Abstract: A multilingual exchange system and message transmission method that enables exchange of messages in a foreign language in a simple manner without any special translation skill, and, simultaneously, allowing for dramatic improvement in grammatical accuracy when converting messages into a second language. In an outer box 106, which is a display area for a modifier in a third sentence pattern sentence inner boxes 106A, 106B, and 106C are allocated to each part of speech in English, which is the second language, and inner boxes 106A, 106B, and 106C can be recognized as separate inner boxes. Even if display content in a first language is divided into three inner boxes, there is no need to divide display content in second language into three inner boxes, and even if substitutions of parts of speech or changes in content occur in one language, it is possible to maintain grammatical accuracy in another language.

Abstract: The subject matter of this specification can be implemented in, among other things, a method that includes storing translator linguistic resources for a personal account and project linguistic resources for a corporate account. The method includes creating a private profile for the personal account in response to the personal account being added to a team for the corporate account or being assigned to a translation project for the corporate account. A graphical user interface provides access to the translator linguistic resources within the translation project for the private profile without providing access to the translator linguistic resources within the translation project for the corporate account. The graphical user interface provides translation suggestions from the project linguistic resources within the translation project for the private profile without providing full access to the project linguistic resources within the translation project for the private profile.

Abstract: A device and method is disclosed for automatically grouping data based on characteristics. The device and method include a communication interface for receiving data for component information, a storage medium for storing the received data, and a processor for performing an analysis on the received data to determine parameters included within the data and processing the data using correlations within the data to group or isolate data points. The correlations enable the processor to identify attributes associated with particular groups of the data.

Abstract: Sentence pairs can be filtered using dual monolingual cross-entropy-delta filtering. The filtered sentence pairs can be used to train a model for performing a language processing task, such as machine translation.

Abstract: Methods, systems, and computer readable media for machine translation between Arabic language and Arabic Sign Language are described.

Abstract: Input data is received, by a server, for a user from one or more data sources. A set of user language features associated with the input data are determined. The set of user language features is compared to multiple sets of known language features. Each set of known language features includes language features associated with the use of a non-native language by a speaker of a respective native language. A native language of the user is determined based on the comparing. A personalized recommendation is generated based on the determined native language of the user.

Abstract: A method of and system of for compressing and decompressing a localized software resource is disclosed. The method may include receiving a software resource, the software resource being in a first language, receiving a localized software resource for compression, where the software resource in the first language is a counterpart of the localized software resource in the second language. Upon receiving the software resources creating a first local dictionary for the localized software resource based at least in part on one or more first language words in the software resource and on data from a global dictionary, and compressing the localized software resource based on the local dictionary.

Abstract: Disclosed are augmented reality (AR) personalization systems to enable a user to edit and personalize presentations of real-world typography in real-time. The AR personalization system captures an image depicting a physical location via a camera coupled to a client device. For example, the client device may include a mobile device that includes a camera configured to record and display images (e.g., photos, videos) in real-time. The AR personalization system causes display of the image at the client device, and scans the image to detect occurrences of typography within the image (e.g., signs, billboards, posters, graffiti).

Abstract: A sentence evaluation apparatus evaluates a sentence which is input. The sentence evaluation apparatus includes an acquisition device and a processor. The acquisition device acquires information indicating a first input sentence and information indicating a second input sentence. The processor executes information processing on the information acquired by the acquisition device, using an algorithm based on machine learning. The processor includes a first encoder that recognizes the first input sentence and a second encoder that recognizes the second input sentence, in the algorithm based on the machine learning. The processor generates evaluation information indicating evaluation on the first input sentence with reference to the second input sentence, based on a result of recognition by the first encoder on the first input sentence and a result of recognition by the second encoder on the second input sentence.

Abstract: Devices, systems, and methods for data analysis and diagnostics utilizing Dynamic Word Entropy. A method includes: obtaining a text of a user; determining word entropy values which correspond to different lengths of text-portions of the text of the user; generating a Dynamic Word Entropy table which corresponds to the text of the user; analyzing the table, and determining whether or not the user has a particular medical condition, or determining whether or not a particular intervention has positively affected the user or has negatively affected the user or has not affected the user.

