Abstract: Methods for improved fraudster detection in a call center. A subset of a plurality of voiceprints from a plurality of interactions between callers and agents at a call center can be used as the basis for fraudster detection. A plurality of connected components that represents one or more voiceprints can be determined based on the subset of the plurality of voiceprints.

Abstract: A text search query including one or more words may be received. An ASR index created for an audio recording may be searched over using the query to produce ASR search results including words, each word associated with a confidence score. For each of the words in the ASR search results associated with a confidence score below a threshold (and in some cases having one or more preceding words in the ASR index and one or more subsequent words in the ASR index), a phonetic representation of the audio recording may be searched for the word having the confidence score below the threshold, where it occurs in the audio recording, possibly after the one or more preceding words and in the audio recording before the one or more subsequent words, to produce phonetic search results. Search results may be returned include ASR and phonetic results.

Abstract: A system and method for segmenting or dividing a series of computer-based actions, for example into sentences, may provide a sequence of subsets of the series of actions to a neural network using a sliding window, and divide or segment the series actions into segments at points where the loss of the neural network is above a threshold. The dividing may include, for each of a sequence of computer-based actions within a sliding window determining if the sequence when provided to the neural network corresponds to a loss above or equal to a threshold, and if so, determining that an action in the sequence of actions within the sliding window should not be part of a segment or sentence being created.

Abstract: A system and method for determining an embedding for a document (e.g. representing the document in vector space) by determining for the document a preliminary document embedding; determining for the document a document topic embedding based on a set of nearest topics to the preliminary document embedding; determining for each phrase in the document a topic relevancy score based on the document topic embedding and the embedding associated with the phrase; using a ranking algorithm to determine a saliency score for each phrase in the document, each saliency score based on the topic relevancy score for the phrase, and an inverse frequency score for the phrase; and calculating an embedding for the document using the saliency scores and embedding, for the phrases in the document.

Abstract: A system and method for concealing sensitive information may include receiving a screenshot, determining a part of the screenshot includes sensitive information, concealing the sensitive information to thus generate a modified screenshot, and recording the modified screenshot.

Abstract: A system and method may classify a plurality of interactions, by: obtaining a plurality of voiceprints of the plurality of interactions, wherein each voiceprint of the plurality of voiceprints represents a speaker participating in an interaction of the plurality of interactions; calculating, for each interaction, a plurality of scores, wherein each score of the plurality of scores is indicative of a similarity between the voiceprint of the interaction and one voiceprint of a set of benchmark voiceprints; calculating, for each interaction, statistics of the scores; and determining that a plurality of interactions pertain to a single cluster of interactions based on statistics of the scores of the interactions in the cluster.

Abstract: Systems and methods for measuring the effectiveness of an agent coaching program calculate a rate of change in a first Key Performance Indicator for a first agent in a first coaching program during a period of time; select a control group of agents in which agents in the control group of agents were not exposed to the first coaching program; calculate an average rate of change in the first Key Performance Indicator for the control group of agents during the period of time; and calculate a first coaching impact of the first coaching program on the first Key Performance Indicator for the first agent relative to the average rate of change in the first Key Performance Indicator for the control group of agents during the period of time.

Abstract: Systems and methods for generating and implementing a real-time multi-factor authentication policy across multiple channels, are configured to: during a pre-authentication stage: receive, via a user interface, information defining one or more scenarios; receive, via the user interface, information defining one or more authentication flows; for each of the one or more scenarios, map one of the one or more authentication flows to a given scenario; and generate a multi-factor authentication policy associated with each of the one or more scenarios; and during a real-time authentication stage: upon receiving an interaction, identify, by a decision engine, a relevant scenario of the one or more scenarios; implement, by the decision engine, the multi-factor authentication policy associated with the relevant scenario; and determine, by the decision engine, an authentication result.

Abstract: A system and method for training and using a text embedding model may include creating structured phrases from an input text; creating turn input samples from the input text, each turn input sample based on only or consisting of input from a single turn within the text and being formed by removing structure from structured phrases; and training an embedding model using the structured phrases and turn input samples. Call input samples may be created based on input from more than one turn within the text. At each level of resolution (e.g. phrase, speaker, call), a different level of resolution may be used to create input samples. At inference an embedding may be based on a weighted combination of the sub-terms within an input phrase, each weight being based on an inverse document frequency measure for the sub-term associated with the weight.

Abstract: A computerized-method for calculating an After-Call-Work (ACW) factor of an interaction in a contact center, by which a related recording may be filtered for evaluation is provided herein. The method includes an After-Call-Work (ACW) factor calculation module. The operating of the ACW factor calculation module includes: (i) receiving agent recording of the interaction; (ii) aggregating data fields associated with: (a) the interaction; and (b) the customer; (iii) retrieving ACW time of the interaction; (iv) forwarding the aggregated data fields to a machine learning model; (v) operating the machine learning model to calculate a predicted ACW time, based on the aggregated data fields; (vi) calculating an ACW factor based on the received time of ACW and the calculated predicted ACW time; and (vii) sending the calculated ACW factor to a platform by which the platform is preconfigured to distribute the interaction for evaluation, based on the ACW factor.

Abstract: Systems and methods for parallel recording of events on a screen receive start recording requests to start recording the events, each start recording request having an associated interaction identifier; capture the events on the screen in a series of video fragment files until a stop recording request with a corresponding associated interaction identifier has been received for each start recording request; update for each video fragment file, a metadata manifest with information about each interaction having events captured in the given video fragment file; and send the video fragment files, along with the updated metadata manifest to a server as each video fragment file is completed; wherein the video fragment files are forked to interaction files based on the updated metadata manifest, each interaction file associated with a different interaction; and wherein, upon receipt of a stop recording request, the files are merged into a single video recording file.

