Abstract: Implementations of the present disclosure are directed to a method, a system, and a computer program storage device for identifying a language in a message. Non-language characters are removed from a text message to generate a sanitized text message. An alphabet and/or a script are detected in the sanitized text message by performing at least one of (i) an alphabet-based language detection test to determine a first set of scores and (ii) a script-based language detection test to determine a second set of scores. Each score in the first set of scores represents a likelihood that the sanitized text message includes the alphabet for one of a plurality of different languages. Each score in the second set of scores represents a likelihood that the sanitized text message includes the script for one of the plurality of different languages.

Abstract: A grammatically diverse test set of natural language sentences for a deep question answering system is provided by analyzing a given sentence to characterize its syntactical classification, and adding the sentence to the test set if its classification is sufficiently different from other sentences already in the test set. A particular sentence may be selected for inclusion according to a desired syntactic distribution. Multiple sentences having the exact same classification may be allowed subject to a maximum number of such sentences. The test set is adapted to an element of interest by characterizing each syntactical classification relative to the element of interest. The analysis derives a parse tree, identifies a particular node of the tree corresponding to the element of interest, and extracts syntactic information by traversing the tree starting at the particular node and ending at the root node of the tree according to different traversal schemes.

Abstract: Systems and methods for domain classification in natural language processing based on domains are disclosed. The method includes generating a trigram corpus for the purpose of classification based on a trigram analysis of a domain model containing a hierarchical ontology and semantic construction that maps patterns of semantic tokens to syntactic patterns. An input string is parsed within each domain, tokenized in each domain. The resulting trigrams for the input text in each domain are looked up in the corresponding trigram corpus to determine the relevancy of each domain to the input text. The input string is thus classified based on the relevancy determination. The systems and methods avoids having to rely on existing annotated domain corpora for classification and allows for fast regeneration of the classifier when domain models are under frequent update and development.

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for bootstrapping a language translation system. A system configured to practice the method performs a bidirectional web crawl to identify a bilingual website. The system analyzes data on the bilingual website to make a classification decision about whether the root of the bilingual website is an entry point for the bilingual website. The bilingual site can contain pairs of parallel pages. Each pair can include a first website in a first language and a second website in a second language, and a first portion of the first web page corresponds to a second portion of the second web page. Then the system analyzes the first and second web pages to identify corresponding information pairs in the first and second languages, and extracts the corresponding information pairs from the first and second web pages for use in a language translation model.

Abstract: Provided is a method, device, and computer-readable medium for converting a string of characters in a first language into a phonetic representation of a second language using a first data structure that maps graphemes in the first language to one or more universal phonetic representations based on an international phonetic alphabet, wherein the first data structure comprises a plurality of first nodes with each first node of the plurality of first nodes having a respective weight assigned that corresponds to a likely pronunciation of a grapheme, and a second data structure that maps the one or more universal phonetic representations to one or more graphemes in the second language, wherein the second data structure comprises a plurality of second nodes with each second node of the plurality of second nodes having a respective weight assigned that corresponds to a likely representation of a grapheme in the second language.

Abstract: Provided are methods, devices, and computer-readable media for generating a string of characters based on a set of rules; parsing the string of characters into string of graphemes; determining one or more phonetic representations for one or more graphemes in the string of graphemes based on a first data structure; determining at least one grapheme representation for one or more of the one or more phonetic representations based on a second data structure; and constructing the phonetic representation of the string of characters based on the grapheme representation that was determined.

Abstract: A call translation system and a method for operating the management system in a controlled environment is disclosed. The call translation system includes a calling device and a call translation server. The calling device receives authorization from the call translation server to conduct call sessions within the controlled environment. The calling device further receives a call session application from the call translation server that enables the calling device to initiate and establish call sessions through the call translation server. The call session application allows a user of the calling device to select outside parties to add to the call session.

Abstract: Systems, methods, and apparatus for accessing distributed models in automated machine processing, including using large language models in machine translation, speech recognition and other applications.

Abstract: A method to facilitate a language independent search includes displaying a search attribute, including an attribute name and a selection of attribute values, to a user and receiving from the user a search attribute value selection corresponding to the attribute name. Then, the attribute name and attribute value selection are translated into language independent symbols, such as numbers. The numbers are used in a search of a sellers item listings table to select item listings with attribute values that match the search attribute values selected by the user.

Abstract: Embodiments create and label contextual slices from observation data and aggregate slices into a hierarchical storyline for a user. A context is a (possibly partial) specification of what a user was doing in the dimensions of time, place, and activity. A storyline is composed of a time-ordered sequence of contexts that partition a given span of time that are arranged in groups at one or more hierarchical levels. A storyline is created through a process of data collection, slicing, labeling, and aggregating. Raw context data can be collected from a variety of observation sources with various error characteristics. Slicing refines the raw context data into a consistent storyline composed of a sequence of contexts representing homogeneous time intervals. Labeling adds more specific and semantically meaningful data (e.g., geography, venue, activity) to the slices. Aggregation identifies groups of slices that correspond to a single semantic concept.