Abstract: Implementations of this disclosure provide methods and apparatuses for generating recommendation information. An example method includes matching text content from a text content library based on a plurality of predetermined scenario-related words; extracting keywords from the related text content, to generate a plurality of training samples; and for each training sample, providing a source sequence of a sequence pair corresponding to the training sample as an input to a recommendation information generation model, obtaining, from the recommendation information generation model, a predicted word, and adjusting a model parameter of the recommendation information generation model based on a comparison between the predicted word and a corresponding word in a target sequence of the sequence pair corresponding to the training sample, to train the recommendation information generation model.

Abstract: A non-transitory computer-readable storage medium storing a program that causes a computer to execute a process, the process includes learning distributed representations of words included in a word space of a first language using a learner for learning the distributed representations; classifying words included in a word space of a second language different from the first language into words common to words included in the word space of the first language and words not common to words included in the word space of the first language; and replacing distributed representations of the common words included in the word space of the second language with distributed representations of the words, corresponding to the common words, in the first language and adjusting a parameter of the learner.

Abstract: One embodiment provides a method, including: receiving, on a keyboard of an information handling device, typing input from a user, wherein the typing input comprises at least one letter; determining, using a processor, at least one completion suggestion stemming from the at least one letter; and presenting, responsive to the determining, the at least one completion suggestion to the user, wherein the presented at least one completion suggestion comprises at least one animation portraying a signed motion corresponding to the at least one completion suggestion. Other aspects are described and claimed.

Abstract: A voice interaction system performs a voice interaction with a user. The voice interaction system includes: ask-again detection means for detecting ask-again by the user; response-sentence generation means for generating, when the ask-again has been detected by the ask-again detection means, a response sentence for the ask-again in response to the ask-again based on a response sentence responding to the user before the ask-again; and storage means for storing a history of the voice interaction with the user. The response-sentence generation means generates, when the response sentence includes a word whose frequency of appearance in the history of the voice interaction in the storage means is equal to or smaller than a first predetermined value, a response sentence for the ask-again formed of only this word or a response sentence for the ask-again in which this word is emphasized in the response sentence.

Abstract: The exemplary embodiments disclose a method, a computer program product, and a computer system for identifying one or more objects of interest in augmented reality. The exemplary embodiments may include detecting one or more cues selected from a group comprising one or more audio cues and one or more visual cues, identifying one or more objects of interest based on the detected one or more cues and a model, and emphasizing the one or more objects of interest within an augmented reality.

Abstract: A training set is created via creating adjacent classified substrings by using character classes to replace corresponding characters in adjacent substrings in each training character string, and associating each pair of adjacent classified substrings and each pair of adjacent substrings with corresponding labels indicating whether corresponding pairs include any token boundary. The system splits input character string into beginning and ending parts and creates classified beginning part by replacing beginning part character with corresponding class and classified ending part by replacing ending part character with corresponding class. The machine-learning model determines probability of token identification, based on training set to determine count of instances that classified beginning part is paired with classified ending part and count of corresponding labels that indicate inclusion of any token boundary.

Abstract: A computer based system and method for identifying complaint interactions, including: detecting appearances of linguistic structures related to complaints in an interaction; calculating at least one sentiment metric of the interaction; and classifying the interaction as being or not being a complaint interaction based on the detected linguistic structures and the at least one sentiment metric, for example using a trained supervised learning model.

Abstract: A method, system and computer-usable medium are disclosed for the use of a text simplification in a question answering (QA) system to improve ingestion quality and QA performance. A source corpus is processed to generate a parsed source corpus, which in turn is processed to generate a shadow corpus of simplified text. The parsed source corpus and the shadow corpus are then processed to generate derived data resources. A user query is processed to generate a set of merged candidate answer variants which are in turn processed to generate a corresponding converged feature vector representing each merged candidate answer variant. Feature values associated with each converged feature vector are then used to rank the merged candidate answer variants. A ranked set of merged candidate answer variants is then provided to the user.

