14, wherein the keyboard comprises a dynamic keyboard and wherein the keyboard correction comprises changing character labeling of at least one key of the keyboard.

16. The computer program product of claim 15, wherein a character provided to the computer system based on selection of a key of the at least one key remains the same prior to and after the changing, despite a character label of the key being changed to represent a different character.

17. The computer program product of claim 14, wherein the automatically identifying the character input errors comprises constructing and refining feature vectors for a machine learning model based on feature extraction, and applying the feature vectors to the character input by the user, and wherein the method further comprises seeding the machine learning model with documents for analysis, the documents comprising literature specifying common character input errors and common character error classifications from observed individuals.