Abstract: A system that incorporates the subject disclosure may include, for example, detecting in a corrected text message an auto-correction by the first communication device of a target word of a group of words, detecting an input command requesting a transmission of the corrected text message to a second communication device, responsive to the detecting of the input command, presenting a correction alert indicating the target word that has been auto-corrected, and in response to the presenting of the correction alert, presenting an option for modifying the target word, where the transmission of the corrected text message to the second communication device is limited to after the option for modifying is presented at the first communication device. Other embodiments are disclosed.

Abstract: A method and apparatus for providing case restoration in a communication network are disclosed. For example, the method obtains one or more content sources from one or more information feeds, and extracts textual information from the one or more content sources obtained from the one or more information feeds. The method then creates or updates a capitalization model based on the textual information.

Abstract: In accordance with one embodiment, a method of generating language models for speech recognition includes identifying a plurality of utterances in training data corresponding to speech, generating a frequency count of each utterance in the plurality of utterances, generating a high-frequency plurality of utterances from the plurality of utterances having a frequency that exceeds a predetermined frequency threshold, generating a low-frequency plurality of utterances from the plurality of utterances having a frequency that is below the predetermined frequency threshold, generating a grammar-based language model using the high-frequency plurality of utterances as training data, and generating a statistical language model using the low-frequency plurality of utterances as training data.

Abstract: A transcript interface for displaying a plurality of words of a transcript in a text editor can be provided and configured to receive a command to edit the transcript. Limited edits to data corresponding to the transcript can be made in response to commands received via the user interface module. For example, edits may be limited to selection of a single word in the text editor for editing via a given command. The edit may affect an adjacent word in some instances, such as when two adjacent words are merged. In some embodiments, data corresponding to the selected word of the transcript is changed to reflect the edit without changing data defining the relative timing of those words of the transcript that are not adjacent to the selected word.

Abstract: A method includes receiving an indication in an information handling device of an autocorrected word in a string of text, and providing an indication proximate the autocorrected word to a display.

Abstract: In one embodiment, a method includes accessing a string of symbols by a computing device. The string is divided into one or more string components each including at least one of the symbols, and each string component is associated with at least one string-position identifier. The string components and their respective associated string-position identifiers are stored for the string of symbols.

Abstract: A support apparatus, computer-implemented method, and article of manufacture for supporting a processing of a character string in a document. The support apparatus includes: a document storage unit; a designation reception unit; a document acquisition unit; a historical-information storage unit; a reference-information generator; and a presentation unit. The method includes: receiving designation from a user; acquiring a document from the document storage device based on the designation by the user; reading pieces of historical information related to the acquired document from the historical-information storage device, and thereby generating reference information for helping the user to select a processing of the character string in the document; and presenting, to the user, a screen for receiving the designation of the processing of the character string in the acquired document, together with the reference information.

Abstract: A method includes receiving a current set of characters in response to a sequence of user commands. The current set of characters includes a current sequence of words. A respective word in the current sequence of words includes one or more characters in the current set of characters. The user commands include character entry commands. Additional user commands that specify characters to delete from the current set of characters and characters to add to the current set of characters are received. Replay recommendations for additional words to be added to the current set of characters in accordance with one or more deleted words that correspond to the deleted characters are provided.

Abstract: A computer-assisted language correction system including spelling correction functionality, misused word correction functionality, grammar correction functionality and vocabulary enhancement functionality utilizing contextual feature-sequence functionality employing an internet corpus.

Abstract: User input is received, specifying a continuous traced path across a keyboard presented on a touch sensitive display. An input sequence is resolved, including traced keys and auxiliary keys proximate to the traced keys by prescribed criteria. For each of one or more candidate entries of a prescribed vocabulary, a set-edit-distance metric is computed between said input sequence and the candidate entry. Various rules specify when penalties are imposed, or not, in computing the set-edit-distance metric. Candidate entries are ranked and displayed according to the computed metric.