Abstract: A method includes receiving one or more natural language dependency parse trees as input. A hardware processor is used for processing the dependency parse trees by creating a mapping from nodes of the one or more dependency parse trees into actions, roles and contextual predicates. The mapping is used for information extraction. The actions include the verbs along with attributes of the verbs. The roles include arguments for the verbs. The contextual predicates include modifiers for the verbs.

Abstract: Specialized language processing engines can use author-specific or reader-specific language models to improve language processing results by selecting phrases most likely to be used by an author or by tailoring output to language with which the reader is familiar. Language models that are author-specific can be generated by identifying characteristics of an author or author type such as age, gender, and location. An author-specific language model can be built using, as training data, language items written by users with the identified characteristics. Language models that are reader-specific can be generated using, as training data, language items written by or viewed by that reader. When implementing a specialized machine translation engine, multiple possible translations can be generated. An author-specific language model or a reader-specific language model can provide scores for possible translations, which can be used to select the best translation.

Abstract: A method, apparatus, and computer program product for describing motion. The method may include receiving a set of eventualities (114). The set of eventualities (114) may describe at least one of a domain event and a domain state. The at least one of the domain event and the domain state may be derived from a set of spatio-temporal data (102) and the set of eventualities (114) may be associated with a particular region and a particular time period. The method may include organizing the set of eventualities to generate a document plan. The method may further include generating, using a processor, a linguistic representation of the set of eventualities using the document plan.

Abstract: Enterprise data sources can be monitored to detect metric conditions via rules, and alerts can be generated. The alerts can be presented as natural language descriptions of metric conditions. From an alert, the reader can navigate to a story page that presents additional detail and allows further navigation within the data. Additional detail presented can include a drill down synopsis, strategies for overcoming a negative condition, links to discussions within the organization about the condition, options for sharing or collaborating about the condition, or the like.

Abstract: A system provides on a computer display screen, a graphical image of an initial parsing as generated by a computerized algorithm. Individual clauses are encased in expandable containers that a user can shift and resize on the screen using a mouse or other cursor control device to thereby improve the initial parsing.

Abstract: Concepts and technologies are described herein for providing topically aware word suggestions. Using a text input, the system determines a conditional count and an unconditional count. The system then determines an adjustment factor for a pair of words of the plurality of words based on the unconditional count and the conditional count. The system then generates a data structure defining a plurality of word clusters. The system then reconstructs the adjustment factor of the pair of words based on a number of common clusters between individual words of the pair of words. The adjustment factor is combined with other data, such as data from a language model dictionary and a freshness factor from an average cluster activation state table to determine a probability associated with a word candidate, which is displayed to a user.

Abstract: A method and server for pushing a cellular lexicon. The server is used for receiving input method log information sent by a client; acquiring a client identifier and a client-selected word in the input method log information; searching for an undownloaded cellular lexicon in an undownloaded cellular lexicon list corresponding to the client identifier, acquiring a first matching lexicon comprising the client-selected word in the undownloaded cellular lexicon, and adding a first weight value to a lexicon weight value of the first matching lexicon corresponding to the client identifier; pushing the first matching lexicon to a corresponding client when the lexicon weight value is greater than a pushing threshold value; and deleting the pushed first matching lexicon. The server can push a cellular lexicon actually needed by the client, thereby increasing the input precision regarding a certain category of entries, and improving the input efficiency of the client.

Abstract: Provided are methods, devices, and computer-readable media for accessing a string of characters; parsing the string of characters into string of graphemes; determining one or more phonetic representations for one or more graphemes in the string of graphemes based on a first data structure; determining at least one grapheme representation for one or more of the one or more phonetic representations based on a second data structure; and constructing the phonetic representation of the string of characters based on the grapheme representation that was determined.

Abstract: Method for text processing executable at a computing device, comprising appreciating a source phrase comprised of source words; appreciating a target phrase comprised of target words; associating a respective source word feature set with each one of the source words; associating a respective target word feature set with each one of the target words; analyzing source word feature sets and target word feature sets; and based on the analysis, generating one or more phrase transformation rules for transforming the source phrase into the target phrase. Also a server and non-transitory computer-readable medium storing program instructions for carrying out the method.

Abstract: Various embodiments described herein facilitate multi-lingual communications. The systems and methods of some embodiments enable multi-lingual communications through different modes of communication including, for example, Internet-based chat, e-mail, text-based mobile phone communications, postings to online forums, postings to online social media services, and the like. Certain embodiments implement communication systems and methods that translate text between two or more languages. Users of the systems and methods may be incentivized to submit corrections for inaccurate or erroneous translations, and may receive a reward for these submissions. Systems and methods for assessing the accuracy of translations are described.