Abstract: A natural language understanding server includes grammars specified in a modified extended Backus-Naur form (MEBNF) that includes an agglutination metasymbol not supported by conventional EBNF grammar parsers, as well as an agglutination preprocessor. The agglutination preprocessor applies one or more sets of agglutination rewrite rules to the MEBNF grammars, transforming them to EBNF grammars that can be processed by conventional EBNF grammar parsers. Permitting grammars to be specified in MEBNF form greatly simplifies the authoring and maintenance of grammars supporting inflected forms of words in the languages described by the grammars.

Abstract: A query generation system receives, from a first device, a first input and a first project identifier and receives, from a second device, a second input and a second project identifier. The first and second inputs are the same and are in a natural language format that is not compatible with a downstream database management system. The system generates, based on the first input, a first database query. The system generates, based on the second input, a second database query. The first and second database queries are compatible with the downstream database management system. The system receives a first response to the first database query and a second response to the second database query from the downstream database management system. The system transmits the first response to the first device and the second response to the second device.

Abstract: Embodiments of the present invention are directed to a computer-implemented method for translating a content item. Aspects include obtaining a user profile for a reader of the content item, the user profile comprising a native language of the reader, a second language of the reader and skill level associated with the second language. Aspects also include obtaining the content item, wherein the content item is in the second language. Aspects further include creating a translated content item by determining, for each phrase in the content item, a confidence score that the reader will understand the phrase based at least in part on the user profile and replacing each phrase in the content item having the confidence score less than a threshold level with an alternate phrase. Aspects also include presenting the translated content item to the reader.

Abstract: Embodiments of the disclosure provide systems and methods for segmenting a sentence. The method may include identifying a first phrase in the sentence associated with a first segmentation path, determining a first group of derivative phrases semantically associated with the first phrase, determining a first evaluation score based on modified sentences generated by replacing the first phrase with the respective derivative phrase in the first group, and segmenting the sentence based on the first evaluation score.

Abstract: An approach is provided to use a first translation attribute that is received at a user interface from a user to automatically translate a document. The source document that is in a source natural language is translated to a target document that is in a target natural language by using a machine translator that utilizes the first translation attribute, such as adequacy or fluency. The target document is analyzed with the analysis resulting in a second translation attribute (e.g., either adequacy or fluency, whichever is different from the first translation attribute). The target (translated) document and the second translation attribute are then provided to the user, such as at the user interface.

Abstract: A media drive 32 is loaded with a data disc 44b in which a plurality of files composing application software is recorded. A language information holding section 110 holds use language information configured to specify a use language selected by a user in the information processing apparatus 10 concerned. For causing application to be in an executable state, a recording processing section 104 copies a language-dependent file recorded in the data disc 44b to an auxiliary storage apparatus 2 on the basis of the use language information held in the language information holding section 110.

Abstract: A display device for a vehicle is provided, the device including personal information acquisition section, environment information acquisition section, information conversion section and display section. The personal information acquisition section acquires personal information about a driver of the vehicle from an identification tag in which the personal information of the driver is recorded. The environment information acquisition section acquires environment information about a vicinity of the vehicle. The information conversion section converts the environment information in accordance with the personal information of the driver. The display section displays the converted information to the interior of a passenger compartment of the vehicle.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for neural machine translation. One of the systems includes an encoder neural network comprising: an input forward long short-term memory (LSTM) layer configured to process each input token in the input sequence in a forward order to generate a respective forward representation of each input token, an input backward LSTM layer configured to process each input token in a backward order to generate a respective backward representation of each input token and a plurality of hidden LSTM layers configured to process a respective combined representation of each of the input tokens in the forward order to generate a respective encoded representation of each of the input tokens; and a decoder subsystem configured to receive the respective encoded representations and to process the encoded representations to generate an output sequence.