Abstract: Methods and systems for large-scale statistical misspelling correction are provided. A method implemented in a computer infrastructure includes reviewing the input text to detect spelling errors in one or more words and calculating a variable cost distance of letters associated with the one or more words. Furthermore, the method can detect space-deletion errors and space-insertion errors. The method further includes determining a best candidate solution for correcting the spelling errors based on the variable cost distance. The method includes correcting the spelling errors using the best candidate solution.

Abstract: An improved spell check function and handheld electronic device provide a spell checking feature that is available in a plurality of data entry fields of a plurality of applications. A memory on the handheld electronic device has a plurality of linguistic data sources which are searched for linguistic data during a spell check operation. Generic linguistic data is stored in a static data source, and other linguistic data is stored in a dynamic data source. An identification of a misspelled text entry results in an editing session being automatically opened on the misspelled text entry. Proposed spell check interpretations of a text entry are output in a list, and an actuation of a <MENU> key further results in the outputting of a menu of selectable spell check options. At least some of the selectable spell check options are selectable either with a track ball or with a key of a keypad.

Abstract: An electronic device is configured for retroactive word correction. The device includes memory, a processor, a network interface and display. The device also includes an operating system managing access to the memory, processor, network interface and display. The device yet further includes a message transmitting application and a spell checker operable to perform spell checking on text in messages composed within the message transmitting application. Finally, the device includes a retroactive word correction module coupled to the spell checker. The module includes program code enabled to identify both an incorrectly spelled word in a previously transmitted message in the message transmitting application and also a correctly spelled form of the incorrectly spelled word, and to respond to the identifying (A) by locating the incorrectly spelled word in the previously transmitted message and (B) by replacing the located incorrectly spelled word with the correctly spelled form of the incorrectly spelled word.

Abstract: A spelling correction system and method are provided. The system includes at least an input unit, a correct keyword candidate determining unit, and a misspelling correction unit. In the method, the input unit detects an input keyword entered by a user. If the input keyword is a misspelled keyword, the correct keyword candidate determining unit selects one or more correct keyword candidates for the input keyword and then returns the selected correct keyword candidates. The misspelling correction unit obtains a misspelling appearance probability of a pair of the input keyword and each correct keyword candidate, and also obtains a word appearance probability of each correct keyword candidate. Then the misspelling correction unit selects a specific correct keyword from among the correct keyword candidates by using the misspelling appearance probability and the word appearance probability.

Abstract: A document having multiple data entry areas can be presented, and a list of suggestions for a designated data entry area can be generated. Generating the list can include analyzing a first set of data previously entered into one or more data entry areas of the document other than the designated data entry area and analyzing a second set of data including data outside the first set of data. The list of suggestions can be presented without having received manual entry of data in the designated data entry area, and user input selecting one of the suggestions can be received. Data represented by the selected suggestion can be entered in the data entry area in response to receiving the user input selecting the selected suggestion.

Abstract: The present invention provides a word input method including the steps of: selectively inputting one or more letters containing the first letter of the word to be inputted; extracting, as a first priority, a word having less letters than a predetermined number, and which coincides with the word to be inputted, in a sequence from the first letter, from words containing one or more letters, from a word storage unit; extracting, as a second priority, a word having less letters than the predetermined number, from words containing one or more letters, from the word storage unit; extracting, as a third priority, a word having more letters than the predetermined number, and which coincides with the word to be inputted, in a sequence from the first letter, from words containing one or more letters, from the word storage unit; extracting, as a fourth priority, a word having more letters than the predetermined number, from words containing one or more letters, from the word storage unit; dividing the words of the first

Abstract: Methods, computer program products and systems for updating a dictionary on a computer are provided including identifying a set of words included with an application program being installed on the computer and identifying one or more installed dictionaries within the computer. The dictionaries that are identified, typically by searching the computer for installed dictionaries, may then be updated by adding one or more words of the set to at least one of the installed dictionaries. Optionally, one or more installed dictionaries may be identified by the user responding to a query asking the user to identify dictionaries or applications that contain dictionaries installed on the computer. An application programming interface may add the words to the dictionaries by associating the installed dictionaries with one or more installed application programs and identifying an API that is specific for the application and using the API to update the dictionaries.

