Abstract: Systems for distributed graph data management that use graph structures to model data and respond to different types of queries. Systems that provide analytic tools and machine learning based on graph structures. Systems for distributed graph data management that provide packaged solutions for different application areas.

Abstract: A processing method includes acquiring first parallel text data which is a pair of a first text described in a first language and a first translation of the first text into a second language; evaluating whether the first parallel text data has a possibility of mistranslation on the basis of words included in the first text and the first translation; and outputting information based on the evaluation result. It is evaluated whether the first parallel text data has a possibility of mistranslation on the basis of results of (1) a first determination as to whether the first text includes either or both of a word with a first meaning and an antonym of the word and (2) a second determination as to whether the first translation includes either or both of a translation word with the first meaning and an antonym of the translation word.

Abstract: Embodiments of the present invention provide methods, systems, apparatuses, and computer program products for aiding user access to technical data for an item. In one embodiment a method is provided comprising providing a model of the item, wherein the model comprises a plurality of visual representations, each visual representation representing one or more components of the item; receiving input selecting at least one of the one or more visual representations; responsive to receiving input selecting the at least one of the one or more visual representations, identifying one or more topics associated with the selected visual representation; providing at least a portion of the one or more topics associated with the selected visual representation; receiving input selecting at least one of the one or more topics; and responsive to receiving input selecting a topic of the one or more topics, providing technical data associated with the topic.

Abstract: A system and method for improving accuracy of natural language processing using a plurality of speech recognition engines, a data fusion model to identify a correct result from the plurality of speech recognition engines and a semantic understanding model, separate and distinct from the speech recognition model, to process the correct results. A corpus is developed using the correct results and the corpus is used to train the data fusion model and the semantic understanding model.

Abstract: Embodiments are directed to a computer-implemented method for ranking a proximity of data sources with one or more reference entities, which may be authoritative entities. The method comprises identifying several data sources and one or more reference entities. For each of the identified reference entities a vector of terms representative of said each of the reference entities is obtained. For each data source of the identified data sources, terms from said each data source are selected, an array comprising the selected terms can be accordingly populated, and a term frequency of terms of the vector obtained for said each of the reference entities is determined, in said array. The term frequency corresponds to the number of occurrences of said each of the terms in said array. The method ranks a proximity of the identified data sources with each of the one or more reference entities.

Abstract: Mechanisms are provided to perform embedding of content of a natural language document. The mechanisms receive a document data object of an electronic document and analyze a structure of the electronic document to identify one or more structural document elements that have a relationship with the document data object. A dependency data structure is generated, representing the electronic document, where edges define relationships between document elements and at least one edge represents at least one relationship between the one or more structural document elements and the document data object. The mechanisms embed the document data object based on the at least one relationship to thereby represent the document data object as a vector data structure. The mechanisms perform natural language processing on the portion of natural language content based on the vector data structure. The one or more structural document elements are non-local non-contiguous with the document data object.

Abstract: A mechanism is provided in a data processing system comprising at least one processor and at least one memory, the at least one memory comprising instructions executed by the at least one processor to cause the at least one processor to implement a cognitive natural language processing system. The cognitive natural language processing (NLP) system analyzes a portion of natural language text to identify an attribute specified in the natural language text. The cognitive NLP system analyzes the portion of natural language text to determine whether a known negation trigger is present in the natural language text in association with the attribute. In response to determining that the natural language text does not contain a known negation trigger in association with the attribute, the cognitive NLP system determines whether the attribute is negated based on instances of the attribute in other natural language content similar to the natural language text.

Abstract: An advertisement translation device is a device for translating an advertisement text displayed in a display area of a predetermined size in advertisement content, and displaying the translated advertisement text in the display area. The advertisement translation device includes: an advertisement text obtainment unit which obtains the advertisement text; a linguistic information obtainment unit which obtains linguistic information indicating a target language into which the advertisement text is to be translated; a translation unit which create translated text candidates by translating the advertisement text into the target language indicated by the linguistic information; and a selection unit which selects and outputs as a translated text, at least one of the translated text candidates a full text of which is displayable in a predetermined character size in the display area.

