Abstract: Generating automated conversation responses by receiving a conversation input message, determining an intent associated with the conversation input message, detecting content associated with the intent in a data stream in response to determining the intent, and generating a conversation output according to the content and the intent.

Abstract: Methods and systems provide for dynamically generated topic segments for a communication session. In one embodiment, the system connects to a communication session with a number of participants; receives a list of topics; receives a transcript of a conversation between the participants produced during the communication session, the transcript including timestamps for a number of utterances associated with speaking participants; for each topic in the list of topics, segments the utterances into one or more topic segments based on the topic; for each of the segments, classifies whether the topic segment is related to the topic, and transmits, to one or more client devices, a list of the topic segments for the communication session.

Abstract: A method for redacting sensitive information from an audio stream, such as a voice signal in a telephone call, in real time is provided. The method includes: receiving an audio stream; conveying the audio stream through a channel that includes a valve; detecting, from within the audio stream, a first event that indicates an onset of sensitive information; closing the valve so that the conveying of the audio stream through the channel is temporarily stopped; detecting, from within the audio stream, a second event that indicates an ending of the sensitive information; and reopening the valve so that the conveying of the audio stream through the channel is resumed. The sensitive information may include payment card industry (PCI) information, such as a card number and/or a card verification value (CVV).

Abstract: This document relates to machine learning. One example includes a method or technique that can be performed on a computing device. The method or technique can include obtaining a task-semantically-conditioned generative model that has been pretrained based at least on a first training data set having unlabeled training examples and semantically conditioned based at least on a second training data set having dialog act-labeled utterances. The method or technique can also include inputting dialog acts into the semantically-conditioned generative model and obtaining synthetic utterances that are output by the semantically-conditioned generative model. The method or technique can also include outputting the synthetic utterances.

Abstract: A system and method are disclosed that enable rapid and cost-effective human-in-the-loop synthesis of domain-specific textual training data for a deep learning model. The data augmentation process incorporates a sentence generator, a sentence classifier, and weak-supervision by a domain expert that is ‘in the loop.’ Generally, both the sentence generator and the sentence classifier are implemented as machine learning models. The sentence generator generates new sentences based on manually labeled sentences and the sentence classifier generates labels for the newly generated sentences. The new sentences are corrected or verified by a domain expert and then used to retrain one or both of the sentence generator and the sentence classifier.

Abstract: Methods, systems, and apparatuses for improved speech recognition and transcription of user utterances are described herein. A user utterance may be processed by a speech recognition computing device. One or more acoustic features associated with the user utterance may be used to determine whether one or more actions are to be performed based on a transcription of the user utterance.

Abstract: A method, performed by at least one processor, and an apparatus for cross-lingual text-to-SQL semantic parsing is provided. The method and computer program code performed by the at least one processor include: generating a contextual representation of a source language utterances, a target language utterances, and a database schema; generating a mixed representation of the target language utterances and the database schema based on the contextual representation of the source language utterances, the target language utterances, and the database schema; concatenating the mixed representation of the target language utterances and the database schema; encoding the concatenated mixed representation of the target language utterances and the database schema, based on k-layer transformers; and generating SQL queries token-by-token based on the encoded concatenated mixed representation of the target language utterances and the database schema.

Abstract: The disclosure relates to an artificial intelligence (AI) system that simulates functions such as cognition and judgment of the human brain by utilizing machine learning algorithms such as deep learning and its applications. A method in which a device provides translation information of an application includes obtaining a resource file of an application through an operating system (OS) of the device; translating text to be displayed on an execution screen of the application from the obtained resource file by using an artificial intelligence model; generating the execution screen of the application by using the resource file and the translated text; and displaying the generated execution screen, wherein the translating includes translating the text of a first language into a second language by applying at least a part of data in the resource file to the artificial intelligence model trained to translate the text to be displayed on the execution screen of the application.

Abstract: The present disclosure discloses a method for generating a styled sentence by a computer device. The method includes: obtaining a to-be-converted natural sentence, classifying, by inputting the natural sentence into a first encoding model, having a classification capability of classifying the natural sentence into a target content vector and a style vector of the natural sentence, the target content vector indicating a meaning of the natural sentence, and the style vector of the natural sentence indicating a language style of the natural sentence. The method also include determining, from at least one style vector according to a set target language style, a target style vector corresponding to the target language style; and inputting the target content vector and the target style vector into a first decoding model, and generating a styled sentence corresponding to the natural sentence.