Abstract: The disclosure generally describes computer-implemented methods, software, and systems for detecting and correcting errors in documents. A computer-implemented method includes: (i) identifying a string policy enforcement rules set associated with at least one set of documents, each rule in the rules set including one or more specific rule strings and instructions relating to string-based policies; (ii) identifying at least one document containing at least one string, the at least one identified document associated with at least one rule included in the identified string policy enforcement rules set; and (iii) validating the at least one identified document, wherein validating the at least one identified document includes validating the at least one string in the at least one identified document using the at least one rule from the string policy enforcement rules set associated with the at least one identified document.

Abstract: A fast client-side spell checker is provided that builds efficient structures out of dictionary and a common misspelling lists and uses the structures to prune the number of searches required to identify misspelled words and provide suggestions for correcting the misspelled words. The spell checker is a browser-based application, which is provided by a server to a client device. The server also sends the dictionary and a list of common misspellings to the client device in the form of efficient data structures. The spell checker utilizes a set of rules to identify the words that are not in the dictionary but are intended to be correct as typed. The spell checker is used by different browser-based applications that utilize the same spell checker regardless of the browser platform used to access the applications. In this way, the spell checker provides a uniform spell checking user experience across different browser platforms.

Abstract: Systems, methods, and programs embodied in a computer readable medium are provided for index extraction. Stored in a database are ground truth documents that are organized according to a plurality of classifications, each classification having a group of predefined indices. A document to be indexed is classified by drawing an association between the document and one of the classifications. An attempt is made to extract from the document at least a subset of the group of predefined indices associated with the one of the classifications. Upon a failure to extract the subset of the group of predefined indices, attempts are made to find and correct at least one text recognition error in the document based upon a salient dictionary associated with the one of the classifications.

Abstract: A handheld electronic device includes a reduced QWERTY keyboard and is enabled with disambiguation software that is operable to disambiguate text input. In addition to identifying and outputting representations of language objects that are stored in the memory and that correspond with a text input, the device is able to generate artificial variants in certain circumstances. Each artificial variant is compared with N-gram data on the handheld electronic device and is suppressed from being output if the artificial variant is determined to have a low probability of being the input intended by a user.

Abstract: The present disclosure is directed to a method for spell checking. The method includes the step of generating a geographic based list of words. The method also includes the step of appending the geographic based list of words to a spell check list. A further step of the method includes detecting a change in position. The method also includes the step of generating a new geographic based list of words when a change in position occurs.

Abstract: A non-standard and standard clause detection system imports raw input data or contractual documents, and extracts non-standard and standard clauses that are semantically linked. One embodiment of a disclosed configuration is a system and a method for identifying non-standard and standard clauses in contractual documents. The system and the method comprise of generating a primary policy and a secondary policy, obtaining a first feature data set by applying the primary policy to a semantic language evaluator, and obtaining a second feature data set by applying the secondary policy to the semantic language evaluator. The first feature data set obtained is the aggregation of the standard clauses used in the document. Furthermore, the second feature data set encompasses the first feature data set, thus the difference between the first feature data set and the second feature data set is the aggregation of the non-standard clauses.

Abstract: In one embodiment, there is provided a keyword presenting device including: an extraction unit configured to extract a plurality of keywords from a browsing document; a determination unit configured to arrange keywords with spellings similar to each other among the plurality of keywords to obtain a plurality of groups of similar keyword; an integration unit configured to classify the keywords into main keywords that are titles and the other sub-keywords for each group of similar keywords, and to integrate the sub-keywords into the main keywords; and a presentation unit configured to present the main keywords to a user.

Abstract: Methods, handheld electronic devices, and computer-readable media for disambiguating an input into a handheld electronic device, are disclosed. The method includes, for example, detecting an ambiguous input including one or more selections of one or more input characters, generating one or more prefix objects corresponding with the ambiguous input, generating an output set comprising at least some of the one or more prefix objects, each of the at least some of the one or more prefix objects associated with an identified corresponding word object, determining the quantity of prefix objects in the output set is fewer than a predetermined quantity, and, based on the determination, adding as an orphan prefix object to the output set at a position corresponding with a relatively low frequency an additional prefix object of the one or more of prefix objects for which a corresponding word object was not identified, and outputting the output set.