Abstract: The program directs a computer processor to resolve an anaphor in electronic natural language text. The program detects a plurality of entities and an anaphor in a span of parsed natural language text comprising one or more sentences, and extracts pairs of related entities based on domain knowledge. The program constructs a set of tuples, wherein each tuple is a data type comprising an anaphor, an antecedent entity (AE) appearing before the anaphor in the span of parsed natural language text, and an entity (E) appearing after the anaphor in the span of parsed natural language text, wherein the anaphor refers to the AE and relates the AE to the E. The program resolves the anaphor by determining which entity in the plurality of entities the anaphor references, using the constructed set of tuples, and selecting an AE among one or more candidate AEs.

Abstract: Examples of techniques for the real-time capture and translation of human thoughts and ideas into structured patterns are disclosed. In one example implementation according to aspects of the present disclosure, a computer-implemented method may include capturing, by a processing device, unstructured data. The method may also include extracting key terms from the unstructured data. Additionally, the method may include assigning an attribute to at least one of the key terms. The method may further include generating, by the processing device, a structured pattern based on the key terms and the attributes.

Abstract: Mechanisms receive natural language content and analyze the natural language content to generate a parse tree data structure. The mechanisms process the parse tree data structure to identify one or more instances of candidate hypothetical spans in the natural language content. Hypothetical spans are terms or phrases indicative of a hypothetical statement. The mechanisms calculate, for each candidate hypothetical span, a confidence score value indicative of a confidence that the candidate hypothetical span is an actual hypothetical span based on a personalized hypothetical dictionary data structure associated with a source of the natural language content. The mechanisms perform an operation based on the natural language content.

Abstract: The present disclosure contemplates a variety of improved methods and systems for enabling set up of a variety of disparate IoT devices coupled to the framework and/or ambient operating system of an assistant device. The described solution includes a voice-driven assistant device setup process. An assistant device can determine the IoT devices in its physical environment and provide setup instructions to the user. The setup including determining voice activatable commands, device information and the adapter to allow the assistant device to operate the one or more IoT devices in response to user instructions.

Abstract: A method includes hierarchically identifying occurrences of some of the words in the set of sentences; creating a first index for each of some of the words based on the upper hierarchy of occurrences identified for each word; receiving input of a queried word; hierarchically identifying occurrences of the queried word in the set of sentences; creating a second index based on the upper hierarchy of occurrences identified for the queried word; comparing the first index and the second index to calculate an estimated value for the number of occurrences of a word in the neighborhood of the queried word; and calculating the actual value of the number of occurrences of a word in the neighborhood of the queried word based on an upper hierarchy and lower hierarchy of the occurrences on condition that the estimated value is equal to or greater than a predetermined number.

Abstract: The present teaching relates to automatic formality classification and transformation of online text items. In one example, a request is received for transforming a formality level of a text item in an online communication. A current formality level of the text item is obtained. The current formality level represents a current degree of formality of the text item. A target formality level is determined for the text item based on the request. The target formality level represents a targeted degree of formality for the text item. The text item having the current formality level is transformed to a transformed text item having the target formality level. The transformed text item has a same literal meaning as the text item. The transformed text item is provided as a response to the request.

Abstract: An apparatus has a network interface circuit to receive a source sentence from a network connected client device. A processor is connected to the network interface circuit. A memory is connected to the processor. The memory stores translation data and instructions executed by the processor. The instructions executed by the processor operate a neural machine translation system. A translation hypothesis is formed from a prefix of a target sentence comprising an initial sequence of target words supplied by a user through an interface. The hypothesis is generated by the neural machine translation system that performs a constrained prefix decoding that repeatedly predicts a next word from previous target words. A suffix of the target sentence comprising a final sequence of words corresponding to a final sequence of words in the source sentence is formed using a beam search that constrains translation to match the prefix.

