Abstract: A system for reinforcement learning based controlled natural language generation is disclosed. The system includes a token generator subsystem to generate an initial output phrase including a sequence of output tokens. The system includes trained models associated with corresponding predefined tasks. Each trained model includes an attention layer to compute attention-based weights for each output token. The trained models include a scoring layer to generate a phrase sequence level score for the output phrase. The trained models include a reward generation layer to generate dense rewards for each output token based on the attention-based weights and the phrase sequence level score. The trained models include a feedback score generation layer to generate a feedback score based on the dense rewards and reward weights assigned to the dense rewards of the corresponding trained models. The feedback score generation layer provides the feedback score iteratively to the token generator subsystem.

Abstract: A method is disclosed. The method includes receiving a text phrase from a user and parsing the text phrase using a natural language parser to generate distinct words. A vector of values relating to the distinct words is generated and compared to vectors in a graph database which relate to learned communities of language. Based on the comparing, a most similar community in the learned communities of language is determined and an action set action set associated with the most similar community is queried. A response to the text phrase from the user is then generated based on the action set and provided to the user.

Abstract: Systems and methods for conversational user experiences and conversational database analysis disclosed herein improve the efficiency and accessibility of low-latency database analytics.

Abstract: Briefly stated, the invention is directed to retrieving a semantically matched knowledge structure. A question and answer pair is received, wherein the answer is received from a query of a search engine. A question is constraint-matched with the answer based on maximizing a plurality of constraints, wherein at least one of the plurality of the constraints is a similarity score between question and answer, wherein the constraint matching generates a matched sequence. For one or more answer sequences, a subsequence is found that are not parsed as answer slots. Query results are obtained from another search engine based on a combination of the answer or question, and the non-answer subsequence. And a KB based is refined on the query results and the constraint matching and based on a neural network training, for a further subsequent semantic matching, wherein the KB includes a dense semantic vector indication of concepts.

Abstract: The technology disclosed provides a so-called “pointer sentinel mixture architecture” for neural network sequence models that has the ability to either reproduce a token from a recent context or produce a token from a predefined vocabulary. In one implementation, a pointer sentinel-LSTM architecture achieves state of the art language modeling performance of 70.9 perplexity on the Penn Treebank dataset, while using far fewer parameters than a standard softmax LSTM.

Abstract: Due to the high language use variability in real-life, manual construction of semantic resources to cover all synonyms is prohibitively expensive and may result in limited coverage. Described herein are systems and methods that automate the process of synonymy resource development, including both formal entities and noisy descriptions from end-users. Embodiments of a multi-task model with hierarchical task relationship are presented that learn more representative entity/term embeddings and apply them to synonym prediction. In model embodiments, a skip-gram word embedding model is extended by introducing an auxiliary task “neighboring word/term semantic type prediction” and hierarchically organize them based on the task complexity. In one or more embodiments, existing term-term synonymous knowledge is integrated into the word embedding learning framework.

Abstract: Systems, devices, and methods of the present invention create a searchable index that includes informative portions of text. In an example, a computer-implemented method creates a discourse tree from a body of text. For each non-terminal node in the discourse tree, the method identifies a rhetorical relationship associated with the non-terminal node. The method labels each terminal node associated with the non-terminal node as either a nucleus or a satellite. The method further accesses a rule associated with the rhetorical relationship, and selects, based on the rule, selects the fragment associated with the nucleus. The method creates a searchable index including the selected fragments.

Abstract: A natural language model can be primed utilizing optimized examples generated from a labeled knowledge graph corresponding to an independently developed application program. Parsing of the labeled knowledge graph can include the identification of triples, comprising a source node, a destination node, and a link between them, each of which can be labeled. One or more natural language input examples can be generated from an individual triple by concatenating the natural language words or phrases utilized to label the source node in the link. Determinations that subsequently received natural language user input is similar to the generated examples can result in an identification of the triple, which can, in turn, trigger the performance of a function associated with the destination node of the triple. Labels can include preferred labels and alternative labels, and various permutations thereof can be concatenated to generate alternative natural language input examples.