Abstract: Various embodiments related to the correction of typographical errors on touch displays are disclosed. One disclosed embodiment includes a portable computing device comprising a touch-sensitive display device, a logic subsystem, and memory comprising instructions executable by the logic subsystem to perform a method of correcting typographical errors displayed on the touch-sensitive display device. The instructions are executable to identify a typographical error within text displayed on the display, and upon identifying the typographical error, highlight on the display the typographical error.

Abstract: A character correcting method of a display apparatus includes displaying a sentence including at least one word in a character input window; displaying, if one of the at least one word is selected, a first user interface (UI) in which the at least one selected word is displayed by a character unit; and, displaying, if one of a plurality of characters displayed in the first UI is selected, a second UI around the selected character, the second UI is configured to edit the selected character.

Abstract: A device and method for its operation comprising a display and input device, a processor to receive character input maintain an input context of words and a dictionary comprising a plurality of word occurrence indicators for selection of candidate words that may follow context words. A word occurrence indicator is associated with a kth order combination of a context word and a candidate word that indicates the frequency by which a respective candidate word follows the corresponding context word by k words in a text corpus. The processor is able to select a list of candidate words where the context word of at least one kth order combination matches the kth last word of input context. A score is determined depending on an associated word occurrence indicator for at least one relevant kth order combination.

Abstract: In one aspect, a spellchecker is provided that differentiates between code and natural language parts of a source code file. In operation, the spellchecker identifies a string in the source code file and attempts to classify the string as inducing programming language instructions or natural language words. If the sting is classified as including natural language words, the spellchecker outputs an indication of a spelling mistake in the string. Otherwise, if the string is not classified as including natural language words, the spellchecker does not output indications of spelling mistakes found in the string.

Abstract: Word predictions in a message are selected or prioritized based on the recipient of the message and a previous location of use by a user. An input history is created based on messages sent to the recipient from the user at a particular location (e.g., global positioning system coordinates). As the user composes subsequent messages, a current location of the user is determined. Word predictions are performed based on a comparison of the current location to the previous locations, and based on the recipient(s). In further embodiments, location-aware spell-check functionality is provided for the messages.

Abstract: Methods, systems, and apparatus for performing simhash based spell correction are provided. A character string is simhashed to generate a simhashed character string. A plurality of substrings is extracted from the character string by applying a sliding window of at least two characters to the character string. The plurality of substrings are hashed to produce a plurality of corresponding hash values. Each hash value is processed to generate a simhashed character string. The simhashed character string is then compared with character strings within a simhashed dictionary dataset to determine at least one candidate to replace the character string. Processing each hash value includes extracting a set of lowest bits from each hash value, and mapping each set of lowest bits to the bitmask.

Abstract: Text input is corrected on a touch-sensitive display by presenting a list of candidate words in the interface which can be selected by touch input. The candidate list can include candidate words having two or more character types (e.g., Roman, kana, kanji). In one aspect, the candidate list can be scrolled using a finger gesture. When a user's finger traverses a candidate word and the touch is released, the candidate word is inserted into a document being edited. In another aspect, characters can be erased by touching a key (e.g., a backspace or delete key) and making a sliding, swiping, or other finger gesture. A number of characters proportional to a distance (e.g., a linear distance) of the finger gesture across the display are erased. If there are characters in a text input area, those characters are erased first, followed by characters in the document being edited.

Abstract: A system and method are provided to process listing data in a network-based commerce system, for example, a network-based auction facility. The method may comprise providing a user with a graphic user interface to enter listing data in the network-based commerce system, the listing data including at least one entered word. A database including a plurality of reference words that relate to listings provided by the network-based commerce system may then be investigated. If at least one entered word is misspelled, the method may include identifying at least one suggested word from the plurality of reference words and communicating the at least one suggested word to the user. In one embodiment, the reference words include words frequently misspelled and the at least one suggested word is the correct spelling of the word. The at least one suggested word may be associated with at least one category associated with a listing.

