Abstract: A method for error reduction in intent classification according to an embodiment includes determining a set of possible intent classes for a section of input text, mapping each of the possible intent classes into a respective group of a plurality of groups, determining which groups of the plurality of groups are contradictory to one or more other groups of the plurality of groups, splitting the section of input text into a plurality of sub-parts, performing intent classification on the section of input text to determine an intent of the section of input text and on each sub-part of the plurality of sub-parts to determine a respective intent of each sub-part, evaluating the determined intent of the section of input text and the respective intents of the sub-parts for intent contradiction, and overriding the determined intent of the section of input text in response to identifying an intent contradiction.

Abstract: The present invention relates generally to the technological field of search engines. More specifically, the present invention relates to detecting statistical irregularities in data and retrieving records pertaining to these irregularities. The invention represents a system and methods of retrieving records from an indexed database which provide an improvement of the technological field of search engines and information retrieval by making the detection and retrieval of records representing statistical irregularities in the database faster, more efficient, and less computationally complex.

Abstract: A method for performing time constrained queries for event sequence visualization may include receiving an event sequence dataset including event sequences. Each event sequence includes at least two sequentially occurring events. The method further includes generating a forward index having event pattern sets that include pattern occurrence data for each event sequence positioned according to a fixed positional order. The method includes generating a time constraint search structure, such as a B?tree, B+tree, or table, that includes keys and pointer values. The method also includes inserting, into the search structure, a sequence identifier corresponding to each event sequence based on the fixed positional order and time interval data for the at least two sequentially occurring events of each event sequence.

Abstract: A method of leveraging machine learning to predict empathy for improved contact center interactions according to an embodiment includes receiving, by a computing system, at least one user message from a real-time contact center interaction with a user, generating, by an artificial intelligence system of the computing system, at least one empathy score based on the at least one message using the machine learning, wherein each of the at least one empathy score is indicative of a real-time empathy of the user, generating, by the artificial intelligence system of the computing system, an empathetic text response to the at least one user message based on the at least one empathy score, and responding to the at least one user message in the real-time contact center interaction based on the empathetic text response generated by the artificial intelligence system of the computing system.

Abstract: A method for selecting forecasting models for generating timeseries workload forecasts for a contact center covering varying timeseries granularities and operating horizons. The method includes selecting, via a first selection process, a first select forecasting model from first candidate forecasting models for forecasting a workload level in accordance with a lower-granularity timeseries.

Abstract: A method for generating a customer bot and using the customer bot to train agents, where a first process generates the customer bot and a second process uses the customer bot to train the agents. The first process includes: gathering conversation data; mining intents from the conversation data; constructing, from the mined intents, a dialog engine simulating an interaction type; uploading the customer bot to an automated training module for use thereby; and periodically repeating the previous steps so to update the customer bot with recent conversation data. The second process includes: monitoring for triggering events; initiating the training by initiating a virtual communication to a user device of a first agent; connecting the virtual communication to the customer bot; conducting a simulated interaction; and analyzing one or more statements received from the first agent to derive a performance assessment.

Abstract: A system and method of determining a sentiment of a participant in an interaction may include: obtaining a plurality of textual segments, each representing a portion of the interaction, and labeled according to a specific participant; inferring a language model on one or more textual segments of the plurality of textual segments, to generate respective semantic embedding vectors, each representing a semantic meaning of the respective textual segment in a semantic vector space; compiling a semantic vector set that includes (i) a target semantic embedding vector, corresponding to a target textual segment of a target participant, and (ii) one or more peripheral semantic embedding vectors, respectively corresponding to one or more peripheral textual segments of the plurality of textual segments; and inferring a composite machine-learning (ML)-based model on the semantic vector set, to classify a sentiment of the target participant, as expressed in the target textual segment.

Abstract: A process for generating a query in response to a customer indicating a point of interest on a webpage. The query generation process includes: queuing, in a web content message queue, web content messages of web content used to render the webpage and storing in a datastore as stored web content messages; receiving input from the customer via a point of interest indicator selecting a visual element on the webpage as the point of interest; determining a web element associated with the visual element for transmitting to a point of interest backend service; initiating a search of the stored web content messages using the web element to determine matching data; gathering associated data related to the matching data for indicating a context of the first query; and providing the associated data to the chatbot feature for generating the first query.