Abstract: Text is extracted from and information resource such as documents, emails, relational database tables and other digitized information sources. The extracted text is processed using a decomposition function to create. Nodes are a particular data structure that stores elemental units of information. The nodes can convey meaning because they relate a subject term or phrase to an attribute term or phrase. Removed from the node data structure, the node contents are or can become a text fragment which conveys meaning, i.e., a note. The notes generated from each digital resource are associated with the digital resource from which they are captured. The notes are then stored, organized and presented in several ways which facilitate knowledge acquisition and utilization by a user.

Abstract: A semantic extraction method based on a natural language is provided. In a character library based on a single-character granularity, a character of an input natural language is queried, to obtain first semantic vectors representing semantic descriptions of the character. Candidate semantic intentions corresponding to second semantic vectors and weight vectors corresponding to the candidate semantic intentions are obtained, wherein the second semantic vectors are obtained by using a neural network and correspond to different permutations and combinations of the first semantic vectors. A candidate semantic intention is selected as a final semantic intention of the natural language based on a preset condition, and key information is obtained corresponding to the final semantic intention according to a weight vector corresponding to the final semantic intention.

Abstract: Provided is a system and method for retrieving a translation of a source word based on context. For example, the context may include other words in a same file as the source word. The context can be used to identify the correct semantic meaning of the source word when the word has multiple contextual meanings. In one example, the method may include identifying a source word from a data file, determining a plurality of translation candidates for the source word which translate the source word from a source language into a different language, retrieving a target translation candidate for the source word from among the plurality of translation candidates based on context of the source word included in the data file, and outputting the translated target translation candidate for display via a display device.

Abstract: A multilingual translation device and method. The multilingual translation device may obtain a multilingual sentence, obtain vector values corresponding to words included in the multilingual sentence by using a multilingual translation model, convert the obtained vector values into vector values corresponding to a target language, and obtain a sentence in the target language, based on the vector values corresponding to the target language.

Abstract: An audio processing device (100) includes a correction microphone (110), a reference microphone (120), a classification unit (130), an active noise cancellation (ANC) unit (140), and a speaker (150). The correction microphone is arranged in an area adjacent to an ear of a user of the device and picks up acoustic signals in the area and outputs same as a correction signal. The reference microphone picks up surrounding noises (124) from a surrounding area of the user and outputs same as a reference signal. The classification unit receives the reference signal and splits the reference signal corresponding to a spectrum and amplitude thereof into signal components of two or more classes and outputs a classified signal. The ANC unit receives and processes the correction signal based on the classified signal and outputs a corresponding audio signal. The speaker receives the audio signal and provides a corresponding acoustic signal output.

Abstract: A method includes determining a trainable model to provide to a trainer, the trainable model determined based on modification of one or more models of a plurality of models. The plurality of models is generated based on a genetic algorithm and corresponds to a first epoch of the genetic algorithm. Each of the plurality of models includes data representative of a neural network. The method also includes providing the trainable model to the trainer. The method further includes adding a trained model, output by the trainer based on the trainable model, as input to a second epoch of the genetic algorithm, the second epoch subsequent to the first epoch.

Abstract: Methods, systems, and computer readable mediums for network hardware table management including obtaining, by a network device table manager of a network device, a first feature table entry published by a first feature; obtaining, by the network device table manager, a second feature table entry published by a second feature; making a first determination that the first feature table entry and the second table feature entry each comprise a common prefix; and based on the first determination, adding a first combined feature table entry to a combined feature table, the first combined feature table entry comprising the common prefix, a first feature action of the first feature table entry and a second feature action of the second feature table entry.