Abstract: A toolbox application presents a user interface on a client device, and receives a natural language search query provided by a user of the client device that identifies a desired functionality to be performed for the user. The toolbox application determines whether any of a plurality of accessible utility components accessible is capable of performing the desired functionality specified in the search query. Responsive to determining that none of the accessible utility components is capable of performing the desired functionality, the toolbox application sends a request comprising the natural language search query to a server that maintains library of a plurality of utility components.

Abstract: An apparatus and method for generating an article, the apparatus including at least a processor and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to receive a content category, receive a data file comprising content data, import the content data from the data file, generate an article as a function of the imported content data and content category, and export the article.

Abstract: Techniques are provided to validate a digitized audio signal that is generated by a conference participant. Reference speech features of the conference participant are obtained, either via samples provided explicitly by the participant, or collected passively via prior conferences. The speech features include one or more of word choices, filler words, common grammatical errors, idioms, common phrases, pace of speech, or other features. The reference speech features are compared to features observed in the digitized audio signal. If the reference speech features are sufficiently similar to the observed speech features, the digitized audio signal is validated and the conference participant is allowed to remain in the conference. If the validation is not successful, a variety of possible actions are taken, including alerting an administrator and/or terminating the participant's attendance in the conference.

Abstract: Techniques for providing and implementing a single skill associated with custom functionality and system-provided functionality are described. The skill may be used to invoke functionality in response to a user input without requiring a user remember exact formulations to cause the functionality to be performed. The skill may be associated with more than one domain. For example, the skill may be associated with custom sample user inputs (corresponding to the custom functionality) that correspond to a custom domain while the skill may also be associated with system-provided sample user inputs (corresponding to the system-provided functionality) associated with a non-custom domain.

Abstract: The disclosed technology is generally directed to chatbot conversation management. In one example of the technology, a generic model associated with a first property associated with a chatbot manager is received. The generic model is generated based on machine learning. A refined model that is associated with the first property is generated based on the generic model and a first plurality of phrases. A first conversation is held between a chatbot and an end user. Upon determining that the value of the first property for the end user is not stored in the database, via the chatbot, during the first conversation, the refined model is used to dynamically ask questions to the end user to determine a value of the first property for the end user. In a second conversation between the chatbot and the end user, the value of the first property for the end user is referenced.

Abstract: In an embodiment, a textual description of a situation of a first user is received. A first set of vector embeddings is determined based on the textual description. A set of ethical texts is received based on an input from a second user. A second set of vector embeddings is determined based on the set of ethical texts. A set of antonym words and a set of synonym words are determined with respect to the first set of vector embeddings, based on the second set of vector embeddings. A set of sentences is determined based on the set of antonym words and the set of synonym words. A first sentence is selected from the set of sentences based on parts-of-speech in each sentence. By using a GAN model, an image is generated based on the first sentence. The image is rendered on a display device associated with the second user.

Abstract: A computer-implemented method, computer program product and computing system for normalizing descriptors within a master website dataset; comparing descriptors within a subordinate website dataset to descriptors within the master website dataset to define a similarity score for each descriptor within subordinate website dataset; and normalizing one or more descriptors of the subordinate website dataset if the similarity score is above a similarity threshold.

Abstract: Some embodiments may obtain a natural language question, determine a context of the natural language question, and generate a first vector based on the natural language question using encoder neural network layers. Some embodiments may access a data table comprising column names, generate vectors based on the column names, and determine attention scores based on the vectors. Some embodiments may update the vectors based on the attention scores, generating a second vector based on the natural language question, determine a set of strings comprising a name of the column names and a database language operator based on the vectors. Some embodiments may determine a values based on the determined database language operator, the name, using a transformer neural network model. Some embodiments may generate a query based on the set of strings and the values.