Abstract: The present invention provides a biometric notification system for selectively sending messages to interested recipients. In various embodiments, message trigger criteria, interested recipients, and message content may vary depending upon, among other things, the service being provided.

Abstract: An example method comprises: receiving a natural language text; transforming, by a neural network, the natural language text into a numeric representation comprising a plurality of numeric values; discarding the natural language text; and performing, using the numeric representation of the natural language text, an information extraction task.

Abstract: A method includes accessing, by a healthcare practitioner via an EHR software application loaded on an electronic device, patient information for a patient; determining, by an agent application loaded on the electronic device, the patient for which the patient information was accessed; communicating, from the agent application to an agent service, a request for data regarding the patient; communicating, from the agent service to the agent application, accessed community data for the patient; displaying, in an interface of the agent application, one or more data items based on the accessed community data; receiving user input corresponding to toggling of the agent application from a filtered mode to an unfiltered mode; in response to the toggling, displaying additional data items coded visual indications that the respective data items are newly displayed, the coded visual indications being coded to indicate a reason why each respective data item was not previously displayed.

Abstract: In some embodiments, a content analysis system accesses input content associated with a user of an online platform. The content analysis system extracts entity tags for entities contained in the input content and links the identities to standard entities in a knowledge base to generate linked entities. The content analysis system further generates a knowledge graph to include the linked entities and other standard entities connected to the linked entities as nodes and edges connecting these nodes. Based on the knowledge graph, the content analysis system identifies related entities that are similar to the linked entities and cause the online platform to be modified based on the related entities.

Abstract: System and methods of creating and using a transparent, computable contractual natural language are disclosed in which a set of legal contracts are text mined to obtain a structured contractual database. A set of categorized contractual phrases are assembled from the structured contractual database. A transparent knowledge representation language is defined having a set of syntax rules, a set of semantic rules and a set of inference rules. A transparent, computable contractual natural language is the set of contractual phrases that map to the transparent knowledge representation language. A user writes computable legal documents comprised of phrases contained in the transparent, computable contractual natural language.

Abstract: A method includes receiving, by a question-answer system, a question in a first language and the question in a second language and predicting, by the question-answer system, a first answer to the question in the first language and a second answer to the question in the second language. The method also includes generating, by the question-answer system, a first vector representing the question in the first language and a second vector representing the question in the second language and adjusting the question-answer system based on the first and second answers and the first and second vectors such that when the question-answer system subsequently generates a third vector representing the question in the first language and a fourth vector representing the question in the second language, a distance between the third and fourth vectors is less than a distance between the first and second vectors.

Abstract: An in-cloud wake-up method and system, a terminal and a non-volatile computer-readable storage medium are provided. The in-cloud wake-up method includes: receiving wake-up voice input by a user, and transmitting the wake-up voice to cloud; performing an decoding operation on the wake-up voice in cloud to generate a wake-up text; identifying wake-up characters in the wake-up text; and providing a wake-up instruction according to an identification result. With the in-cloud wake-up method, voice can be identified in cloud, and a smart device can be waken up in cloud according to the identified voice, so that resource consumption and power loss of the terminal device are reduced in the case of ensuring the quality of wake-up.

Abstract: Systems and methods identify and provide interesting facts about an entity. An example method includes selecting documents associated with at least one unique fact trigger, the documents being from a document repository. The method also includes generating entity-sentence pairs from the documents and, for a first entity of the entities represented by the entity-sentence pairs, clustering the entity-sentence pairs for the first entity using salient terms occurring in the sentence. The method also includes determining a representative sentence for each of the clusters and providing at least one of the representative sentences in response to a query that identifies the first entity. Another example method includes determining that a query relates to an entity in a knowledge base, determining that the entity has an associated unique fact list, and providing at least one of the unique facts in the list in response to the query.

Abstract: Artificial intelligence (AI) technology can be used in combination with composable communication goal statements to facilitate a user's ability to quickly structure story outlines using “explanation” communication goals in a manner usable by an NLG narrative generation system without any need for the user to directly author computer code. This AI technology permits NLG systems to determine the appropriate content for inclusion in a narrative story about a data set in a manner that will satisfy a desired explanation communication goal such that the narratives will express various ideas that are deemed relevant to a given explanation communication goal.