Abstract: A system and method for providing an adaptive Interaction Logging functionality to help agents reduce the time spent documenting contact center interactions. In a preferred embodiment the system uses a pipeline comprising audio capture of a telephone conversation, automatic speech transcription, text normalization, transcript generation and candidate call log generation based on Real-time and Global Models. The contact center agent edits the candidate call log to create the final call log. The models are updated based on analysis of user feedback in the form of the editing of the candidate call log done by the contact center agents or supervisors. The pipeline yields a candidate call log which the agents can edit in less time than it would take them to generate a call log manually.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing autocomplete suggestions. A computing device can receive an n-gram of characters. Cached autocomplete suggestions can be identified in a local cache of the computing device. Additional autocomplete suggestions can be requested at a first time. The computing device can prevent presentation of the cached autocomplete suggestions until a presentation event occurs. The presentation event can include (i) receiving the requested additional autocomplete suggestions by the computing device or (ii) the end of a predetermined period after the first time. The computing device can determine that that the presentation event has occurred, and in response, can display one or more autocomplete suggestions.

Abstract: A method for word correction on a multi-touch device, including: detecting a selection by a user of a word displayed in a user interface of the multi-touch device; breaking the word into logical segments; presenting the logical segments in the user interface; detecting a user-selected segment of one of the logical segments; displaying at least one alternative segment for the user-selected segment in the user interface; and altering the selected segment in response to receiving a user-selected replacement from the at least one alternative segment.

Abstract: The disclosed embodiments relate to a system that corrects user-interface shift errors. During operation of the system, a user provides an input through a user interface. In response, the system compares the input to a pre-determined set of phrases associated with user-interface errors. In particular, the user-interface errors may include shift errors in a position of a finger or a hand of the user relative to the user interface. Based on the comparisons, the system identifies a correct input, thereby facilitating correction of the shift error that occurred when the user provided the input through the user interface. Next, the system provides feedback to the user. For example, the feedback may correct the received input or may provide a suggestion to the user that includes the suggested correct input.

Abstract: A computer-implemented method for determining whether a target text-string is correctly spelled is provided. The target text-string is compared to a corpus to determine a set of contexts which each include an occurrence of the target text-string. Using heuristics, each context of the set is characterized based on occurrences in the corpus of the target text-string and a reference text-string. Contexts are characterized as including a correct spelling of the target text-string, an incorrect spelling of the reference text-string, or including an indeterminate usage of the target text-string. A likelihood that the target text-string is a misspelling of the reference text-string is computed as a function of the quantity of contexts including a correct spelling of the target text-string and the quantity of contexts including an incorrect spelling of a reference text-string. In one application, the target text-string is received in a search query, the search executed following a spell-check.

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: For relieving typing burdens caused by incorrect spellings, typing errors, unknown spellings, and characters with diacritical marks, and for enhancing a typing efficiency of a typist with some simplified techniques, several candidate generating methods are provided for assisting the typist to pick a candidate word from a list of generated candidate words, or for selecting a candidate word from said list of generated candidate words in an automatic-selection manner. A proper-designed user interface may also be utilized for implementing the candidate generating methods.

Abstract: A method and apparatus for representing and controlling documents including rich text for Web based applications and browsers is provided so that editing of rich text can be facilitated within the browsers. The rich text is represented in a memory structure so that various formats may be flexible maintained. Text, images, tables, links and the like are represented in the memory structure, which may be maintained in databases for eventual editing. A controller class and subsidiary classes represent the rich text and provide methods to convert html to the memory structure and back, representing the rich text in a relational database, retrieving the rich text from a relational database, and presenting the rich text for editing. A spell checking facility for the rich text is included.

Abstract: Provided herein is a character processing device that converts an input character formed of an input alphanumeric or symbol to an extended Latin character similar to the input character including: a display unit displaying as an editing character the input character with a cursor attached thereto; a conversion target distinction unit discerning whether or not the editing character is convertible to the extended Latin character; and a notification unit indicating that the editing character is convertible to the extended Latin character when the editing character is regarded as convertible.

