Data processing: artificial intelligence – Knowledge processing system – Knowledge representation and reasoning technique
Reexamination Certificate
2000-01-18
2003-11-25
Gordon, Paul P. (Department: 2121)
Data processing: artificial intelligence
Knowledge processing system
Knowledge representation and reasoning technique
C706S011000, C706S046000
Reexamination Certificate
active
06654733
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to devices having keyboards, such as personal-digital-assistant (PDA) devices having soft keyboards as well as desktop and laptop computers having real, physical keyboards, and more particularly to determining in conjunction with such keyboards a most likely intended-to-be-typed keystroke or sequence of keystrokes.
BACKGROUND OF THE INVENTION
Small computerized devices that fit into the palm of one's hand, generally referred to as personal-digital-assistant (PDA) devices, have become increasingly popular. They allow users to look up and record information where computer access is inconvenient, such as when in one's car, while shopping at the mall, during a business meeting, etc. Unlike a typical desktop or laptop computer, which has an actual physical, or “real,” keyboard for text and data entry, a PDA device usually only has a small number of buttons, and a small touch-sensitive area on which a pen-like implement called a stylus can be used. For text entry on such a device, the user is able to write letters or symbols corresponding to letters on the touch-sensitive area with the stylus, which the device attempts to recognize.
However, for a variety of reasons, some users prefer not to input text into these devices in this manner. Therefore, most PDA devices also permit a second form of text and data entry. An image of a regular keyboard is displayed on the small display of a PDA device, such that a user is able to select which key to “type” next by tapping the stylus on the part of the image of the keyboard corresponding to the desired key. Such an image of a keyboard being displayed on the display of a device such as a PDA device, to provide for text and data entry on the part of the user by tapping the parts of the image corresponding to keys of the keyboard, is known as a “soft” keyboard.
A disadvantage with such soft keyboards is, however, that the displayed keys of the keyboard on the display of the PDA device are very small, since the display of the device as well as the device itself are small. This means that users are apt to tap incorrect keys, especially when “typing” quickly, and also that users are apt to type more slowly using soft keyboards, because of the dexterity required in order to achieve a high accuracy rate when typing. This limits the usefulness of these PDA devices, since they cannot very conveniently be used for quick and/or large amounts of text and data entry. For this and other reasons, there is a need for the present invention.
SUMMARY OF THE INVENTION
The invention relates to fuzzy keyboards, to determine a most likely intended-to-be-typed keystroke or keystrokes. In one embodiment, a computer-implemented method adds each of one or more keys to each of a current list of key sequence “hypotheses,” to create a new list of key sequence hypotheses, upon a key press event. For a soft keyboard, for example, a key press event may be the actuation of a location within the soft keyboard rendered on a touch-sensitive display. For a hard keyboard, for example, a key press event may be the actuation of an actual key on the keyboard. The method determines a likelihood probability for each hypothesis in the new list, and then removes any hypothesis that fails to satisfy any of one or more thresholds. In one embodiment, the most likely key sequence of the new list is displayed.
In another embodiment specifically relating to soft keyboards, a computerized system includes a touch-sensitive display, a soft keyboard rendered thereon, and a computer program. The computer program detects actuation of a location on the soft keyboard, and based at least thereon determines a most likely intended-to-be-typed key on the soft keyboard. In another embodiment specifically relating to hard keyboards, a computerized system includes a keyboard and a computer program. The computer program detects actuation of a key on the keyboard, and determines a most likely intended-to-be-typed key of the keyboard, based on at least the key actuated and either a key sequence of keys already determined to have been typed, a list of key sequence hypotheses, or both.
Embodiments of the invention provide for advantages not found within the prior art. With respect to hard keyboards, the user may be typing fast, for example, and press the “d” key instead of the “e” key. However, the most likely key sequence hypothesis may be such that the typing of the “e” key is much more probable than the typing of the “d” key. As an example, the user may have initially typed “Th” prior to the “d” key. Because the “d” key is close to the “e” key, and few words begin “Thd” while many words begin “The,” and also because the user was typing fast, the key sequence “The” may be displayed instead of “Thd.”
With respect to soft keyboards, keys that are determined as more likely to be entered by the user have a greater probability that they are intended to be typed when the user actuates a location close to those keys. For example, if the user actuates the soft keyboard in the vicinity between the “d” and the “f” keys, and the most likely key sequence hypothesis is such that the “d” key is more likely than the “f” key to have been typed by the user, then the “d” key may be displayed. In some embodiments of the invention, past keystrokes may also be changed, up to a predetermined number of keystrokes before the current keystroke.
Furthermore, it is noted that in one embodiment of the invention, the fuzzy keyboard as described herein can be implemented in conjunction with the predictive keyboard described in the copending, coassigned, and cofiled application entitled “Predictive Keyboard” having U.S. patent application Ser. No. 09/484,079. However, the invention itself is not so limited.
The invention includes computer-implemented methods, machine-readable media, computerized devices, computerized systems, and computers of varying scopes. Other aspects, embodiments and advantages of the invention, beyond those described here, will become apparent by reading the detailed description and with reference to the drawings.
REFERENCES:
patent: 5718590 (1998-02-01), Choate
patent: 5963671 (1999-10-01), Comerford et al.
Karen Kukich, Techniques for Automatically Correcting Words in Text; Dec. 1992; ACM; vol. 24, No. 4; 377-439.*
Hermann Ney, Ute Essen, Reinhard Kneser, On structuring probabilistic dependences in stochastic language modelling, Computer Speech & Language 8, 1-38, 1994.
I. Scott MacKenzie, Shawn Zhang, The design and evaluation of a high-performance soft keyboard, CHI '99, Pittsburgh, PA, May 15-20, 1999.
P. Vale, U.S. patent application entitled Dynamic Keyboard, Ser. No. 09/146,769, Sep. 3, 1998.
Dario D. Salvucci, Inferring Intent in Eye-Based Interfaces: Tracing Eye Movements with Process Models, CHI '99, Pittsburgh, PA, May 15-20, 1999.
S. Chen and J. Goodman, An Empirical Study of Smoothing Techniques for Language Modeling, Computer Science Group, Harvard University, TR-10-98, Aug. 1998.
Andrew W. Golding, Yves Schabes, Combining Trigram-based and Feature Based Methods, Mitsubishi Technical Report, TR-96-03a, May 1996.
John W. Miller, Fil Alleva, Evaluation of a Language Model using a Clustered Model Backoff, Proccedings of ICSLIP 96, pp. 390-393, 1996.
Goodman Joshua
Huang Xuedong
Venolia Daniel
Gordon Paul P.
Hirl Joseph P.
Koehler S.
Microsoft Corporation
Westman Champlin & Kelly P.A.
LandOfFree
Fuzzy keyboard does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Fuzzy keyboard, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Fuzzy keyboard will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3182486