Abstract: Systems, methods, and apparatuses for contextual natural language understanding are detailed. An exemplary method includes receiving a user utterance provided by a user within a multi-turn chat dialog between the user and a conversational agent; providing to a contextual natural language understanding framework: the user utterance, and contextual information associated with one or more previous turns of the multi-turn chat dialog, the contextual information associated with each turn of the one or more previous turns including a previous intent, a previous dialog act, and an elicited slot; and obtaining, from the contextual natural language understanding framework, an intent classification and one or more slot labels.

Abstract: Tools, strategies, and techniques are provided for evaluating the identities of different entities to protect individual consumers, business enterprises, and other organizations from identity theft and fraud. Risks associated with various entities can be analyzed and assessed based on analysis of social network data, professional network data, or other networking connections, among other data sources. In various embodiments, the risk assessment may include calculating an authenticity score based on the collected network data.

Abstract: A natural language processing method, comprising: receiving multiple input words; and reducing the multiple words into one or more subject word data structures according to sets stored in a database, wherein one of the subject word data structures includes a first input word and a second input words among the input words, wherein one of the sets includes a compatible relation between the first input word and the second input word, wherein the compatible relation between the first input word and the second word includes a compatible property for denoting an intensity representing occurrences of the first input word and the second word in a training corpus.

Abstract: Provided are systems and methods for mapping customer feedback to one or more processes. In one example, the method may include receiving text content that includes feedback, mapping, via execution of a mapping algorithm, the text content to a plurality of processes based on a synonym graph, generating mapping values for the plurality of processes, where a mapping value for a process is generated based on distance within the synonym graph of a mapping between a word in the text content and a word identifying the process, and determining a process among the plurality of processes that is most correlated to the feedback based on the generated mapping values.

Abstract: A mechanism is provided for implementing a dynamic context-based collaborative medical concept interpreter for automatically generating and presenting summarized explanations of medical concepts. The dynamic context-based collaborative medical concept interpreter performs natural language processing on a real-time patient-provider communication to identify one or more medical concepts referred to in the communication. The dynamic context-based collaborative medical concept interpreter adjusts one or more previous explanations for the one or more medical concepts referred to in the communication using a set of contextual factors. The dynamic context-based collaborative medical concept interpreter generates an abstractive summary that summarizes ranked explanations of the one or more medical concepts based on an original language used in the one or more previous explanations.

Abstract: Systems, methods, devices, instructions, and other examples are described for natural language processing. One example includes accessing natural language processing general encoder data, where the encoder data is generated from a general-domain dataset that is not domain specific. A domain specific dataset is accessed and filtered encoder data using a subset of the encoder data is generated. The filtered encoder data is trained using the domain specific dataset to generate distilled encoder data, and tuning values for the distilled encoder data are generated to configure task outputs associated with the domain specific dataset.

Abstract: Embodiments relate to an intelligent computer platform to identify a lexical answer type (LAT), a first concept relevant to the received request and a second concept related to the identified first concept. The LAT, together with the first and second concepts are utilized to create a first and second cluster. Documents are selectively populated into the clusters. The clusters are subject to sorting based on a relevancy protocol.

Abstract: This disclosure covers methods, systems, and computer-readable media that select answer choices from potential answer choices for a digital question based on responses to other digital questions and/or embedded user data. In certain embodiments, the disclosed systems select answer choices from potential answer choices for a digital question based on a multiple choice response. Furthermore, in some embodiments, the disclosed systems select answer choices from potential answer choices for a digital question based on keywords and/or sentiment values identified by analyzing a text response. In some embodiments, the disclosed systems select answer choices for a digital question from a dynamic choice reference dataset that comprises potential answer choices. Additionally, in one or more embodiments, the disclosed systems train and/or utilize a machine-learning model to select answer choices from potential answer choices for a digital question based on a response.

