Abstract: A contact center server receives an input from a customer device as part of a conversation. The contact center server (CCS) identifies at least one of: one or more intents, one or more entities, or one or more entity values from the input. The CCS detects one or more escalation conditions based on the input and pauses the conversation based on the detected one or more escalation conditions. The CCS outputs a conversation transcript and the identified one or more intents, the one or more entities, or the one or more entity values to an agent device. The CCS receives agent-identified-information or agent-modified-information from the agent device. Subsequently, the CCS resumes the conversation by providing a response to the input based on the received agent-identified-information or the agent-modified-information.

Abstract: Disclosed herein are system, method, and computer program product embodiments for providing traffic visibility in a network. An embodiment operates by a third-party component receiving a copy of a first data packet during a first period of time. The third-party component extracts a first network parameter associated with the first period of time from the copy of the first data packet. The third-party component then predicts a baseline of normalcy for the first network parameter during a second period of time after the first period of time based on data associated with a copy of a second data packet and the first network parameter. Thereafter, the third-party component receives a copy of a third data packet during the second period of time, and extracts a second network parameter from the copy of the third data packet. The third-party component then determines that the second network parameter of the copy of the second data packet is an anomaly based on the baseline of normalcy for the first network parameter.

Abstract: A machine-learning model (MLM) is trained to identify a given item identifier for an item and shelf dimensions of an empty space associated with the item from training images of a shelf. After training, real-time images of the shelf are provided as input to the MLM and the output provided by the MLM includes empty space identifiers, dimensions or pixel coordinates for each empty space identifier, and an item identifier for each empty space identifier. A quantity of each item identifier is determined based on known shelf dimensions that the corresponding item should occupy on a fully stocked shelf and based on the corresponding empty space dimensions for the empty space associated with the item. A real-time report is sent to store personnel and/or published on a website monitored by the store personnel. The report identifies the items, the shelves, and restocking item quantities that need restocked in the store.

Abstract: Processes and devices for finding signals within a frequency range include converting an analog input signal to a digital domain using an analog-to-digital converter to create a digital input signal. A first fast Fourier transform (FFT) converts the digital input signal to a frequency input signal, which is binned based on frequency and a threshold to create a binary filter. The digital input signal (from the analog-to-digital converter) is sorted into containers (e.g., using the Chinese remainder theorem) and then run through a second FFT to convert the digital input signal to a frequency intermediate signal. The frequency intermediate signal is sorted into bins and multiplied using the binary filter to create an output frequency signal.

Abstract: A virtual assistant server extracts concept(s) from a received user query and correlates the extracted concept(s) to a plurality of nodes in a knowledge graph to identify qualifying path(s) in the knowledge graph. Further, when the two or more qualifying paths are identified, the virtual assistant server addresses one or more ambiguities in the knowledge graph by: prompting the user device to resolve a current one of the one or more ambiguities progressing from contextual nodes towards leaf nodes in the knowledge graph; receiving a response to the prompt from the user device; and repeating the prompting and the receiving until each of the one or more ambiguities is resolved and a matching path to the received query is identified. Finally, the virtual assistant server outputs content associated with a leaf node of the matched path to the user device.

Abstract: A contact center server receives an utterance from a customer device as part of a conversation. The contact center server identifies utterance parameters of the utterance, generates a first response to the utterance based on the utterance parameters, and outputs the utterance parameters and the first response to an agent device. The contact center server receives agent-identified-information corresponding to the utterance or agent-modified-information corresponding to the utterance parameters from the agent device. Subsequently, the contact center server generates a second response to the utterance based on the agent-identified-information or the agent-modified-information and outputs the second response to the agent device.

Abstract: Disclosed herein are system, method, and computer program product embodiments for providing traffic visibility in a network. An embodiment operates by a third-party component in communication with a network component—each located at a network's edge—maintaining a rule table including a first rule comprising first identifiers and a first action for deriving a first packet characteristic. The third-party component receives a first packet copy including second identifiers from the network component. Upon the second identifiers matching the first identifiers, the third-party component determines the rule table's second rule includes a second action for deriving a second packet characteristic. Thereafter, the third-party component receives a second packet copy comprising third identifiers from the network component.

Abstract: A virtual assistant server receives a user input from a virtual assistant interface available on a client system associated with a user identified by a user identifier. Upon receiving the user input, the virtual assistant server creates a search network of repositories not associated with the user identifier by determining, for the received user input, a priority score of each repository not associated with the user identifier based on interactions of the user identifier through the virtual assistant interface. Further, the virtual assistant server searches, for the received user input, in a repository associated with the user identifier and in the repositories in the search network. Based on the searching, a group of one or more results corresponding to the received user input are identified and returned to the client system associated with the user identifier.

Abstract: A virtual assistant server determines a subset of test data corresponding to changes between a first version of training data of a virtual assistant and a second version of the training data of the virtual assistant. Subsequently, the virtual assistant server creates a test suite with the subset of test data and runs the test suite on a second language model of the virtual assistant created using the second version of the training data. Based on the running the test suite, the virtual assistant server generates one or more executable corrective actions to be implemented at the user device and provides the one or more executable corrective actions to the user device to implement to train the virtual assistant.