Abstract: In one embodiment, a method includes receiving a user request to automatically debug a natural-language understanding (NLU) model, accessing a plurality of predicted semantic representations generated by the NLU model, wherein the plurality of predicted semantic representations are associated with a plurality of dialog sessions, respectively, wherein each dialog session is between a user and an assistant xbot associated with the NLU model, generating a plurality of expected semantic representations associated with the plurality of dialog sessions based on an auto-correction model, wherein the auto-correction model is learned from dialog training samples generated based on active learning, identifying incorrect semantic representations of the predicted semantic representations based on a comparison between the predicted semantic representations and the expected semantic representations, and automatically correcting the incorrect semantic representations by replacing them with respective expected semantic repre

Abstract: A computer-implemented technique for characterizing lexical concreteness in narrative includes receiving data encapsulating narrative text having a plurality of words. Thereafter, the function words can be removed from the narrative text to result in only content words. A concreteness score can then be assigned to each content word by polling a database to identify matching words and to use concreteness scores associated with such matching words as specified by the database. Data can then be provided which characterizes the assigned concreteness scores. Related apparatus, systems, techniques and articles are also described.

Abstract: An event detection system is configured to access a repository that contains a collection of media content. The media content may for example include images, videos, audio clips, and the like, wherein the media content comprises features that include: tags (e.g., hashtags or other similar mechanisms to label and sort content); captions that comprises one or more words or phrases; continuous numerical values; geolocation data (e.g., geo-hash, check-in data, coordinates); as well as temporal data (e.g., timestamps).

Abstract: Provided is a word processing system which includes: a first generation unit which generates, based on sentence information including a plurality of sentences, hierarchy data indicating syntax tree data for each hierarchy with regard to each sentence; a second generation unit which acquires, from a plurality of hierarchy data generated by the first generation unit, hierarchy data of a second sentence similar to hierarchy data of a first sentence generated by the first generation unit, extracts a difference between the hierarchy data of the first sentence and the hierarchy data of the second sentence, and generates, as paraphrasing rule data, first expression data as a difference in the first sentence and second expression data as a difference in the second sentence; and a storage unit which stores the paraphrasing rule data generated by the second generation unit in a storage unit.

Abstract: In an embodiment, a programmed computer system implemented via client-server Software as a Service (SaaS) techniques provides an interactive user interface for identifying specific portions of a digital document susceptible for review and improvement. A server computer may receive a representation of a digital document, such as an email, comprising words arranged into sentences arranged into paragraphs. An embodiment may tokenize a set of all sentences comprising the sequence of sentences into a document-specific vocabulary, then compute a corresponding first and second score for each sentence of the sequence of sentences. The first score may represent a calculated probability of semantic importance of the corresponding sentence to an overall meaning of the digital document. The second score may represent a calculated likelihood that the corresponding sentence will be read by a future reader of the digital document. An embodiment may identify key sentences using the first scores and second scores.

Abstract: Systems and methods are provided for processing a group of essays to develop a classifier that detects nonsensical computer-generated essays. A data structure associated with a group of essays is accessed, wherein the group of essays includes nonsensical computer-generated essays and good-faith essays. Both the nonsensical computer-generated essays and the good-faith essays are assigned feature values. The distribution of feature values between the nonsensical computer-generated essays and the good-faith essays is measured. A classifier that detects whether an essay is a nonsensical computer-generated essay is developed, wherein the classifier is developed using the distribution of feature values.

Abstract: Systems and methods for generating strings based on a seed string are disclosed. Machine learning models are trained using domain-specific training data. Random walk models are derived from the trained machine learning models. A seed string is input into each of the random walk models, and each of the random walk models iteratively generate one or more next characters for the seed string to generate at least one term from each of the random walk models. A predicted class for the at least one term generated by each of the random walk models can be determined, and a ranked order for the at least one term generated by each of the random walk models with the predicted classes can be output to a graphical user interface.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating document vector representations. One of the methods includes obtaining a new document; and determining a vector representation for the new document using a trained neural network system, wherein the trained neural network system has been trained to receive an input document and a sequence of words from the input document and to generate a respective word score for each word in a set of words, wherein each of the respective word scores represents a predicted likelihood that the corresponding word follows a last word in the sequence in the input document, and wherein determining the vector representation for the new document using the trained neural network system comprises iteratively providing each of the plurality of sequences of words to the trained neural network system to determine the vector representation for the new document using gradient descent.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media that can leverage a natural language model to determine a most probable candidate sequence of tokens and thereby generate a predicted user activity. In particular, the disclosed systems can tokenize activity event vectors to generate a series of sequential tokens that correspond to recent user activity of one or more user accounts. In addition, the disclosed systems can, for each candidate (e.g., hypothetical) user activity, augment the series of sequential tokens to include a corresponding token. Based on respective probability scores for each of the augmented series of sequential tokens, the disclosed systems can identify as the predicted user activity, a candidate user activity corresponding to one of the augmented series of sequential tokens associated with a highest probability score. Based on the predicted user activity, the disclosed systems can surface one or more suggestions to a client device.