Abstract: Methods, systems, and devices supporting querying disparate data sources are described. Querying disparate data sources may include receiving an input for data stored at a first data source from a plurality of data sources, selecting a first data connector from a plurality of data connectors, wherein the first data connector corresponds to the first data source, and identifying a first query language corresponding to the first data source from a plurality of query languages. Querying the disparate data sources may further include generating a converted query based at least in part on the first query language and retrieving the data from the first data source using the first data connector based at least in part on the converted query.

Abstract: Artificial intelligence (AI) technology can be used in combination with composable communication goal statements to facilitate a user's ability to quickly structure story outlines in a manner usable by an NLG narrative generation system without any need for the user to directly author computer code. Narrative analytics that are linked to communication goal statements can employ a conditional outcome framework that allows the content and structure of resulting narratives to intelligently adapt as a function of the nature of the data under consideration. This AI technology permits NLG systems to determine the appropriate content for inclusion in a narrative story about a data set in a manner that will satisfy a desired communication goal.

Abstract: A system retrains a natural language understanding (NLU) model by regularly analyzing electronic documents including web publications such as online newspapers, blogs, social media posts, etc. to understand how word and phrase usage is evolving. Generally, the system determines the frequency of words and phrases in the electronic documents and updates an NLU dictionary depending on whether certain words or phrases are being used more frequently or less frequently. This dictionary is then used to retrain the NLU model, which is then applied to predict the meaning of text or speech communicated by a people group. By analyzing electronic documents such as web publications, the system is able to stay up-to-date on the vocabulary of the people group and make correct predictions as the vocabulary changes (e.g., due to natural disaster). In this manner, the safety of the people is improved.

Abstract: A machine learning based model generates a feature representation of a text sequence, for example, a natural language sentence or phrase. The system trains the machine learning based model by receiving an input text sequence and perturbing the input text sequence by masking a subset of tokens. The machine learning based model is used to predict the masked tokens. A predicted text sequence is generated based on the predictions of the masked tokens. The system processes the predicted text sequence using the machine learning based model to determine whether a token was predicted or an original token. The parameters of the machine learning based model are adjusted to minimize an aggregate loss based on prediction of the correct word for a masked token and a classification of a word as original or replaced.

Abstract: An approach is described for addressing propagation of inaccurate information in a social networking environment. An associated method may include identifying inaccurate information within the social networking environment, facilitating creation of countering content to address the inaccurate information, and disseminating the countering content. The countering content may be determined by identifying behavior of one or more users among a plurality of users within the social networking environment. Identifying the inaccurate information within the social networking environment may include receiving information provided within the social networking environment. Upon determining that the received information is factual and thus objectively verifiable, it may be determined whether the received information matches analogous information verified as accurate.

Abstract: Provided are a computer program product, system, and method for using a machine learning module to rank technical solutions to user described technical problems to provide to a user. A solutions database is searched to determine a result set of answer files having keywords related to keywords of a technical problem in a user description. The result set of answer files, the user description of the technical problem, user profile information, and technical specifications of the computing system are provided as input to a machine learning module. The machine learning module outputs ranks of the answer files in the result set. A subset of answer files in the result set is selected based on ranks of the answer files in the result set and returned to the user.

Abstract: A sentiment-based score is determined for a contact center engagement between a first contact center service operator and a contact center user. The sentiment-based score is indicated within a graphical user interface displaying information associated with multiple contact center engagements at a device of a second contact center service operator. Based on a request to participate in the contact center engagement received from the device of the second contact center service operator via the graphical user interface, information associated with the contact center engagement is transmitted to the device of the second contact center service operator, and a contact center session involving a device of the contact center user and the device of the second contact center service operator is established.

Abstract: Apparatus for generating a putative ontology from a data structure associated with a data store, the apparatus including an electronic processing device that generates a putative ontology by determining at least one concept table in the data structure, determining at least one validated attribute within the at least one concept table, determining at least one selected attribute value from the at least one validated attribute and generating at least one ontology class using the at least one attribute value.

