Abstract: A company may implement automated workflows for convenience of users or to reduce support costs. For example, allowing a user to change an address using an automated workflow may be faster or less expensive than with a human agent. In some instances, a first communications session may be started between a first user and a second user. During the first communications session, one or more communications may be processed to select an intent of the first user and a value of an information item communicated by the first user. An automated workflow may be selected to continue assisting the first user, and the first user may be transferred to a second communications session with the automated workflow. The automated workflow may be initialized with the value of the information item that was provided during the first communications session so that the first user does not need to repeat information.

Abstract: A user using a messaging application may be in conversations with multiple people and may inadvertently send a message intended for a first person to a second person. The user may be warned before making such mistakes by processing the text of an entered message and/or the text of the conversations with a mathematical model. A match score may be computed that indicates the match between the entered message and the conversation in which it was entered. Where the match score indicates a possible mistake, a warning may be presented to the user. In some implementations, a match score may be computed using a conversation encoding vector and a message encoding vector. In some implementations, a match score may be computed by processing a sequence of tokens for the conversation and the entered message that includes special token separators.

Abstract: A description of a conversation may be generated to allow a person to understand important aspects of the conversation without needing to review the conversation. The conversation description may be generated by identifying one or more events that occurred in the conversation and then generating the description using the identified events. A set of possible events may be determined in advance for a particular application. The events may be identified by using an event neural network for each event. Each event neural network may process the messages of the conversation to generate an event score that indicates a match between the conversation and the corresponding event. The event scores may then be used to select one or more events. Message scores from the event neural network of a selected event may then be used to select one or more messages of the conversation as a rationale for the selected event.

Abstract: A first named entity recognition (NER) system may be adapted to create a second NER system that is able to recognize a new named entity using few-shot learning. The second NER system may process support tokens that provide one or more examples of the new named entity and may process input tokens that may contain the new named entity. The second NER system may use a classifier of the first NER system to compute support token embeddings from the support tokens and input token embeddings from the input tokens. The second NER system may then recognize the new named entity in the input tokens using abstract tag transition probabilities and/or distances between the support token embeddings and the input token embeddings.

Abstract: A graphical user interface (GUI) page may be represented as GUI page encoding to facilitate processing of the GUI page in an application of GUI pages. A GUI page encoding may be computed by processing a GUI with a GUI page encoding model, and a GUI page encoding model may be trained by processing a training corpus of sequences of GUI pages. The training process may include obtaining first and second GUI pages from the training corpus, computing first and second GUI page encodings with the GUI page encoding model, computing a predicted GUI page encoding by processing the first GUI page encoding with a page predictor model, computing an error value be comparing the predicted GUI page encoding and the second GUI page encoding, and updating parameters of the GUI page encoding model by performing back propagation using the error value.

Abstract: Audio signals of speech may be processed using an acoustic model. An acoustic model may be implemented with multiple streams of processing where different streams perform processing using different dilation rates. For example, a first stream may process features of the audio signal with one or more convolutional neural network layers having a first dilation rate, and a second stream may process features of the audio signal with one or more convolutional neural network layers having a second dilation rate. Each stream may compute a stream vector, and the stream vectors may be combined to a vector of speech unit scores, where the vector of speech unit scores provides information about the acoustic content of the audio signal. The vector of speech unit scores may be used for any appropriate application of speech, such as automatic speech recognition.

Abstract: Medical records may be analyzed to identify important items in the text of the medical record. Actionable content may be identified and may be emphasized or extracted from the medical record. Actionable content may be categorized into one or more categories. Identification may include processing using trained models that use contextual information and position information to determine sentence labels.

Abstract: A graph model of a graphical user interface (GUI) may be generated by processing usage data of the GUI where the usage data comprises sequences of GUI pages and actions between GUI pages. The nodes of the graph model may be determined by obtaining GUI pages from the usage data, identifying dynamic GUI elements in the GUI pages, generating canonical GUI pages by modifying the GUI pages using the dynamic GUI elements, and creating graph nodes using the canonical GUI pages. The edges of the graph may be determined by processing actions from the GUI data that were performed by users to transition from one GUI page to another GUI page. The graph model of the GUI may be used for any appropriate application, such as determining statistics relating to the GUI or statistics relating to individual users of the GUI.

Abstract: The third-party company may provide a user interface to a customer to submit a customer support request regarding a first company. The third-party company may receive a customer identifier from the first device and may request prediction data from the first company using the customer identifier. The third-party company may process the prediction data to predict a customer support request of the customer and generate user interface data corresponding to the predicted customer support request. The user interface data may be transmitted to the user interface before a customer support request from the customer is received.

Abstract: Machine learning-based approaches are used to automatically establish customer support sessions and/or assign customer support requests to customer support agents or other appropriate resources. For example, during a customer support session between a customer and a customer support agent, a trained model can process session data obtained during the session to determine prediction information (e.g., a next message prediction score, a predicted time for receiving a next message, a capacity score, etc.) The prediction information can be compared to an appropriate threshold to determine whether to establish a customer support session and/or assign a customer support request to the agent, even though the agent may otherwise be considered at capacity. In the situation it is determined to establish a session and/or assign a request to the agent, a session can be established and/or a request can be assigned to the agent based on scheduling or other information.