Abstract: A system and method may determine, e.g. for suggestion to an automation designer, a next action in an automation sequence of actions being created. The automation sequence may include actions executed by a computer system which substitute for actions taken by a user operating a computer system. A sequence of actions may be received, a database of action sequences may be searched for a set of action sequences corresponding to the sequence of actions, and the next action may be presented or determined as being the for example a subsequent new action immediately after the last action matching any action from the received sequence of actions in an action sequence in the set of action sequences.

Abstract: Systems for and methods of determining the priority of a call interaction include receiving a call interaction from a call center; validating, by a validation and transcription engine, that the call interaction is authentic; converting, by the validation and transcription engine, the call interaction into text; extracting, by a data calculation engine, organization, location, and time information from the text; calculating, by the data calculation engine, a priority of the call interaction from the extracted information and the text; determining that the call interaction should be transmitted to a queue of the call center for initial handling by a call center agent; and transmitting the call interaction, the calculated priority, and the extracted information to the call center.

Abstract: A computerized-method for generating a machine-learning model to determine genuineness of customer feedback to filter-out, non-genuine agent recording segments from evaluation. The computerized-method includes generating a Genuineness Opinion Score (GOS) model. The generating of GOS model includes: (a) a data manipulation phase; (b) a data visualization and analysis phase, and (c) a feature augmentation phase for sorting the variables in a set of unique and relevant variables into two categories: estimated variables and anticipated variables. The estimated variables are used for calculation of a GOS of an interaction that is received in a contact center and the anticipated variables are used for calculation of a threshold of said GOS.

Abstract: A system is provided for optimized selection of a plurality of processing units for resource intensive processing operations. The system includes a processor and a computer readable medium operably coupled thereto, to perform the scheduling operations which include receiving a processing operation for a data input that requires processing in a computing environment, determining at least one constraint requirement imposed on performing the processing operation that are all required to be fulfilled for successful completion of the processing operation, accessing a routing table associated with the computing environment, determining one of the plurality of processing units from the routing table based on fulfilling all of the at least one constraint requirement, and assigning the processing operation to the one of the plurality of processing units on the least costly basis or other optimization consideration. The processing units are serverless in a preferred embodiment.

Abstract: A computerized method for evaluating agent-call-logging, in a contact center, is provided herein. The method includes operating an evaluating agent-call-logging module. The operating of an evaluating agent-call-logging module includes: (i) receiving a recorded interaction between an agent and a customer, stored in the database of recorded interactions and a corresponding agent-call-logging, having ‘n’ sentences, stored in the database of call loggings; (ii) operating a speech-to-text algorithm on the received recorded interaction to yield a summarized text thereof. The yielded summarized text is comprised of sentences expressed by the customer; and (iii) operating a semantic comparison unit, to compare each sentence in the yielded summarized text with a corresponding sentence in the agent-call-logging, to yield a Sentence Similarity Score (SSS) for each sentence of the agent-call-logging.

Abstract: A computer based system and method for automatically detecting frustration in an interaction, may include: identifying in the interaction using a set of linguistic rules, natural language patterns related to frustration, wherein the linguistic rules further define weights associated with the natural language patterns and rule metadata; reviewing the rule metadata associated with the identified natural language patterns to identify override attributes, wherein if the rule metadata does not include override attributes, then a frustration level in the interaction is determined based on the identified natural language patterns and weights associated with the identified natural language patterns; and if the rule metadata includes override attributes than the frustration level is determined based on the identified override attributes.

Abstract: Systems and methods for providing unsupervised model health monitoring extract from an interaction database, first and second random samples of interaction data relating to first and second sets of interactions during first and second periods of time; score each interaction of the first and second sets of interactions by applying a predictive model to the related interaction data to produce first and second sets of interaction scores; identify a plurality of sub-populations among the first and second sets of interaction scores by applying a clustering model to the first and second sets of interaction scores; measure distances between each of the plurality of sub-populations among the first and second sets of interaction scores; compare the distances of the first period of time and the distances of the second period of time; and generate an alert when the comparison exceeds a predefined threshold.

Abstract: A system is provided to provide near real-time updates to data tables for a data warehouse using a micro-batch manager. The system includes a processor and a computer readable medium operably coupled thereto, to perform table updating operations which include accessing a first staging data table comprising a plurality of records that was generated by the micro-batch manager for a first destination table of the data tables for the data warehouse, reading a merge instruction column of the first staging data table for the plurality of records, determining a merge operation for each of the plurality of records from reading the merge instruction column, merging the plurality of records from the first staging data table to the first destination table using the merge operation for each of the plurality of records, and deleting the plurality of records from the first staging data table after merging.

Abstract: A combination of two or more offers may be selected from a plurality of offers to be served to potential respondents, in a method performed using one or more processors in a computing system. The method of selecting may comprises receiving a request for a selection of a combination of two or more offers from the sets of offers; in response to the request: performing a series of selection operations each resulting in the selection of one offer to be included in the combination, each selection operation being carried out according to a respective model including one or more variables. The variables of the model for the second and any subsequent selection operation may include at least one selected offer from a previous selection operation. A signal identifying the selected offers may be output to cause the combination of the selected offers to be served to a respondent.