Abstract: Embodiments of the present disclosure relate to a dialogue interaction method and apparatus, a device and a storage medium, and relate to the field of artificial intelligence technology. The method may include: determining a first semantic encoding of received user information according to a sentence tree; determining a second semantic encoding for responding to the user information from a dialogue tree according to the first semantic encoding, the sentence tree and the dialogue tree being trained and obtained through sentence node information and/or word node information in a logical brain map sample; and determining a target response sentence of the second semantic encoding from the sentence tree, to be used for a dialogue with a user.

Abstract: An apparatus generates an index including positions of morphemes included in a target text data and semantic attributes between the morphemes corresponding to the positions. The apparatus gives information including positions of morphemes included in an input query and semantic attributes between the morphemes corresponding to the positions to the query, and executes a retrieval on the target text data, based on the information given to the query and the index.

Abstract: A method for training a language model based on various word vectors, a device and a medium, which relate to the field of natural language processing technologies in artificial intelligence, are disclosed. An implementation includes inputting a first sample text language material including a first word mask into the language model, and outputting a context vector of the first word mask via the language model; acquiring a first probability distribution matrix of the first word mask based on the context vector of the first word mask and a first word vector parameter matrix, and a second probability distribution matrix of the first word mask based on the context vector of the first word mask and a second word vector parameter matrix; and training the language model based on a word vector corresponding to the first word mask.

Abstract: Estimation accuracies of a conversation satisfaction and a speech satisfaction are improved. A learning data storage unit (10) stores learning data including a conversation voice containing a conversation including a plurality of speeches, a correct answer value of a conversation satisfaction for the conversation, and a correct answer value of a speech satisfaction for each speech included in the conversation.

Abstract: Disclosed are a computer implemented method, system and platform for improving the readability and/or coherency of a conversation transcript, which include the applying of a speech disfluency detection model to identify speech disfluencies in a text transcript and to provide a corrected and/or annotated version of the conversation transcript indicating the edits made vis-à-vis the inputted text transcript.

Abstract: Systems, devices, and methods discussed herein provide improved autonomous agent applications that are configured to provide explanations in response to user-submitted questions. Training data comprising a question, and an explanation pair may be accessed. A discourse tree and an explanation chain can be constructed from the explanation. The explanation chain may identify logical relationships between two entities of elementary discourse units identified from the discourse tree. A query may be submitted for the two entities, and a set of search results can be mined to identify text linking the two entities. An additional discourse tree can be generated from the text of a search result. The additional discourse tree can be combined with the original discourse tree to generate a complete discourse tree. A model may be trained using this augmented data (e.g., the complete discourse tree) to improve the quality of explanations provided by the autonomous agent application.

Abstract: Unstructured text is identified as larger than a threshold size. Named-entity recognition analysis is executed on the unstructured text. One or more anchor entities of the unstructured text are determined that each occur more than a threshold amount of times within the unstructured text. Two or more instances of the one or more anchor entities that are separated by at least a threshold amount of text of the unstructured text are identified. The unstructured text is partitioned into at least three sections. The unstructured text is partitioned at respective natural language demarcation points associated with each of the two or more instances such that each of the at least three sections is smaller than the threshold size. Separate coreference analyses are performed in parallel on each of the at least three sections.

Abstract: A text processing method includes: word segmentation is performed on a first text to be processed to determine multiple words of the first text; one word is removed from the first text to obtain one second text; each word is removed from the first text one by one to obtain a second text set; the first text and each second text in the second text set are analyzed respectively to obtain a first tree structure of the first text and a second tree structure of each second text; and a summary text of the first text is determined according to the first tree structure and each second tree structure.

Abstract: A method of automatically generating content summaries for topics includes receiving a taxonomy for a concept and a text corpus. The method further includes generating an annotated dataset having term annotations corresponding to the concept from the text corpus based on the taxonomy, parsing the annotated dataset into a custom generated document object having a structured layout, determining features for the term annotations, and extracting snippets from the custom generated document object, where each of the snippets corresponds to a section of the custom generated document object. The method further includes scoring the snippets based on the features such that each of the snippets corresponds to a score, filtering one or more snippets from the snippets when one or more snippet filtering conditions is met, ranking the snippets into an ordered list for the concept based on the score, and providing, to a user computing device, the ordered list.

