Alphanumeric data entry apparatus and method using multicharacter keys of a keypad

• US 20070237310A1
• Filed: 03/30/2006
• Published: 10/11/2007
• Est. Priority Date: 03/30/2006
• Status: Active Grant

• Current Assignee: Harris Global Communications, Inc. (L3Harris Technologies, Inc.)
• Original Assignee: Harris Corporation (L3Harris Technologies, Inc.)
• Inventors: Cahoon, William, Schmiedlin, Joshua
• Granted Patent: US 8,296,484 B2
• Time in Patent Office: Days
• US Class Current: 379/88.10
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques

Abstract

A method and apparatus for predicting the desired alphanumeric character of a depressed multi-character key of a reduced-key keypad based upon the previous selection of characters. Rule trees defining the prediction associated with a depressed key based upon the previous selection of characters are dynamically built upon use to readily function with all language styles and vocabularies. Moreover, the rule trees are weighted to increase the probability of predicting the desired character and to be quickly adaptable to different users having different language styles or vocabularies.

Claims

1. A method for predicting desired alphanumeric characters of a depressed multi-character key of a reduced-key keypad, comprising the steps of:

predicting a current desired character based upon at least one immediately-preceding selected character; and

dynamically weighting subsequent predictions of other desired characters based upon the number of times the previous predictions correctly predicted the desired character.

2. The method as set forth in claim 1, wherein the step of predicting a current desired character based upon at least one immediately-preceding selected character comprises populating a rule tree comprising a plurality of rules that predict a current desired character based upon at least one immediately-preceding selected character.

3. The method as set forth in claim 2, wherein the step of dynamically weighting subsequent predictions of other desired characters based upon the number of times the previous predictions correctly predicted the desired character comprises the step of dynamically weighting each rule that previously correctly predicted the desired character.

4. The method as set forth in claim 3, wherein the step of dynamically weighting subsequent predictions of other desired characters based upon the number of times the previous predictions correctly predicted the desired character comprises the step of not dynamically weighting each rule each time that the rule previously incorrectly predicted the desired character.

5. The method as set forth in claim 3, wherein the step for dynamically weighting the rule tree based upon the number of times the rule tree previously correctly predicted the desired character comprises the step of, for each previously-correct prediction following at least one immediately previous depressed keys, incrementing a counter to weight the prediction associated with a subsequent prediction following the same previously-depressed keys.

6. The method as set forth in claim 3, wherein the step for dynamically weighting the rule tree based upon the number of times the rule tree previously correctly predicted the desired character comprises the step of, for each previously-incorrect prediction following at least one immediately previous depressed keys, not incrementing the counter to weight the prediction associated with a subsequent prediction following the same previously-depressed keys.

7. The method as set forth in claim 2, further comprising the step of updating the rule tree after each character selected.

8. The method as set forth in claim 2, further comprising the step of updating the rule tree after all the desired characters of a desired word is selected.

9. The method as set forth in claim 2, further comprising the step of updating the rule tree after all the desired characters of a plurality of desired words are selected.

10. The method as set forth in claim 1, wherein the step of predicting a current desired character based upon at least one immediately-preceding selected character comprises the step of predicting a current desired character based upon at least two immediately-preceding selected characters.

11. The method as set forth in claim 1, wherein the step of predicting a current desired character based upon at least one immediately-preceding selected character comprises the step of predicting a current desired character based upon at least three immediately-preceding selected characters.

12. The method as set forth in claim 1, further comprising the step of normalizing the dynamic weighting of subsequent predictions of other desired characters based upon the number of times the previous predictions correctly predicted the desired character.

13. The method as set forth in claim 12, further including the step of deleting from the rule tree one or more rules having the lowest weights.

14. The method as set forth in claim 1, wherein the steps of predicting a current desired character based upon at least one immediately-preceding selected character and dynamically weighting subsequent predictions of other desired characters based upon the number of times the previous predictions correctly predicted the desired character are language-independent.

15. The method as set forth in claim 1, wherein the steps of predicting a current desired character based upon at least one immediately-preceding selected character and dynamically weighting subsequent predictions of other desired characters based upon the number of times the previous predictions correctly predicted the desired character employ limited resources.