Abstract: A device may receive text data, audio data, and video data associated with a user, and may process the received data, with a first model, to determine a stress level of the user. The device may process the received data, with second models, to determine depression levels of the user, and may combine the depression levels to identify an overall depression level. The device may process the received data, with a third model, to determine a continuous affect prediction, and may process the received data, with a fourth model, to determine an emotion of the user. The device may process the received data, with a fifth model, to determine a response to the user, and may utilize a sixth model to determine a context for the response. The device may utilize seventh models to generate contextual conversation data, and may perform actions based on the contextual conversational data.

Abstract: Techniques for table row identification using machine learning are disclosed herein. For example, a method can include detecting a table body in a document by processing the document using a machine learning (ML)-based table body model; predicting an initial table row index for one or more words among a plurality of words obtained in the document, wherein the one or more words are determined to be within the table body; and determining a table row index for the one or more words using an ML-based table row model that is trained based on the predicted initial table row index for the one or more words.

Abstract: A technique making use of a few-shot model to determine whether a query text content belongs to a same language as a small set of examples, or alternatively provide a next member in the same language to the small set of examples. The related few-shot model makes use of convolutional models that are trained in a “learning-to-learn” fashion such that the models know how to evaluate few-shots that belong to the same language. The term “language” in this usage is broader than spoken languages (e.g., English, Spanish, German, etc.). “Language” refers to a category, or data domain, of expression through characters. Belonging to a given language is not specifically based on what the language is, but the customs or traits expressed in that language.

Abstract: Disclosed herein are various embodiments for pronoun-based natural language processing. An embodiment operates by receiving a plurality of text-based sentences each comprising a plurality of words, and each text-based sentence including a pronoun. A plurality of candidate nouns are identified amongst the plurality of words. A trigger word is identified from the plurality of words, wherein the trigger word is associated with both the pronoun and one of the plurality of candidate nouns. A score for each of the candidate nouns is received based on a relationship with the trigger word. The candidate noun with a highest score is selected as being associated with the pronoun.

Abstract: A print bot service analyzes a character string received from a message application service and input from a user by using a keyword for each language associated with printing, determines a language used in the character string, and in accordance with the determined language, switches a language used in a message transmitted to the message application service.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a text classification machine learning model. One of the methods includes training a model having a plurality of parameters and configured to generate a classification of a text sample comprising a plurality of words by processing a model input that includes a combined feature representation of the plurality of words in the text sample, wherein the training comprises receiving a text sample and a target classification for the text sample; generating a plurality of perturbed combined feature representations; determining, based on the plurality of perturbed combined feature representations, a region in the embedding space; and determining an update to the parameters based on an adversarial objective that encourages the model to assign the target classification for the text sample for all of the combined feature representations in the region in the embedding space.

Abstract: A computing device has a display, one or more processors, and memory. The memory stores one or more programs configured for execution by the one or more processors. The computing device receives, via a first region of a graphical user interface, a first natural language input related to a data source. The first natural language input includes one or more terms. In response to receiving the first natural language input, computing device generates one or more interpretations for the first natural language input, including: determining a first term of the one or more terms that is excluded from the one or more interpretations, and visually de-emphasizing the first term in the first region.