Abstract: A system and method for monitoring conversations of a community of users to identify topics of interest is provided. A user community which is based partly on social networking connections relative to a first user is identified. Conversations involving at least one member of the identified user community are monitored. Based in part on an aggregated analysis of the monitored conversations, keywords are selected to present to the first user. The first user is then provided with a display in which the selected keywords associated with the user community are presented to the first user such that the first user can select a keyword to access content associated therewith.

Abstract: Systems and methods are disclosed for extracting structured relation information from Chinese clinical notes. Structured relation information in the form of entity-feature-value (EFV) triples is disclosed. The EFV triples may be used for advanced data abstraction. An ontology-driven method for data extraction is disclosed. The ontology-driven method may include lexical and syntactic analysis followed by semantic analysis to generate the EFV triples.

Abstract: Systems and methods of automated ontology development include a corpus of communication data. The corpus of communication data includes communication data from a plurality of interactions and is processed. A plurality of terms are extracted from the corpus. Each term of the plurality is a plurality of words that identify a single concept within the corpus. An ontology is automatedly generated from the extracted terms.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, can aggregate content from one or more electronic publication documents. In one aspect, a method includes obtaining information including an index to content items of an electronic publication including a discrete package of received data that is stored locally, the content items being less than all content in the electronic publication received; retrieving, by a computer based on one or more criteria, two or more of the content items provided in disparate portions of the electronic publication, including the discrete package of received data, using the index; and presenting the two or more of the content items together on an output device in a user interface format different from that of the electronic publication.

Abstract: A method and apparatus for training a neural network language model, and a method and apparatus for recognizing speech data based on a trained language model are provided. The method of training a language model involves converting, using a processor, training data into error-containing training data, and training a neural network language model using the error-containing training data.

Abstract: A display apparatus includes an input interface that receives a user input for selecting a channel, a tuner that receives a broadcast signal corresponding to the channel, a display that displays a content included in the broadcast signal, a storage configured to store channel information obtained from the broadcast signal, a communicator configured to communicate with an external apparatus, and a controller that selects a channel based on the user input, displays content corresponding to the changed channel on the display, acquires a channel number, a source ID and a channel name with respect to the channel, and transmits the channel number, the source ID and the channel name to a server.

Abstract: A computer-implemented method for automatically generating passwords that comply with password restrictions may include (1) maintaining a database that stores password criteria for a plurality of websites, (2) determining that a user is accessing a website that requests creation of a password, (3) determining a set of password complexity rules for the website by querying the database that stores the password criteria, (4) using the set of password complexity rules to automatically generate the password for the website such that the password complies with the password criteria for the website, and (5) providing the password for use in the website that requested creation of the password. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: Systems and methods for content aggregation creation are disclosed herein. The system can include memory having a content database and an aggregation database. The system can include a user device having a first network interface and a first I/O subsystem. The system can include a server that can: provide content to the user device via a first electrical signal; receive a selection of a portion of the provided content from the user device via a second electrical signal; automatically extract sentences from the selected portion of the provided content via a natural language processor; automatically generate a parse tree for one of the automatically extracted sentences; identify noun phrases from the part of speech tags within the parse tree; place content associated with one of the noun phrase in a content aggregation; and output the content aggregation to the user device.

Abstract: A system for performing semantic search receives an electronic text corpus and separates the text corpus into a plurality of sentences. The system parses and converts each sentence into a sentence tree. The system receives a search query and matches the search query with one or more of the sentence trees.

Abstract: An auction apparatus according to the application includes a plan receiving unit and an auction providing unit. The plan receiving unit receives plan information indicating a plan of content from a content creator who creates the content. The auction providing unit presents the plan information received by the plan receiving unit and provides an auction service in which bid targets are a right to display an advertisement in an advertising space that is set to the content and a right to replace the advertisement displayed in the advertising space.

Abstract: In one implementation, a computer-implemented method for providing a graphical user interface (GUI) that teaches correct placement of accent marks on Spanish language words through a particular sequence of graphical elements includes receiving, through a GUI presented by a computing device, a selection of a graphical word element corresponding to a word to test as part of a Spanish language teaching program; identifying, by the computing device, a particular word set that corresponds to the word from among a plurality of word sets; and outputting, in the GUI presented by the computing device and based on the particular word set, a first graphical prompt element for the user to identify whether the word has a particular ending characteristic that corresponds to the particular word set.