Abstract: A method for entering keys in a small key pad is provided. The method comprising the steps of: providing at least a part of keyboard having a plurality of keys; and predetermining a first probability of a user striking a key among the plurality of keys. The method further uses a dictionary of selected words associated with the key pad and/or a user.

Abstract: Methods and systems for large-scale statistical misspelling correction are provided. A method implemented in a computer infrastructure includes reviewing the input text to detect spelling errors in one or more words and calculating a variable cost distance of letters associated with the one or more words. Furthermore, the method can detect space-deletion errors and space-insertion errors. The method further includes determining a best candidate solution for correcting the spelling errors based on the variable cost distance. The method includes correcting the spelling errors using the best candidate solution.

Abstract: A method and apparatus for representing and controlling documents including rich text for Web based applications and browsers is provided so that editing of rich text can be facilitated within the browsers. The rich text is represented in a memory structure so that various formats may be flexible maintained. Text, images, tables, links and the like are represented in the memory structure, which may be maintained in databases for eventual editing. A controller class and subsidiary classes represent the rich text and provide methods to convert html to the memory structure and back, representing the rich text in a relational database, retrieving the rich text from a relational database, and presenting the rich text for editing. A spell checking facility for the rich text is included.

Abstract: The disclosure is directed to a method and system for accessing textual widgets. A method in accordance with an embodiment includes: entering a string expression into a document; invoking a spell-checker to check a spelling of the string expression; marking the string expression as misspelled; identifying a textual widget based on the misspelling of the string expression; evaluating the misspelled string expression using the identified textual widget, the identified textual widget returning at least one result of the evaluation; displaying the at least one result of the evaluation; selecting a result of the evaluation; and replacing the string expression in the document with the selected result of the evaluation.

Abstract: A method for providing suggestions includes capturing audio that includes speech and receiving textual content from a speech recognition engine. The speech recognition engine performs speech recognition on the audio signal to obtain the textual content, which includes one or more passages. The method also includes receiving a selection of a portion of a first word in a passage in the textual content, wherein the passage includes multiple words, and retrieving a set of suggestions that can potentially replace the first word. At least one suggestion from the set of suggestions provides a multi-word suggestion for potentially replacing the first word. The method further includes displaying, on a display device, the set of suggestions, and highlighting a portion of the textual content, as displayed on the display device, for potentially changing to one of the suggestions from the set of suggestions.

Abstract: A portable device may attempt to predict words input by a user. In one implementation, the portable device may generate candidate words that correspond to predictions of the word being entered by the user. The candidate words being generated based on the characters input by the user of the portable device, characters in proximity to a last entered one of the input characters, and distance factors calculated based on distances between a touch-point on the a virtual keyboard and characters in proximity to the touch-point.

Abstract: Described are computer-based methods and apparatuses, including computer program products, for automation of auditing claims. Data indicative of an insurance company name is received, the data comprising one or more words. The data is processed through one or more processing steps to generate processed data comprising one or more processed words. One or more candidate word strings are selected based on the one or more processed words. Matching information is associated with each of the one or more candidate word strings. Analysis information is generated for each of the one or more candidate word strings based on the associated matching information. An insurance company identifier is associated with received data based on the analysis information and one or more matching rules.

Abstract: Systems and methods for character error correction are provided, useful for a user of mobile appliances to produce written text with reduced errors. The system includes an interface, a word prediction engine, a statistical engine, an editing distance calculator, and a selector. A string of characters, known as the inputted word, may be entered into the mobile device via the interface. The word prediction engine may then generate word candidates similar to the inputted word using fuzzy logic and user preferences generated from past user behavior. The statistical engine may then generate variable is error costs determined by the probability of erroneously inputting any given character. The editing distance calculator may then determine the editing distance between the inputted word and each of the word candidates by grid comparison using the variable error costs. The selector may choose one or more preferred candidates from the word candidates using the editing distances.