Abstract: Systems described herein may dynamically add one or more proxy data protection agents to a cloud data storage system to process a data protection job. Upon completion of the job or at some other appropriate interval, the system can power down and decommission the proxy data protection agents and/or the virtual machines on which the data protection proxies reside according to a cleanup schedule (e.g., at hourly or minute intervals). In order to improve the allocation of computing resources, the system takes into account currently existing proxies or virtual machines when processing a backup request to determine the need for new proxies to service the backup request. In this manner the system can save costs and computing resources through efficient virtual machine deployment and retirement.

Abstract: A virtual assistant server receives a web request such as a HTTP request with one or more call parameters corresponding to a call redirected from an interactive voice response server. The virtual assistant server inputs the received one or more call parameters to a predictive model, which identifies, based on the one or more call parameters, an intelligent communication mode to route the redirected call to. Subsequently, the virtual assistant server routes the redirected call to the intelligent communication mode.

Abstract: A virtual assistant server receives a selection of at least two versions of a virtual assistant for comparison from a client device and compares metadata corresponding to each component of the at least two versions. The virtual assistant server presents to the client device, based on the compared metadata, a list of components of the at least two versions and a type of change made to one or more components of the list of components. The virtual assistant server receives from the client device, a selection corresponding to the one or more components of the list of components and presents to the client device, a comparison of definitions of the at least two versions of the selected one or more components of the list of components.

Abstract: A virtual assistant server receives an utterance from an input mechanism. Upon receiving the utterance, the virtual assistant server, evaluates the utterance to identify a plurality of intents corresponding to the utterance and calculates common scores using natural language processing techniques for each of the identified plurality of intents. The virtual assistant server ranks the identified plurality of intents based on the calculated common scores and based on the ranking identifies a first winning intent and a second winning intent. Subsequently, the virtual assistant server prioritizes one of the first winning intent or the second winning intent to identify a final winning intent based on context information. The virtual assistant server executes the final winning intent and forwards a response to the input mechanism.

Abstract: A computer-implemented method includes: receiving, by a computing device, data which is associated with a user; generating, by the computing device, a personalized recommendation for the data by at least one artificial intelligence (AI) application; training, by the computing device, a machine learning (ML) model using the data from the at least one AI application; and generating, by the computing device, a trained fitness model for predicting safety issues based on the trained ML model using the data from the at least one AI application.

Abstract: A radiation-sensitive device is disclosed. The radiation-sensitive device includes: a plurality of single photon avalanche diodes (SPADs), and processing circuitry configured to determine an intensity of incident radiation using at least one of the plurality SPADs. An amount of the SPADs used to determine the intensity of the incident radiation varies in relation to the intensity of the incident radiation. Also disclosed in an associated method of determining an intensity of radiation incident upon such a radiation-sensitive device, and uses of the radiation-sensitive device in an electronic-nose or point-of-care apparatus, or for ambient light sensing.

Abstract: In some examples, a system may be configured to generate one or more query attributes for a search query received from a computing device of a user. Additionally, the system may be configured to, based at least in part on historical data of the user including data characterizing one or more items associated with the user, generate relevant item data. In various examples, the relevant item data characterizing a set of relevant items. Moreover, the system may be configured to, based on the relevant item data, the historical data of the user and the one or more query attributes, implement a set of operations that generate a set of personalized search results associated with the search query.

Abstract: A radiation-sensitive device is disclosed. The device includes an array of single photon avalanche diodes (SPADs) and circuitry configured to measure an intensity of incident radiation from the array of SPADs with a plurality of different measurement windows to provide an associated plurality of results. The circuitry is configured to determine the intensity of the incident radiation from one of the plurality of results, a selection of the result determined by whether the result exceeds a maximum count defined, at least in part, by a duration of the measurement window associated with the result. An associated method of increasing a dynamic range of a radiation-sensitive device comprising an array of SPADs is also disclosed.

Abstract: Implementations disclosed describe systems and methods to optimize allocation of wireless subcarriers to station devices in wireless networks. In an example implementation, the disclosed techniques may include determining, by a station device in a wireless network, that a current set of wireless subcarriers, allocated by an access point device of the wireless network for the station device, is to be changed, generating, by the station device, a feedback information characterizing a current state of one or more of the wireless subcarriers, and sending, by the station device, the feedback information to the access point device.

Abstract: Systems described herein may dynamically add one or more proxy data protection agents to a cloud data storage system to process a data protection job. Upon completion of the job or at some other appropriate interval, the system can power down and decommission the proxy data protection agents and/or the virtual machines on which the data protection proxies reside according to a cleanup schedule (e.g., at hourly or minute intervals). In order to improve the allocation of computing resources, the system takes into account currently existing proxies or virtual machines when processing a backup request to determine the need for new proxies to service the backup request. In this manner the system can save costs and computing resources through efficient virtual machine deployment and retirement.

Abstract: A radiation-sensitive device is disclosed. The device comprises a plurality of single photon avalanche diodes (SPADs) and circuitry configured to adapt a read-out rate of the plurality of SPADs in relation to an intensity of incident radiation. Also disclosed is an associated method of increasing a dynamic range of a radiation-sensitive device comprising a plurality of SPADs. The method comprises adapting a read-out rate of the plurality of SPADs in relation to an intensity of incident radiation.