Abstract: Scoring candidate evidence passages for criteria validation. Evidence data associated with a criteria, such that the evidence data entries include a decision indicator indicating that the criteria is either met or not met by the evidence data, is collected. Candidate evidences, making up a corpus of data associated with the criteria, against which the criteria is to be validated, are generated. Each candidate evidence is evaluated against the evidence data. A score indicating the validity of the criteria with respect to the candidate evidence is generated, based on the decision indicators associated with the evidence data entries.

Abstract: According to an embodiment, a tagging support apparatus includes a first acquirer, an estimator, a first storage, a second acquirer, and a presenter. The first acquirer acquires a spoken sentence concerning an utterance of a user. The estimator estimates an utterance intention of the spoken sentence to obtain one or more intention candidates of the utterance intention. The first storage stores an intention system having a hierarchical structure of intentions used in a dialog system. The second acquirer acquires, based on the intention candidates, part of the intention system as one or more hierarchical intention candidates. The presenter presents the hierarchical intention candidates.

Abstract: Systems and methods of natural language processing in an environment with no existing corpus are disclosed. The method includes defining an input grammar specific to a chosen domain, the input grammar having a domain specific knowledge and general grammatical knowledge. Groups of tokens are identified within the input grammar having syntactic and semantic equivalence. The identified groups are assembled into hypernyms, wherein the hypernyms include a semantic output for each token in the hypernyms. A list of fields is then combined with the hypernyms for combination with the hypernyms. A corpus of possible combinations of hypernyms and fields is created. A data structure mapping each possible combination to a partial semantic output is generated and the data structure is saved for use in later processing.

Abstract: A system, and computer program product for adapting tabular data for narration are provided in the illustrative embodiments. A set of categories used to organize data is identified in a first tabular portion of a document. A structure of the categories is analyzed. An inference is drawn about data in a first cell in the first tabular portion based on a position of the first cell in the structure. The first tabular portion of the document is transformed into a first narrative form using the inference.

Abstract: Aspects of the present disclosure include a system comprising a machine-readable storage medium storing at least one program and computer-implemented methods for detecting a language of a text string. Consistent with some embodiments, the method may include applying multiple language identification models to a text string. Each language identification model provides a predicted language of the text string and a confidence score associated with the predicted language. The method may further include weighting each associated confidence score based on historical performance of the corresponding language identification model in predicting languages of other text strings. The method may further include selecting a predicted language of the text string from among the multiple predicted languages provided by the multiple language identification models based on a result of the weighting of the confidence score associated with the particular predicted language.

Abstract: This disclosure describes a distributed reader architecture for a mobile computing device such as cellular telephone handset.

Abstract: An encryption platform may identify language characters in database fields and determine which of multiple encoding schemes more efficiently encodes the language characters. The encryption platform may use the selected encoding scheme to encode plaintext into a byte array and then encrypt the byte array into ciphertext. The encryption platform may assign encoding flags to the ciphertext identifying the encoding scheme used for encoding the plaintext. The encryption platform then may use the encoding flag to determine which encoding schemes to use for decoding the data back into plaintext. Using the different encoding schemes allows encrypted data to fit better into database storage structures. The encryption platform also may enable users to select different fields for different selectable objects for encoding and encrypting.

Abstract: An approach is provided for classifying natural language input into plurality of query processing levels. The approach involves parsing at least one natural language input into one or more words, wherein the at least one natural language input represents at least one cartographic query. The approach also involves mapping the one or more words to at least one function of a cartography system. The approach further involves determining an availability of the at least one function, one or more parameters associated with the at least one function, or a combination thereof with respect to at least one component of the cartography system. The approach also involves classifying the at least one cartographic query into at least one of a plurality of query processing levels based, at least in part, on the availability.