Abstract: A system and method for implementing a machine learning-based virtual dialogue agent includes computing an input embedding based on receiving a user input; computing, via a pre-trained machine learning language model, an embedding response inference based on the input embedding; searching, based on the embedding response inference, a response imprintation embedding space that includes a plurality of distinct embedding representations of potential text-based responses to the user input, wherein each of the plurality of distinct embedding representations is tethered to a distinct human-imprinted media response, and searching the response imprintation embedding space includes: searching the response imprintation embedding space based on an embedding search query, and returning a target embedding representation from the response imprintation embedding space based on the searching of the response imprintation embedding space; and executing, via a user interface of the machine learning-based virtual dialogue agent,

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-readable storage media, for providing information cards using semantic graph data. In some implementations, semantic graph data for a semantic graph is stored, where the semantic graph data indicates objects and relationships among the objects, and the objects include a card object that represents characteristics of an information card. A request is received from a client device, and the request is processed using the semantic graph data. Data for the information card is provided to the client device based on the card object indicated by the semantic graph data.

Abstract: There is provided a method and apparatus for generating a competition commentary based on artificial intelligence, and a storage medium. The method comprises: obtaining commentator's words commentaries and structured data of historical competitions; generating a commentating model according to obtained information; during live broadcast of a competition, determining a corresponding words commentary according to the commentating model with respect to the structured data obtained each time.

Abstract: An information processing system comprising a processor programmed to: receive a question asked by a questioner, an answer provided by an answerer to the question, and a rating by a rater with respect to at least one of the question and the answer; manage relationship information, the relationship information being related to the questioner, the answerer, and the rating by the rater; acquire attribute information about each of the questioner, the answerer, and the rater; and present rating information based on the relationship information and in response to a condition specified by a requester with respect to the attribute information.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for optimized graph traversal are disclosed. In one aspect, a method includes the actions of receiving a given phrase that is input through a user interface by a digital component provider. The actions further include determining an entity that is being referred to by the given phrase. The actions further include identifying properties of the entity. The actions further include selecting a subset of the properties that were identified for the entity. The actions further include identifying additional phrases. The actions further include updating the user interface to present at least some of the additional phrases with programmatic controls that assign one or more of the additional phrase as distribution criteria for digital components of the digital component provider in response to activation of the programmatic controls.

Abstract: This disclosure describes converting computer-executable predicate-argument structures for a specific field to field-specific predicated-argument structures to improve execution. In some implementations, a method can be performed by one or more processors of a computing device, and can include receiving one or more predicate-argument structures (PASs) associated with taxation-specific text and converting the one or more PASs into one or more tax-specific predicate-argument structures (TPASs). Converting the one or more PASs to one or more TPASs may include one or more of: defining terms in a segment based on a definition of the term from a different segment or line description (including from a different document); reordering nodes, replacing nodes, or removing nodes of a segment (such as based on one or more single segment tree traversal rules); or combining multiple PASs for multiple segments of a single line description based on one or more multiple segment tree traversal rules.

Abstract: In one embodiment, a method includes receiving a user request from a client system associated with a first user, wherein the user request is associated with a semantic-intent, identifying one or more dialog-intents associated with the user request based on the semantic-intent and context information associated with the user request, wherein each dialog-intent is a sub-intent of the semantic-intent, determining one or more agents for executing one or more tasks associated with the one or more dialog-intents, and sending instructions for presenting information returned from the one or more agents responsive to executing the one or more tasks to the client system.

Abstract: The present specification provides a method and apparatus for determining feature words and a server. The method includes: obtaining text data; extracting a first feature word from the text data; updating a word segmentation library based on the first feature word to obtain an updated word segmentation library, the word segmentation library including a plurality of predetermined feature words for representing predetermined attribute types; and extracting a second feature word from the text data based on the updated word segmentation library and the predetermined feature words.