Abstract: Disclosed are a method and apparatus for generating prediction information, and an electronic device and a medium. One embodiment of the method comprises: acquiring at least one input word; generating a word vector of each input word of the at least one input word to obtain a word vector set, wherein the at least one input word is obtained by performing word segmentation on target input text; generating an input text vector on the basis of the word vector set; and on the basis of the input text vector and a user vector, generating prediction information for predicting a user intention, wherein the user vector is obtained on the basis of user historical record information. In this embodiment, prediction information for predicting a user intention is generated, such that the popping up of unnecessary information is reduced. A user can be prevented from being disturbed, thereby improving the user experience.

Abstract: Described herein is a system and method of processing data of a digital twins object model. Data associated with a node of the digital twins object model is parsed to identify metadata associated with the node. The data can comprise telemetry data received from an IoT device associated with the node. User-defined function(s) are determined that match the identified metadata. The data and metadata can be provided to the user-defined function(s). The determined user-defined function(s) are executed. Also described herein is a method of creating a user-defined function for processing data of a digital twins object model. Information regarding business logic is received from a user defining a user-defined function. Information regarding telemetry condition(s) to which the user-defined function applies is received defining a matcher. The user-defined function and matcher are stored as objects within a spatial intelligence graph associated with the digital twins object model.

Abstract: Representative embodiments disclose mechanisms to map natural language input to an application programming interface (API) call. The natural language input is first mapped to an API frame, which is a representation of the API call without any API call formatting. The mapping from natural language input to API frame is performed using a trained sequence to sequence neural model. The sequence to sequence neural model is decomposed into small prediction units called modules. Each module is highly specialized at predicting a pre-defined kind of sequence output. The output of the modules can be displayed in an interactive user interface that allows the user to add, remove, and/or modify the output of the individual modules. The user input can be used as further training data. The API frame is mapped to an API call using a deterministic mapping.

Abstract: The described technology is generally directed towards processing various customer input data to extract frequently recurring customer experience themes, including positive and negative sentiment regarding customer experiences. Natural language processing, image processing, speech recognition and/or computer vision techniques can be used on customer-related data to determine themes, tests and scenarios, as well as discover insights that can be used to improve customer experiences. The technology can be used to recreate a customer engagement, journey and overall experience by designing test scenarios around failure themes.

Abstract: Systems, methods, and computer-readable storage media for aggregating media (and commentary on that media) into a topology. To do so, the system receives first content (and associated metadata) as well as second content (and associated metadata). The system then generates a topology based on a relationship between the first content and the second content, where the topology has a number of dimensions based on the metadata of the different pieces of content. The system then compares the topology and the metadata to previously stored topologies and/or metadata and, based on that comparison, executes a machine learning algorithm. The output of that machine learning algorithm includes predicted future changes to the topology, which the system uses to reduce the number of dimensions within the topology.

Abstract: The present application discloses a VR scene and an interaction method thereof, and a computer-readable storage medium. The method includes: constructing a semantics path, here the semantics path is a trajectory drawn on geometry of the virtual environment agent and composed of nodes and directed connections among the nodes, and information of the nodes includes at least node position information, node behavior semantics information and node environment semantics information; the semantics path, moving the virtual character agent according to its own task and the semantics path, and executing a target action; acquiring action result information according to information of a target action of the virtual character agent at the node of the semantics path, and instructing the semantics path processing unit to update the node environment semantics information of the node of the semantics path. Embodiments of the present application can effectively realize interaction in the VR scene.

Abstract: Methods and systems for analyzing care data are described. The method includes building a virtual model of a physical medical device. Training is provided to a patient associated with the physical medical device to properly use the physical medical device by manipulating the virtual model. First care data associated with a first sensor associated with the physical medical device and second care data associated with a second sensor is received by a cloud service. The cloud service analyzes the first care data to obtain a first care data score and analyzes the second care data to obtain a second care data score. The cloud service scores, using a machine learning algorithm, the first care data score and the second care data score to obtain a combined care score. The cloud service determines whether the combined care score is greater than a threshold. The cloud service triggers an emergency procedure when it is determined that the combined care score is greater than the threshold.