Abstract: Implementations herein relate to methods and devices for generating candidate translations and for quantizing text as well as words. A method may include generating, by a computing device, pending candidate translations of text to be translated based on predetermined translation rules. The computing device may generate translation probabilities from the text to be translated to the pending candidate translations based on features having impacts on translation probabilities of the pending candidate translations and a predetermined translation probability prediction model. The computing device may then select a predetermined number of pending candidate translations that have the translation probabilities higher than other pending candidate translations in the pending candidate translations to be the candidate translations of the text to be translated.

Abstract: A relay apparatus includes a characteristic information storage unit, a first dictionary information storage unit, a first selection unit, and a first conversion unit. The characteristic information storage unit stores pieces of terminal information in association with respective pieces of characteristic information, each indicating a characteristic of a user. The first dictionary information storage unit stores the pieces of characteristic information in association with respective pieces of first dictionary information used for converting voice data into text data. The first selection unit selects, when voice data is received from a second communication terminal to communicate with a first communication terminal, the characteristic information associated with the terminal information for identifying the second communication terminal and selects the first dictionary information associated with the selected characteristic information.

Abstract: Disclosed is a method of provisioning electronic forms based on natural language. The method includes receiving, using a communication device, a natural language input from a builder device, wherein the natural language input represents one or both of a requested data and a presented data associated with a legal process. Further, the method includes analyzing, using a processing device, the natural language input. Yet further, the method includes generating, using the processing device, an electronic form based on the analyzing of the natural language input, wherein the electronic form comprises an input field configured to receive the requested data and an output field configured to present the presented data. Moreover, the method includes storing, using a storage device, the electronic form. Furthermore, the method includes transmitting, using the communication device, the electronic form to one or more user devices.

Abstract: There are disclosed a system and method for providing a word search service based on message according to the present invention. The system for providing a word search service based on message according to the present invention includes a database which is equipped with a plurality of data modules and registers sentence data and word extracted from the sentence data in each data module equipped to manage the sentence data and the word; and a service server which, if receiving a search word from a user terminal, broadcasts the received search word to all data modules and is provided with a search result from each of all data modules in response to broadcasting and thereafter provides the provided search result to the user terminal.

Abstract: Examples of the present disclosure describe processing by an input understanding system/service. A received input is processed to generate a set of alternatives for recognizing the received input. The set of alternatives is filtered. Filtering comprises ranking the set of alternatives and propagating a plurality of the ranked alternatives for additional processing. The propagated alternatives are processed to generate an expanded set of alternatives for potential hypotheses based on the received input. The expanded set of alternatives is filtered. Filtering comprises ranking alternatives of the expanded set and propagating a plurality of the ranked alternatives of the expanded set for additional processing. The propagated alternatives of the expanded set are evaluated based on application of knowledge data fetched from external resources. A response to the received input is generated.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage medium, to facilitate interception of messages that include offensive content. In one aspect, a method includes actions of receiving input on a user device that includes message content, determining, on the user device, whether the message content includes offensive content, and in response to determining, on the user device, that the message content includes offensive content, generating an alert message for display on the user device that provides an indication that the message includes offensive content.

Abstract: A system with methodology for providing multi-language support for dynamic ontology. In one embodiment, for example, a method comprises: storing an ontology for a data store, wherein the ontology comprises a plurality of data types; for each data type of the plurality of data types: storing, in the ontology, one or more display values, wherein each display value is associated with a locale; determining a locale identifier corresponding to a particular locale; selecting a particular display value corresponding to the particular locale; displaying the particular display value.

Abstract: Systems and methods are provided for identifying relevant information for an entity, referred to as a seed entity. A plurality of search queries can be generated each comprising a property of a seed entity or one of the entities associated with the seed entity (seed-linked entities). Preferably, a collection of search queries includes ones representing different properties of the seed entity and properties of different seed-linked entities. Optionally, the collection of search queries is optimized to reduce search burden. Searches can then be conducted with the search queries in one or more data sources to obtain a plurality of search results, wherein each search result comprises a hit entity and one or more entities associated with the hit entity (hit-linked entity).