16. The method as set forth in claim 1, further comprising the step of, whenever the desired character is the first character of a desired word, selecting a key having the first character associated therewith, and then displaying all the key characters associated with the key one at a time allowing selection of the desired character.

17. The method as set forth in claim 16, wherein the step of displaying all the key characters associated with the key one at a time occurs in a predicted sequence.

18. A language-independent method for predicting desired alphanumeric characters of a depressed multi-character key of a reduced-key keypad, comprising predicting a current desired character of a word based solely upon at least one immediately-preceding selected character whereby the prediction may occur irrespective of any language of the word.

19. A method for predicting with limited resources the desired alphanumeric characters of a depressed multi-character key of a reduced-key keypad, comprising predicting, with limited memory, a current desired character of a word based upon at least one immediately-preceding selected character whereby the method may be employed in a resource-limited system.

20. A method for predicting the characters of a desired word by depressing multi-character keys of a reduced-key keypad, comprising the steps of:

(1) accessing a first character rule list corresponding to a first word character of the desired word by;

selecting a first key having a key character that is the first word character of the desired word;

displaying the key characters of the first key one at a time in a predicted sequence and selecting the displayed key character as the first character key when the key character is the first word character of the desired word;

(2) processing a rule tree including a multi-level hierarchical tree of rules for sequentially predicting and displaying of the succeeding word characters of the desired word and either allowing selection thereof when the desired succeeding word character is displayed or allowing selection by multi-tapping of a new desired word character.

21. The method as set forth in claim 20, further comprises the step of dynamically weighting each rule'"'"'s prediction of the succeeding word characters of the desired word based upon the number of times the succeeding word character predicted by the rule was previously selected.

22. The method as set forth in claim 21, wherein the step of dynamically weighting each rule comprises the step of weighting based upon the number of times the succeeding word character was previously selected following at least one immediately-preceding word character selected.

23. The method as set forth in claim 21, wherein the step of dynamically weighting each rule comprises the step of weighting based upon the number of times the succeeding word character was previously selected following at least two immediately-preceding word characters selected.

24. The method as set forth in claim 21, wherein the step of dynamically weighting each rule comprises the step of weighting based upon the number of times the succeeding word character was previously selected following at least three immediately-preceding word characters selected.

25. The method as set forth in claim 21, wherein the step of dynamically weighting each rule comprises the step of not dynamically weighting each rule each time that the rule previously incorrectly predicted the desired character.

26. The method as set forth in claim 21, wherein the step for dynamically weighting the rule tree comprises incrementing a counter to weight the prediction associated with a subsequent prediction following the same previously-depressed keys.

27. The method as set forth in claim 21, wherein the step for dynamically weighting the rule tree comprises the step of, for each previously-incorrect prediction following at least one immediately previous depressed keys, not incrementing the counter to weight the prediction associated with a subsequent prediction following the same previously-depressed keys.

28. The method as set forth in claim 21, further comprising the step of normalizing the dynamic weighting of subsequent predictions of other desired characters based upon the number of times the previous predictions correctly predicted the desired character.

29. The method as set forth in claim 28, further including the step of deleting from the rule tree one or more rules having the lowest weights.

30. The method as set forth in claim 20, further comprising the step of updating the rule tree after each character selected.

31. The method as set forth in claim 20, further comprising the step of updating the rule tree after all the desired characters of a desired word is selected.

32. The method as set forth in claim 20, further comprising the step of updating the rule tree after all the desired characters of a plurality of desired words are selected.

33. The method as set forth in claim 20, wherein the step of displaying the key characters of the first key and selecting the displayed key character as the first character key when the key character is the first word character of the desired word includes the step of weighting the selection based upon the number of times the first character key is selected and then predicting the sequential display of the succeeding first key character based thereon.

34. The method as set forth in claim 33, wherein the step of displaying all the key characters associated with the key one at a time occurs in a predicted sequence.

