Abstract: A method includes displaying a set of one or more suggestions including one or more character strings that are suggested replacements for a first set of one or more entered characters. The method further includes: while displaying the set of suggestions, receiving one or more additional entered characters; and after receiving the additional entered characters, updating the set of suggestions based on an updated set of entered characters that includes the first set of entered characters and the additional entered characters. The updating comprises changing a first suggestion in the set of suggestions from a first character string that is a suggested replacement for the first set of entered characters to a second character string that is a suggested replacement for the updated set of entered characters.

Abstract: A method includes displaying a suggestion region above an on-screen keyboard. The suggestion region includes multiple suggested character strings. The method further includes: detecting a gesture that begins within a predefined key of the on-screen keyboard; and responsive to detecting the gesture: in accordance with a determination that the gesture ends within the predefined key, inserting a first character string into a text field; and in accordance with a determination that the gesture ends outside of the predefined key, inserting a second character string into the text field, wherein the second character string is different from the first character string.

Abstract: Utilization of error vector data representative of errors between the location of actual keystrokes and the location of determined intended keystrokes to compute “bias” data indicative of the magnitude and direction of error vectors for a given location on the virtual keyboard is disclosed. This bias data can then be used to perform a number of operations such as keyboard re-spotting.

Abstract: Systems and processes are disclosed for predicting words in a text entry environment. Candidate words and probabilities associated therewith can be determined by combining a word n-gram language model and a unigram language model. Using the word n-gram language model, based on previously entered words, candidate words can be identified and a probability can be calculated for each candidate word. Using the unigram language model, based on a character entered for a new word, candidate words beginning with the character can be identified along with a probability for each candidate word. In some examples, a geometry score can be included in the unigram probability related to typing geometry on a virtual keyboard. The probabilities of the n-gram language model and unigram model can be combined, and the candidate word or words having the highest probability can be displayed for a user.

Abstract: Text correction processing is disclosed. An initial score is assigned to each of a plurality of candidate sequences of one or more characters, based at least in part on a keyboard geometry-based value associated with the received user input with respect to the candidate key. Further processing is performed with respect to a subset of the candidate sequences having the highest initial score(s) to determine for each candidate sequence in the subset a refined score. A candidate sequence is selected for inclusion in a result set based at least in part on a determination that a refined score of the selected candidate is higher than an initial score of one or more candidate sequences that are not included in the subset and with respect to which the further processing has not been performed.

Abstract: Text correction processing is disclosed. An initial score is assigned to each of a plurality of candidate sequences of one or more characters, based at least in part on a keyboard geometry-based value associated with the received user input with respect to the candidate key. Further processing is performed with respect to a subset of the candidate sequences having the highest initial score(s) to determine for each candidate sequence in the subset a refined score. A candidate sequence is selected for inclusion in a result set based at least in part on a determination that a refined score of the selected candidate is higher than an initial score of one or more candidate sequences that are not included in the subset and with respect to which the further processing has not been performed.

Abstract: Methods and systems of multi-word automatic correction (“autocorrect”) are provided. Autocorrect generally can select a corrected word based on a typed word and a dictionary of correctly-spelled words. Multi-word autocorrect can add to this functionality by revisiting the selection of an initial corrected word if a subsequently-typed word indicates that it would be more appropriate to instead select an additional corrected word. In some cases, an autocorrect system can make a multi-word correction based on a multi-word phrase in a dictionary, such as replacing “new york” with “New York” as described above. In other cases, an autocorrect system can make a multi-word correction to correct a mistakenly-typed delimiter character. In other cases, an autocorrect system can use grammar rules to obtain additional context information with each subsequently-typed word and make multi-word corrections on that basis.

Abstract: Text correction processing is disclosed. An initial score is assigned to each of a plurality of candidate sequences of one or more characters, based at least in part on a keyboard geometry-based value associated with the received user input with respect to the candidate key. Further processing is performed with respect to a subset of the candidate sequences having the highest initial score(s) to determine for each candidate sequence in the subset a refined score. A candidate sequence is selected for inclusion in a result set based at least in part on a determination that a refined score of the selected candidate is higher than an initial score of one or more candidate sequences that are not included in the subset and with respect to which the further processing has not been performed.

Abstract: In one embodiment, the present invention is a method for reward-based learning of improved systems management policies. One embodiment of the inventive method involves obtaining a decision-making entity and a reward mechanism. The decision-making entity manages a plurality of application environments supported by a data processing system, where each application environment operates on data input to the data processing system. The reward mechanism generates numerical measures of value responsive to actions performed in states of the application environments. The decision-making entity and the reward mechanism are applied to the application environments, and results achieved through this application are processed in accordance with reward-based learning to derive a policy. The reward mechanism and the policy are then applied to the application environments, and the results of this application are processed in accordance with reward-based learning to derive a new policy.

Abstract: In one embodiment, the present invention is a method for reward-based learning of improved systems management policies. One embodiment of the inventive method involves supplying a first policy and a reward mechanism. The first policy maps states of at least one component of a data processing system to selected management actions, while the reward mechanism generates numerical measures of value responsive to particular actions (e.g., management actions) performed in particular states of the component(s). The first policy and the reward mechanism are applied to the component(s), and results achieved through this application (e.g., observations of corresponding states, actions and rewards) are processed in accordance with reward-based learning to derive a second policy having improved performance relative to the first policy in at least one state of the component(s).