Abstract: An input method editor (IME) is associated with a local user. Memory stores local data and a processor, coupled to the memory, is configured to receive input from a local, first user, obtain shared data associated with at least a remote, second user from a remote server and generate prediction candidates and conversion candidates based on the input provided by the local, first user and correlation of the input and the obtained shared data.

Abstract: Technologies are described herein for cross-language speech recognition and translation. An example method of speech recognition and translation includes receiving an input utterance in a first language, the input utterance having at least one name of a named entity included therein and being pronounced in a second language, utilizing a customized language model to process at least a portion of the input utterance, and identifying the at least one name of the named entity from the input utterance utilizing a phonetic representation of the at least one name of the named entity. The phonetic representation has a pronunciation of the at least one name in the second language.

Abstract: An environmental context of a user may be taken into account to enhance the user's communication throughput. An “environmental context” can include spatial surroundings of a user, device, and/or sensor of the device and more broadly to denote the context of the user in a multiplicity of environments such as, for example, the surroundings of a user, a digital environment such as the user or other individuals' interactions with or made near a device, etc. The techniques can include obtaining contextual data to provide context-predicted suggestions of words and/or phrases that a user can select to be output on the user's behalf. In some examples, the techniques can also use contextual data to weight, sort, rank, and/or filter word and/or phrase suggestions.

Abstract: A natural language processing system has a hierarchy of user intents related to a domain of interest, the hierarchy having specific intents corresponding to leaf nodes of the hierarchy, and more general intents corresponding to ancestor nodes of the leaf nodes. The system also has a trained understanding model that can classify natural language utterances according to user intent. When the understanding model cannot determine with sufficient confidence that a natural language utterance corresponds to one of the specific intents, the natural language processing system traverses the hierarchy of intents to find a more general user intent that is related to the most applicable specific intent of the utterance and for which there is sufficient confidence. The general intent can then be used to prompt the user with questions applicable to the general intent to obtain the missing information needed for a specific intent.

Abstract: A system and method for translating data from a source language to a target language is provided wherein machine generated target translation of a source sentence is compared to a database of human generated target sentences. If a matching human generated target sentence is found, the human generated target sentence may be used instead of the machine generated sentence, since the human generated target sentence is more likely to be a well-formed sentence than the machine generated sentence. The system and method does not rely on a translation memory containing pairs of sentences in both source and target languages, and minimizes the reliance on a human translator to correct a translation generated by machine translation.

Abstract: A token-based database management system described herein may reduce an amount of comparisons during entity resolution of records. The system includes a token creator configured to create tokens from records, a token-record mapping creator configured to create a token-record mapping of tokens to records, a token importance calculator configured to calculate token importance values for the tokens, a token pruner configured to identify a token of the current record as unimportant based on token importance values of the tokens of the current record, and to remove the unimportant token from the token-record mapping, a record selector configured to select only records sharing at least one common token with the current record, and a record comparator configured to compare the current record with each of the selected records to determine whether the current record matches any of the selected records.

Abstract: A user context generator determines one or both of a location of a user and contextual information for the user. The contextual information is indicative of content of interest to the user. A custom content generator engine generates customized translated content for the user. Generating the customized translated content includes selecting, from translated content stored in a database, based on the one or both of the determined location of the user and the determined contextual information for the user, translated content to be presented to the user. The customized translated content includes a set of phrases in a source language and corresponding translations of phrases, in the set of phrases, from the source language to a target language. The selected translated content is displayed to the user, such that the user is provided with translated content of interest to the user.

Abstract: A method (300) and system (100) is provided to add the creation of examples at a developer level in the generation of Natural Language Understanding (NLU) models, tying the examples into a NLU sentence database (130), automatically validating (310) a correct outcome of using the examples, and automatically resolving (316) problems the user has using the examples. The method (300) can convey examples of what a caller can say to a Natural Language Understanding (NLU) application. The method includes entering at least one example associated with an existing routing destination, and ensuring an NLU model correctly interprets the example unambiguously for correctly routing a call to the routing destination. The method can include presenting the example sentence in a help message (126) within an NLU dialogue as an example of what a caller can say for connecting the caller to a desired routing destination.