Abstract: A set of documents is parsed. Members of the set of documents include a set of text elements and a set of visual elements. A text content stream based on the set of text elements and a visual content stream based on the set of visual elements are produced. For respective documents, a set of respective visual element summarizations is built from the visual content stream. Each visual summarization includes a textual description of a respective visual element in the respective document. A holistic index is created by indexing the text content from the text content stream and the text descriptions of the visual elements for each document in a single search index. The indexing uses a set of semantic relationships between the text content from the text content stream and the textual descriptions of the visual elements. A user interface allows a user to selectively search text content and visual content.

Abstract: For a chat bot, first intent data is provided, including a first instance of sample utterances in a first natural language and intents corresponding to the respective sample utterances. The sample utterances are translated by a computer system to a second natural language as a second instance of the sample utterances. The sample utterances of the second instance are translated by a computer system back to the first natural language as a third instance of the sample utterances. Second intent data is provided including the third instance of the sample utterances and the corresponding intents. An intent classifier of the chat bot is trained to identify respective intents of real time utterances, wherein the identifying is responsive to the intent classifier receiving the real time utterances from a user when the chat bot is in an operating mode and the training includes training the intent classifier on the first and second intent data.

Abstract: Apparatuses, methods, and program products are disclosed for converting multiple language variations. One apparatus includes a processor, a sensor, and a memory that stores code executable by the processor. The code is executable by the processor to: determine a priority ranking corresponding to each language variation of multiple language variations, wherein each language variation includes a language, a dialect, or a combination thereof; detect, by use of the sensor, an audible input; convert the audible input to text based on a first language variation, wherein the priority ranking of the first language variation is a highest priority; and in response to a portion of the text being incorrect, convert the audible input corresponding to the portion of the text to a revised portion of the text based on a second language variation, wherein the priority ranking of the second language variation is a second highest priority.

Abstract: A processor-implemented machine translation method includes translating a source sentence expressed in a source language into a target language to determine a target sentence, determining, based on a reliability of the target sentence, whether the target sentence is appropriate as a translation result of the source sentence, and re-determining a target sentence corresponding to the source sentence in response to a result of the determining being that the target sentence is inappropriate as the translation result.

Abstract: Translation review workflow systems and methods are disclosed herein. Some embodiments include a web content management system having a translation preview agent, the web content management system managing web content for publishing, a translation system that manages translation content of the web content, and a translation preview service that generates a translation preview of the web content with the translated content that is embedded within the web content to maintain context for a translator.

Abstract: A self-learning translation-engine brokering system characterizes a set of language-translation engines by associating each engine with values of a set of engine parameters. The system receives a request to translate text or speech input, along with a set of weightings that identify the relative importance of each engine parameter to the translation requester. The system formats the input into a quantifiable engine-agnostic form and performs an optimization procedure that finds the best fit between the source's weightings and the sets of engine parameters. The system directs the input to the best-fitting translation engine, receives the translated output from the selected engine, directs the output to the translation requester, and then determines how well the translated output meets user expectations. This feedback used to update the best-fitting engine's parametric values and to train the system to make more accurate selections in the future.

Abstract: An information processing apparatus includes a receiving unit, an obtaining unit, and a translating unit. The receiving unit receives a document as a target to be translated. The obtaining unit obtains a non-target region from the document, the non-target region being a region that is not a target to be translated. The translating unit does not perform translation on the non-target region but performs translation on a portion other than the non-target region.

Abstract: A method for automatically detecting meaning patterns in a text that includes input words of at least one sentence, includes a database system containing words of a language, a number of defined categories of meaning in order to describe properties of the words, and meaning signals for all the words stored in the database, wherein a meaning signal is a clear numerical characterization of the meaning of the word using the categories of meaning.

Abstract: A response sentence that is satisfactory to a user is formed. A response device (1) includes: a related term searching section (13) configured to, in a case where an input sentence contains an unknown word, detect a related term, which is a known word that shares, with the unknown word, at least one shared word; and a response sentence forming section (16) configured to form the response sentence whose content is related to the related term that has been detected by the related term searching section (13).

