Genesys Cloud Services, Inc.

A method for generating an inverted search index for event sequence visualization may include determining, from an event sequence dataset containing event sequences and events, event pair combinations based on identified unique singleton events. The method may further include establishing a positional ordering for each singleton event and each event pair combination and generating a respective event sequence set for each event sequence. The method may also include determining whether each unique singleton event and each event pair combination exists or does not exist within each event sequence, and populating, for each respective event sequence set and according to the positional ordering, data indicative of whether each unique singleton event and each event pair combination exists or does not exist within the event sequence for the respective event sequence set. The method may also include generating the inverted search index including each respective event sequence set for each event sequence.

A method for determining representative utterances from input utterances to generate paraphrased utterances with healthy intents which optimizes intent health in a bot flow architecture. The conversational bot flow architecture includes a machine learning model trained for natural language understanding (NLU) within a NLU domain that is defined by a collection of intents and sets of associated utterances. The method includes obtaining, by a relevance processing module an overall relevance score of the input utterances based on a vector similarity, a key phrase quality score, a non-ideal length penalty or any combination thereof; obtaining, by a diversity processing module an overall diversity score between a candidate input utterance and previously selected representative utterances, based on both a token value difference and a token order difference; and retrieving, by an adaptive utterance retrieval module selected representative utterances based on the overall relevance score and the overall diversity, where the selected representative utterances are used to generate paraphrased utterances with healthy intents.

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.

A system and method of identifying occurrence of a semantic variation of a phrase in a passage by at least one processor may include calculating a phrase embedding vector, representing a semantic meaning of the phrase; extracting, from a textual representation of the passage, at least one hierarchical set of nested sequences of words; for each sequence, calculating a corresponding sequence embedding vector, representing a semantic meaning of the sequence; for one or more sequence embedding vectors, calculating a corresponding vector similarity value, representing similarity of the sequence embedding vectors to the phrase embedding vector, identifying a sequence corresponding to a maximal vector similarity value of the one or more vector similarity values; and determining the identified sequence as a semantic variation of the phrase, based on the maximal vector similarity value.

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.

A method for agent training using generative artificial intelligence and machine learning according to an embodiment includes retrieving, by a computing system, original training content for contact center agents, analyzing, by the computing system, the original training content using machine learning based on agent characteristics of a particular agent to determine target content characteristics for training content customized to the particular agent, generating, by the computing system, custom agent training content using a generative artificial intelligence system based on the original training content and the target content characteristics, providing, by the computing system, a virtual training session for the particular agent using the generated custom agent training content, receiving, by the computing system, results data associated with the particular agent’s completion of the virtual training session, and updating, by the computing system, an artificial intelligence model leveraged by the machine learning based on the results data.

A system for generating outbound video using generative artificial intelligence and machine learning according to an embodiment includes retrieving, by a computing system, a plurality of attributes of a target video recipient for each of a plurality of target video recipients, wherein each attribute of the plurality of attributes is indicative of a preference or characteristic of the respective target video recipient, generating, by the computing system, a customized outbound video message for each of the plurality of target video recipients using a generative artificial intelligence system based on the plurality of attributes for the respective target video recipient and at least one artificial intelligence model, and transmitting, by the computing system, each customized outbound video message to the respective target video recipient in response to generating the customized outbound video message for each of the plurality of target video recipients.

A system for carbon emission reduction according to an embodiment includes at least one processor and at least one memory comprising a plurality of instructions stored thereon that, in response to execution by the at least one processor, causes the system to determine whether an application for providing one or more services in a contact center is hosted in an on-premises system or a cloud-based system. The instructions may also cause the system to determine, in response to a determination that the application is hosted in a cloud-based system and as a function of one or more cloud hosting implementation properties, a score indicative of a carbon emission efficiency and provide data indicative of a potential carbon emission reduction for the application as a function of the determined score and at least one modification to the cloud hosting implementation properties.

A system and method of training a machine-learning (ML) based model by at least one processor may include receiving an initial dataset, including a plurality of annotated data samples; based on the initial dataset, training at least one ML-based first-level model to perform a first-level task; based on training the at least one ML-based first-level model, calculating at least one characteristic, representing, for each data sample, a value of contribution into a training of a ML-based second-level model to perform a second-level task; omitting a subset of data samples from the initial dataset based on the at least one characteristic, to obtain a target dataset; and training ML-based second-level model, to perform the second-level task, based on the target dataset.