Method and apparatus for psychomotor and psycholinguistic prediction on touch based device

• US 8,583,421 B2
• Filed: 03/04/2010
• Issued: 11/12/2013
• Est. Priority Date: 03/06/2009
• Status: Expired due to Fees

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Motorola Mobility LLC (Lenovo Group Ltd.)
• Inventors: Slocum, Jeremy, Cheong, Yuen-Keen, Phan Huy, Minh Quan, Phipps, Chad
• Primary Examiner(s): SPOONER, LAMONT M
• Application Number:
  US12/717,555
• Publication Number:
  US 20100228539A1
• Time in Patent Office: 1,349 Days
• Field of Search: 704/1, 704/9, 704/10, 382/209, 382/214, 382/309, 345/169, 345/170, 345/173, 345/179, 707706-708
• US Class Current: 704/9
• 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 40/232 Orthographic correction, e.g. spell checking or vowelisation
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs

Abstract

A method (300 or 400) of key press prediction on a touch-based device or an apparatus (501) can entail registering (302, 402) a position of a touch press attempt on an icon among a plurality of icons presented on the touch-base device, calculating (304, 404) a distance from a center of the icon, calculating (306, 406) a psychomotor probability for each icon in the plurality of icons, calculating (312, 408) a linguistic probability for each icon in the plurality of icons, calculating (314, 410) a predictive index for each icon in the plurality of icons based on the psychomotor probability and the linguistic probability, and registering (310,416) as a user selected icon, an icon with one among a highest predictive index (316, 412) or a probability higher than a predetermined absolute value (308).

Claims

1. A method of key press prediction on a touch-based device, comprising:

registering a position of a touch press attempt on an icon among a plurality of icons presented on the touch-base device;

calculating a distance from a reference point of the icon to the position;

calculating a psychomotor probability for each icon in the plurality of icons;

calculating a linguistic probability for each icon in the plurality of icons;

calculating a predictive index for each icon in the plurality of icons based on the psychomotor probability and the linguistic probability; and

registering as a user selected icon, an icon with one among a highest predictive index and a probability higher than a predetermined absolute value,

wherein for the psychomotor probability calculation, a distribution of finger presses is defined by one among a distance of an x-y position from a center of an icon or key or by a distance of an x-y position from a centroid of pertinent finger presses related to the icon or key, and

wherein the distribution of finger presses is a parameter that is standardized using z-scores which allow the determination of a likelihood of an intended key press for specific keys.

2. The method of claim 1, wherein the method calculates the psychomotor probability and the linguistic probability if there is no single icon with a larger probability than the predetermined absolute value.

3. The method of claim 1, wherein the method performs a disambiguation function if there is no icon with a highest predictive index before registering the user selected icon.

4. The method of claim 1, wherein the method performs a disambiguation function if there is no icon with a highest predictive index and no single icon with the larger probability than the predetermined absolute value before registering the user selected icon.

5. The method of claim 1, wherein the method performs post-hoc auto-correction after detection of completion of a word.

6. The method of claim 1, wherein the method determines for each touch press attempt on an icon if the touch press attempt is ambiguous and creates a list of candidate characters for a given position of the character in the word.

7. The method of claim 6, wherein the method determines if a word is completed and then checks a dictionary for best word matches.

8. The method of claim 1, wherein the variability of finger presses on a QWERTY key arrangement is determined a priori and used for the psychomotor probability calculation.

9. The method of claim 1, wherein the distribution of finger key presses is captured a priori and used for the psychomotor probability calculation.

10. The method of claim 1, wherein for the psychomotor probability calculation, a distribution of finger presses is defined by a distance of an x-y position from a reference point of an icon or a key or from a center of the icon or key, or from a centroid of touch presses.

11. The method of claim 1, wherein for the psychomotor probability calculation, a distribution of finger presses is defined by a distance of an x-y position from a centroid of pertinent finger presses related to the icon or key.

12. The method of claim 1, wherein the method constructs letter frequency tables to determine a likelihood of an occurrence of a letter based on the characters in front of a given letter.

13. The method of claim 1, wherein the predictive index is computed by combining the psychomotor probability with the linguistic probability.

14. The method of claim 1, wherein the method calculates the distance from the reference point of the icon by calculating from a center of the icon as the reference point.

15. The method of claim 1, wherein the method calculates the distance from the reference point of the icon by calculating from a centroid of touch points collected a priori.

16. An electronic device, comprising:

a touch-based presentation device; and

a controller coupled to the touch-based presentation device operable to;

register a position of a touch press attempt on an icon among a plurality of icons presented on the touch-base device;

calculate a distance from a reference point of the icon;

calculate a psychomotor probability for each icon in the plurality of icons;

calculate a linguistic probability for each icon in the plurality of icons;

calculate a predictive index for each icon in the plurality of icons based on the psychomotor probability and the linguistic probability; and

register as a user selected icon, an icon with one among a highest predictive index and a probability higher than a predetermined absolute value,

wherein for the psychomotor probability calculation, a distribution of finger presses is defined by one among a distance of an x-y position from a center of an icon or key or by a distance of an x-y position from a centroid of pertinent finger presses related to the icon or key, and

wherein the distribution of finger presses is a parameter that is standardized using z-scores which allow the determination of a likelihood of an intended key press for specific keys.

17. The electronic device of claim 16, wherein the electronic device is one among a cellular phone, a personal digital assistant, a laptop computer a smart phone, an MP3 player, and a remote controller.

18. A portable communication device, comprising:

a touch-based screen; and

a controller coupled to the touch-based screen operable to;

register a position of a touch press attempt on an icon among a plurality of icons presented on the touch-base device;

calculate a distance from a center of the icon;

calculate a psychomotor probability for each icon in the plurality of icons;

calculate a linguistic probability for each icon in the plurality of icons;

calculate a predictive index for each icon in the plurality of icons based on the psychomotor probability and the linguistic probability; and

register as a user selected icon, an icon with a highest predictive index,

wherein an icon comprises a key,

wherein for the psychomotor probability calculation, a distribution of finger presses is defined by a distance of an x-y position from a center of an key or by a distance of an x-y position from a centroid of pertinent finger presses related to the key, and

wherein the distribution of finger presses is a parameter that is standardized using z-scores which allow the determination of a likelihood of an intended key press for specific keys.