35. A language-independent method for predicting the characters of a desired word by depressing multi-character keys of a reduced-key keypad, comprising accessing a first character rule list corresponding to a first word character of the desired word by selecting a first key having a key character that is the first word character of the desired word and then displaying the key characters of the first key for selection as the first character key when the key character is the first word character of the desired word and then sequentially predicting and displaying of the succeeding word characters of the desired word and either allowing selection thereof when the desired succeeding word character is displayed or allowing selection by multi-tapping of a new desired word character, whereby the desired characters may be predicted irrespective of the language of the word.

36. A method for predicting with limited resources, the characters of a desired word by depressing multi-character keys of a reduced-key keypad, comprising accessing a first character rule list corresponding to a first word character of the desired word by selecting a first key having a key character that is the first word character of the desired word and then displaying the key characters of the first key for selection as the first character key when the key character is the first word character of the desired word and then with limited memory sequentially predicting and displaying of the succeeding word characters of the desired word and either allowing selection thereof when the desired succeeding word character is displayed or allowing selection by multi-tapping of a new desired word character, whereby the desired characters may be predicted in a system with limited resources.

37. An electronic device, comprising in combination:

a reduced-key keypad having multi-character keys;

a system for predicting desired alphanumeric characters of the multi-character keys comprising software that reads the selected keys and predicts a current desired character based upon at least one immediately-preceding selected character and dynamically weights subsequent predictions of other desired characters based upon the number of times the previous predictions correctly predicted the desired character.

38. The device as set forth in claim 37, wherein the software comprises a rule tree having a plurality of rules that predicts a current desired character based upon at least one immediately-preceding selected character.

39. The device as set forth in claim 38, wherein the software dynamically weights each rule that previously correctly predicted the desired character.

40. The device as set forth in claim 39, wherein the software normalizes the dynamic weighting of subsequent predictions of other desired characters based upon the number of times the previous predictions correctly predicted the desired character.

41. The device as set forth in claim 40, wherein the software deletes from the rule tree one or more rules having the lowest weights.

42. The device as set forth in claim 37, wherein the software predicts characters of the desired word of any language.

43. The device as set forth in claim 37, wherein the software is operable with limited resources.

44. The device as set forth in claim 37, wherein the software, whenever the desired character is the first character of a desired word, selects a key having the first character associated therewith, and then displays all the key characters associated with the key one at a time allowing selection of the desired character.

45. The device as set forth in claim 44, wherein the software displays all the key characters associated with the key one at a time occurs in a predicted sequence.

46. A language-independent embedded system for predicting desired alphanumeric characters of a depressed multi-character key of a reduced-key keypad, comprising software to predict a current desired character of a word based solely upon at least one immediately-preceding selected character whereby the prediction may occur irrespective of any language of the word.

47. A system embedded within an electronic device with limited resources, comprising software to predict, with limited memory, a current desired character of a word entered via a depressed multi-character key of a reduced-key keypad based upon at least one immediately-preceding selected character whereby the method may be employed in a resource-limited system.

48. A system for predicting the characters of a desired word by depressing multi-character keys of a reduced-key keypad, comprising in combination:

(1) a first character rule list corresponding to a first word character of the desired word that is accessed by;

selecting a first key having a key character that is the first word character of the desired word;

displaying the key characters of the first key one at a time in a predicted sequence and selecting the displayed key character as the first character key when the key character is the first word character of the desired word;

(2) a rule tree including a multi-level hierarchical tree of rules for sequentially predicting and displaying of the succeeding word characters of the desired word and either allowing selection thereof when the desired succeeding word character is displayed or allowing selection by multi-tapping of a new desired word character.

49. The system as set forth in claim 48, wherein each rule'"'"'s prediction is dynamically weighted based upon the number of times the succeeding word character predicted by the rule was previously selected.

50. The system as set forth in claim 49, wherein the dynamic weighting of the rule tree comprises a counter that is incremented to weight the prediction associated with a subsequent prediction following the same previously-depressed keys.

51. The system as set forth in claim 50, wherein the dynamic weighting of the rules is normalized.

52. The system as set forth in claim 51, wherein rules having the lowest weights are deleted from the rule tree.

53. The system as set forth in claim 48, wherein the first key character displayed is weighted based upon the number of times the first character key had been previously selected.