Abstract: The present disclosure relates to a system, method, and computer program for extracting utterances corresponding to a user problem statement in a conversation between a human agent and a user. The system obtains a set of utterances from a natural language conversation between the human agent and the user. The system uses a problem-statement classifier to obtain machine-generated predictions as to whether each natural language utterance in the set relates to a problem statement. The system selects one or more utterances from the set as corresponding to a problem statement based on the predictions. The system provides the selected utterances to a downstream system for further processing. In certain embodiments, the problem statement classifier includes an encoder that creates an utterance embedding for each utterance and a prediction module that uses the utterance embeddings to predict whether each utterance corresponds to a user problem statement.

Abstract: The amount of time required to train a neural network may be decreased by modifying the neural network to allow for greater parallelization of computations. The computations for cells of the neural network may be modified so that the matrix-vector multiplications of the cell do not depend on a previous cell and thus allowing the matrix-vector computations to be performed outside of the cells. Because the matrix-vector multiplications can be performed outside of the cells, they can be performed in parallel to decrease the computation time required for processing a sequence of training vectors with the neural network. The trained neural network may be applied to a wide variety of applications, such as performing speech recognition, determining a sentiment of text, determining a subject matter of text, answering a question in text, or translating text to another language.

Abstract: The present disclosure describes a system, method, and computer program for automatically discovering goals from conversations using neural networks and deep multi-view clustering. A dataset of conversations is partitioned into two views. Vector representations of each view are then generated and clustered in an alternating fashion between views for a number of iterations (i.e., the system alternates between views in generating and clustering vector representations of a view). A first neural network encoder generates the vector representations for the first view, and a second neural network encoder generates the vector representations for the second view. With each semi-iteration, cluster assignments from one view are used to update the encoder for the other view, thus encouraging the two neural network encoders to yield similar cluster assignments.

Abstract: A user may respond to a request of another user by entering text, such as a customer service representative responding to a customer. As the user enters text, a suggested completion of the text may be suggested to the user so that the user may select the suggested completion instead of continuing to enter text. Previous messages between the two users and other information may be used to determine an appropriate suggested completion to the entered text. A neural network language model and a search graph may be used to select a suggested completion from a search graph of possible suggested completions.

Abstract: A user of an automated natural language system may submit an ambiguous or incomplete request, and interactive techniques may be used to obtain clarification information from the user and then determine a response for presentation to the user. A user's initial request may be processed to compute a category score for each possible category of request. The category scores may be processed to determine if clarification of the request is needed. Where clarification is needed, one or more tags may be selected to determine a clarification question to be presented to the user. For example, a tag clarification score may be computed for each tag that indicates a value of the tag in clarifying the request. After receiving the clarification information from the user, one or more categories may be selected or, where needed, additional clarification information may be obtained.

Abstract: Text may be modified according to a specified attribute value, such as changing a 1-star review to a 5-star review. To modify the text, an input sequence of tokens may be obtained corresponding to the text. Mask scores may be computed for the tokens by processing the input sequence of tokens with masking neural network. One or more tokens may be selected using the mask scores, and a masked sequence of tokens may be generated by replacing the selected tokens with a mask token. The masked sequence of tokens may be processed by a language model neural network to select a replacement token for each of the mask tokens. Modified text may then be generated using the selected replacement tokens. The modified text may be used for any appropriate application, such as suggesting messages to users participating in a conversation.

Abstract: A company may implement automated workflows for convenience of users or to reduce support costs. For example, allowing a user to change an address using an automated workflow may be faster or less expensive than with a human agent. In some instances, a first communications session may be started between a first user and a second user. During the first communications session, one or more communications may be processed to select an intent of the first user and a value of an information item communicated by the first user. An automated workflow may be selected to continue assisting the first user, and the first user may be transferred to a second communications session with the automated workflow. The automated workflow may be initialized with the value of the information item that was provided during the first communications session so that the first user does not need to repeat information.

Abstract: A company may implement automated workflows for convenience of users or to reduce support costs. For example, allowing a user to change an address using an automated workflow may be faster or less expensive than with a human agent. Companies may provide support over different types of communications channels with different capabilities, such as voice channels and text channels. Instead of implementing different workflows for different channels, a company may separate aspects of the workflow that are common to different channels from aspects of the workflow that are different for different channels. For example, a workflow may be implemented to determine an action in response to a received communication where the action may be used with multiple channels. The action may then be used to select an action implementation that is specific to a channel.

Abstract: A graph model of a graphical user interface (GUI) may be generated by processing usage data of the GUI where the usage data comprises sequences of GUI pages and actions between GUI pages. The nodes of the graph model may be determined by obtaining GUI pages from the usage data, identifying dynamic GUI elements in the GUI pages, generating canonical GUI pages by modifying the GUI pages using the dynamic GUI elements, and creating graph nodes using the canonical GUI pages. The edges of the graph may be determined by processing actions from the GUI data that were performed by users to transition from one GUI page to another GUI page. The graph model of the GUI may be used for any appropriate application, such as determining statistics relating to the GUI or statistics relating to individual users of the GUI.

Abstract: In some applications, it may be desired to process a message to determine an intent of the message, where the intent indicates the meaning of the message. An intent classifier may be used to determine the meaning of a message by processing the message to compute a message embedding vector that represents the message in a vector space. Each possible intent may be represented by a prototype vector, and the intent of the message may be determined by comparing the message embedding to one or more prototype vectors, such as by selecting an intent whose prototype vector is closest to the message embedding. An intent classifier may be used, for example, (i) to implement an automated communications system with states where each state is associated with a subset of the possible intents or (ii) for processing usage data of a communications system to update the intents of the communications system.