Abstract: A method of leveraging a unified conversational system according to an embodiment includes receiving, by the unified conversational system, a query statement from a user made to a chatbot, determining, by the unified conversational system, an intent of the query statement received from the user, preparing, by the unified conversational system, a query for an application service based on the intent of the query statement, sending, by the unified conversational system, the query to the application service via an application programming interface (API) of the application service, receiving, by the unified conversational system, a query response from the application service, and providing, by the unified conversational system, the query response to the user via the chatbot.

Abstract: A system for dynamic content generation for contact center interactions using artificial intelligence according to an embodiment includes determining, by a computing system and using an artificial intelligence model, an intent of an interaction between a contact center agent and a client based on a live transcript, identifying, by the computing system in real time, at least one image relevant to the determined intent in response to determining the intent of the live transcript, displaying, by the computing system in real time, the at least one image for the contact center agent on a dynamic interaction widget application executed by an agent device of the contact center agent, and sending, by the computing system, a selected image of the at least one image displayed on the dynamic interaction widget application to the client in response to selection of the selected image by the contact center agent.

Abstract: A method of selectively generating message notifications for chat messages to a user of a chat channel. The method includes: activating a mute feature for the chat channel; activating an auto-unmute feature for the chat channel; receiving a search target defining a message characteristic; receiving a first new message; analyzing the first new message to determine if the message characteristic of the search target is present in the first new message; if the search target is determined to not be present in the first new message, maintaining the mute feature so that, in relation to the first new message, no message notification is generated; and if the search target is determined to be present in the first new message, overriding the mute feature so that, in relation to the first new message, a first message notification is generated.

Abstract: A method for creating a student model from a teacher model for knowledge distillation. The method including: providing a first model; using a first instance of the first model to create the teacher model by training the first instance of the first model on a training dataset; using a second instance of the first model to create the student model by training the second instance of the first model on a subset of the training dataset; identifying corresponding layers in the teacher model and the student model; for each of the corresponding layers, computing a weight similarity criterion; ranking the corresponding layers according to the weight similarity criterion; selecting, based on the ranking, one or more of the corresponding layers for designation as one or more discard layers; removing from the student model the one or more discard layers.

Abstract: A method for verifying anonymization by a first user that the subject text does not contain personal identifiable information (PII). The method includes: receiving the subject text; receiving a non-PII wordlist listing non-PII words; comparing each word appearing in the subject text to the non-PII words to determine matches therebetween so that, via the comparison, the words of the subject text are classified as being either first text, which includes the words in the subject text found to match one of the non-PII words, and second text, which includes the words in the subject text found not to match any of the non-PII words; and generating a first user interface that displays the subject text such that a visual format of the first text differs from a visual format of the second text in accordance with a visual format alteration.

Abstract: A method in a contact center for generating an action classifier model and use thereof in selectively initiating turn set queries of a knowledge base to assist agents in real time during ongoing conversations with customers. The method includes: generating an action classifier model; receiving classification data that classifies a first plurality of the customer actions found in training samples as belonging to a first action category for which a knowledge base search is deemed needed, and a second plurality of the customer actions as belonging to a second action category for which a knowledge base search is deemed not needed; and using the action classifier model and the received classification data to perform a query filtering routine for selectively initiating a turn set query for a present turn set occurring in an ongoing conversation between an agent and customer.

Abstract: A method for managing a queue in a contact center that includes: receiving a previous call from a first customer and determining a unique identifier; tracking a previous queue time; detecting the previous call ending without being connected to an agent; determining the previous wait time accrued; saving in a database a previous call record that includes the previous queue time and the unique identifier; receiving a present call from the first customer and determining a unique identifier; tracking a present queue time; using the unique identifier to check the database to determine the related previous call record; determining the previous queue time; calculating an adjusted total queue time that is the present queue time plus previous queue time; and positioning the first customer in the queue based on the adjusted total queue time.

Abstract: A method of verifying that incoming telephone calls to respective customers originate from a business entity known to the customers that includes: receiving data transmitted from a customer application on a first customer device that notifies of a triggering event occurring; generating a business-side version of the verification code for use in association with the first authentication attempt; generating, by the first customer device, a customer-side version of the verification code for use in association with the first authentication attempt; generating a user interface showing the customer-side version of the verification code; and transmitting the generated business-side version of the verification code to a first agent device for authenticating that the first telephone call originates from a first business entity by enabling a comparison verifying that the customer-side version and the business-side version of the verification code are equal.