Abstract: Persistence and linking of analytic products is provided. Information regarding a plurality of analytic methods is collected. A first process node is generated in a network. The first process node corresponds to a first analytic method. Information is collected regarding a plurality of executions of the first analytic method. A plurality of session nodes is generated in the network corresponding to the plurality of executions. Each of the plurality of session nodes is linked to the first process node. Metadata regarding the plurality of executions is associated with the plurality of session nodes. At least one product node is generated corresponding to a product. The product integrates a result value of at least one of the plurality of executions. The at least one product node is linked to the session node of the plurality of session nodes corresponding to the at least one of the plurality of executions.

Abstract: A method for semantic recognition includes: in response to performing semantic analysis on information acquired by a terminal, a sentence to be processed is acquired. Word recognition is performed on the sentence to be processed, to obtain a plurality of words and part-of-speech information thereof. A target set update operation is determined with a pre-trained word processing model, according to a word to be processed in the set of words to be processed and part-of-speech information of the word to be processed. If a dependency relationship corresponding to the target set update operation is a first dependency relationship, through each of the plurality of preset set update operations, a respective dependency relationship of the word to be processed and a respective confidence level corresponding to the dependency relationship is determined, and a respective update of the set of words to be processed is performed.

Abstract: A learning device includes a learning unit that executes learning of determining a corrected polishing condition by updating an action value function based on state information including at least one polishing condition and a calculation result calculated based on at least one measured value during polishing.

Abstract: When composing an electronic message having at least one intended recipient, a contact manager provides contact information for one or more related recipients. The contact manager may receive an identifier associated with the intended recipient(s), which may be used to lookup and identify contact information for related recipient(s). The contact manager may access a contact database, and, using the identifier and relationship information, determine a contact for the related recipient. The contact manager may then populate the contact information for the contact in a recipient field or other compose surface.

Abstract: A system and method includes identifying unstructured conversational dialogue data sourced from communications between a subscriber and a conversational dialogue agent; automatically mapping, via the one or more computers, one or more distinct unstructured data synthetization requests defined in the unstructured conversational dialogue data to a distinct artifact synthetization objective defined within a synthetization objective distillation layer; generating, via the one or more computers, a plurality of artifact synthetization prompts corresponding to the plurality of unstructured synthetization data requests based on the distinct artifact synthetization objective mapped to each of the one or more distinct unstructured synthetization data requests; and generating, by a target machine learning model, a plurality of synthesized digital artifacts based on an input of the plurality of artifact synthetization prompts generated for the plurality of unstructured synthetization data requests.

Abstract: Systems and methods for personalizing messages in a conversational chatbot are disclosed. An example method may include receiving clickstream event data corresponding to click events by users of an application, generating featurized clickstream data based at least in part on the received clickstream event data, determining one or more predicted intentions for a first user based at least in part on the featurized clickstream data, and generating one or more personalized messages for the first user based at least in part on the one or more predicted user intentions.

Abstract: Systems and methods for computer-generated reality user hailing are described. Some implementations may include accessing sensor data captured using one or more sensors; detecting a person within a distance of a head-mounted display based on the sensor data; detecting a hail event based on the sensor data; and responsive to the hail event, invoking an alert using the head-mounted display.

Abstract: Systems, computer-implemented methods, and computer program products to facilitate a dual learning bridge between text and a knowledge graph are provided. According to an embodiment, a system can comprise a processor that executes computer executable components stored in memory. The computer executable components comprise a model component that employs a model to learn associations between text data and a knowledge graph. The computer executable components further comprise a translation component that uses the model to bidirectionally translate second text data and one or more knowledge graph paths based on the associations.

Abstract: Provided is a setting change assist apparatus for a vehicle configured to: determine whether a setting change request for changing a setting state of driving support control is issued; when determining that the setting change request is issued, notify a driver of confirmation information including information on a content of the setting change request and information on a predetermined approval operation with respect to the setting change request; and change the setting state of the driving support control in accordance with the setting change request when the driver performs the approval operation.

Abstract: Resume data identification implementations are described that identify anonymized resume corpus data pertaining to the same individual. In one implementation, identifying this data in the anonymized resume corpus involves segmenting the corpus into resume snippets and clustering the resume snippets into groups. Within each group the resume snippets potentially pertain to the same individual. In addition, one or more optional filtering operations can be performed to remove snippets from a group that are less likely to be associated with the same person. This filtering is especially useful when the resume corpus is large.