Abstract: A sentence mapping method includes obtaining a source language document in a source language and a target language document in a target language, wherein the target language document is a translation of the source language document, generating a translated document by translating the target language document into the source language, and mapping source language sentences in the source language document and target language sentences with the target language document by comparing the source language document and the translated document.

Abstract: Of the four primary approaches to processing language by computer, only the parsing approach considers the semantic and syntactic components from the start. In doing so, however, the required resources expand rapidly as the scope of the language processed increases. And as that scope increases, the performance of parsing systems decreases. A natural language processor uses a tumbling-frequency phrase chain parser as described herein which circumvents this resource-intensive step in parsing, while quickly and almost effortlessly arriving at the next step in natural-language processing with far more accurate results involving a partitioning dictionary and phrase chains.

Abstract: A method of cognitive translation integrated with context sensitive derivations is provided. The method is implemented via a cognitive translation engine that receives, from an integrated development environment, program-integrated information relationships with respect to a message package. The program-integrated information relationships includes the context sensitive derivations. The cognitive translation engine translates program-integrated information translation requests from a source language into a target language according to the context sensitive derivations. The program-integrated information translation requests are embedded with program-integrated information relationships.

Abstract: A method for text classification involves generating, using a bilingual embedding model, source language embeddings for source language documents; obtaining source language document labels of the source language documents; and training a source language classifier model and a label embedding network, executing on a computing system, using the source language embeddings and the source language document labels. The method further involves generating pseudo-labels for unlabeled target language documents, by: generating, using the bilingual embedding model, target language embeddings for the unlabeled target language documents, and applying the source language classifier model and the label embedding network to the target language embeddings to obtain the pseudo-labels for the unlabeled target language documents. In addition, the method involves training a target language classifier model executing on the computing system using the target language embeddings and the pseudo labels.

Abstract: A processing device identifies an application that is in a first language and has a first binary resource file that includes first resource data for a first set of resources of the application in the first language, determines that the application is translated into a second language and has a second binary resource file that includes second resource data for a second set of resources of the application in the second language, determines differences between functionality of the second set of resources of the application in the second language and functionality of the first plurality of resources of the application in the first language based on the first binary resource file and the second binary resource file, and modifies one or more of the second plurality of resources of the application in the second language based on at least one of the differences between the functionality of the second plurality of resources and the functionality of the first plurality of resources of the application.

Abstract: System and method for providing a computer-assisted translation from a source language to a target language, using a generic NMT model and a translation memory. An input text segment is received, and input context information is received. The input context information is indicative of circumstances in which the input text segment is used, the input text segment being in the source language. An estimated translation of the input text segment into the target language is calculated, using a generic neural machine translation “NMT” model for providing a generalised machine translation from the source language to the target language, and a translation memory comprising translation elements; each translation element comprising a source language text segment, a corresponding target language text segment. The estimated translation is provided to a user, for correction by the user.

Abstract: The present disclosure pertains to modeling free-text clinical documents into a hierarchical graph-like data structure based on semantic relationships among clinical concepts present in the documents. A method comprises parsing, identifying, and annotating clinical terms within free-text clinical documents. This is accomplished by storing identified clinical terms in a concept node. The concept node is a data structure that has a set of properties to categorize stored concepts. Clinical concepts of free-text clinical documents are classified into clinical event classes. The free-text clinical documents include clinical terms that were associated with clinical concept categories. Classifying clinical concepts includes organizing clinical text-free documents into sections that describe a specific aspect of the clinical text-free documents that include one or more of clinical, technical, or administrative aspects of the documents.

Abstract: Original idea extraction from written data is provided by capturing expression as written text data, obtaining a knowledge graph representing concepts and relationships between the concepts automatically topic modeling the written text data to ascertain thought units and identify respective concepts of the thought unit, mapping a thought unit to the knowledge graph, determining that the thought unit is an original idea based on a graph distance in the knowledge graph between correlated concepts represented in the knowledge graph, and based on determining that the thought unit is an original idea, storing a representation of the original idea to an idea repository and invoking processing of at least one computer.