Abstract: A processor may receive user information for a request payload from an external device and data describing a plurality of user interface (UI) elements configured to be presented in a UI of the external device. The processor may select a machine learning (ML) model from a plurality of ML models using a contextual bandit ML model that is trained based on the user information. The processor determines at least one recommended user interface (UI) element with a selected ML model, based on the user information and the data describing the plurality of UI elements. The at least one recommended UI element may be presented in the UI of the external device. The processor may receive event data indicating a user interaction with the at least one recommended UI element in the UI of the external device. The contextual bandit ML model may be re-trained based on the event data.

Abstract: An orchestration layer for execution user defined automation workflows. The orchestration layer may include multiple process instances that host user defined automation workflows that automate processes or tasks. To improve system performance and reduce operating costs, the user defined automation workflows are deployed to the orchestration layer in a standard format that standardizes the user defined workflow configurations. The orchestration layer may also dynamically scale the computational resources allocated to teach process instance based on the properties of each user defined automation workflow.

Abstract: Systems and methods are provided for generating recommendations using contextual bandit models with non-linear oracles.

Abstract: A method for detecting fraudulent financial transactions in information technology networks involves obtaining a multitude of features associated with a financial transaction conducted over an information technology network by an unknown transaction party. The multitude of features includes clickstream data obtained from the unknown transaction party. The clickstream data is associated with data of the financial transaction being entered by the unknown transaction party. The method further involves obtaining a first fraud indicator using a machine learning classifier operating on the multitude of features, obtaining a second fraud indicator using a rule-based classifier operating on the multitude of features, obtaining a fraud prediction for the financial transaction, using the first fraud indicator and the second fraud indicator, and taking an action, in response to the fraud prediction.

Abstract: Certain aspects of the disclosure provide a method, comprising: processing input data with an ensemble of nonlinear machine learning models; generating a sparse high-dimensional embedding based on one or more leaf nodes of each nonlinear machine learning model in the ensemble of nonlinear machine learning models; projecting the high-dimensional embedding into a lower-dimensional embedding, wherein the lower-dimensional embedding is less sparse than the high-dimensional embedding; processing the lower-dimensional embedding with a linear machine learning model to generate a binary class prediction; determining a confidence for the binary class prediction; and outputting: the binary class prediction if the confidence is greater than or equal to a threshold; or a flipped binary class prediction if the confidence is lower than the threshold.

Abstract: A method classifies feedback from transcripts. The method includes receiving an utterance from a transcript from a communication session and processing the utterance with a classifier model to identify a topic label for the utterance. The classifier model is trained to identify topic labels for training utterances. The topic labels correspond to topics of clusters of the training utterances. The training utterances are selected using attention values for the training utterances and clustered using encoder values for the utterances. The method further includes routing the communication session using the topic label for the utterance.

Abstract: Systems and methods for enriching raw user text with a database to identify relevant context, wherein generated prompts provide responses to user queries is provided. A method includes receiving a query, wherein the query comprises the raw text, creating a first embedding based on the query, retrieving a plurality of other embeddings, wherein the plurality of other embeddings are complementary to the first embedding, creating a contextual prompt including context from at least one of the plurality of other embeddings, processing the contextual prompt using a trained machine learning model, thereby generating a response to the query, and causing the response to be displayed by a display device.

Abstract: A method includes obtaining matches between target records in a target dataset and a reference records in a reference dataset, each match of the matches comprising a corresponding confidence level of the match, categorizing the target records into review level categories according to the corresponding confidence level, and presenting a graphical user interface (GUI). The GUI includes a first section for a first review level category showing a first subset of the target records assigned to the first review level category, the first subset comprising target records related, in the GUI, to at least one matching reference record. The GUI includes a second section for a second review level category, wherein the second section shows a second subset of the target records assigned to the second review level category, the second subset comprising target records related, in the GUI, to at least one matching reference record.

Abstract: At least one processor may receive a query response generated by a query machine learning (ML) model, wherein the query response is generated in response to a query from a client device. The at least one processor may generate an evaluated likelihood of the query response being found in a training data set comprising known valid data, wherein the generating is performed using an evaluation ML model. The at least one processor may determine that the evaluated likelihood indicates the query response is likely to include valid data. In response to the determining, the at least one processor may return the query response to the client device.

Abstract: A method classifies feedback from transcripts. The method includes receiving an utterance from a transcript from a communication session and processing the utterance with a classifier model to identify a topic label for the utterance. The classifier model is trained to identify topic labels for training utterances. The topic labels correspond to topics of clusters of the training utterances. The training utterances are selected using attention values for the training utterances and clustered using encoder values for the utterances. The method further includes routing the communication session using the topic label for the utterance.

Abstract: A method implements brand engine for extracting and presenting brand data with user interfaces. The method includes receiving a blueprint with a set of structure blocks extracted from a selected content. A structure block of the set of structure blocks includes a set of style parameter requests for a section of the selected content. The method further includes processing the set of structure blocks with a first set of smart blocks to generate a set of scores. A smart block of the first set of smart blocks includes brand data with style parameter selections. The method further includes selecting a second set of smart blocks, for the set of structure blocks, from the first set of smart blocks, using the set of scores. The method further includes presenting the second set of smart blocks with the brand data.

Abstract: A Large Language Model (LLM) for classifying documents by identifying indicators within the documents. A smart caching mechanism stores document classifications and associated indicators output from the LLM. The database contains document details, classifications, and associated indicators. A classification module classifies a new document by analyzing it for indicators, checking the cache for a match, and querying the database for the indicators if no match is found. The module applies a majority vote based on the classifications associated with the indicators.

Abstract: Methods and systems for assisting entities with improving the effectiveness of their profiles are disclosed. An example method is performed by one or more processors of a system and includes storing profile data including profiles identifying attributes associated with respective entities, obtaining a selection data vector including values each indicating a selection rate for a respective entity, generating, using a trained analysis model, selection prediction data predicting, for each respective change of a set of possible changes to a selected entity's profile, how the selection rate for the selected entity will change if the selected entity's profile is adjusted in accordance with the respective change, selecting, from the selection prediction data, one or more recommended changes likely to result in an increase in the selection rate for the selected entity, and outputting a prompt recommending that the selected entity make one or more recommended changes to the selected entity's profile.

Abstract: Systems and methods for matching received product information with stored product information. Incoming product information has multiple attributes, which may be fuzzy matched with corresponding attributes of stored product information to generate corresponding fuzzy matching scores. Each of the fuzzy matching scores is associated with a weighting factor, which is used to indicate a contribution of the corresponding fuzzy matched attribute to a match between the entire product information. A matching coefficient is initialized and progressively updated by using the weighted fuzzy matching scores. When a desired number of fuzzy matchings between the corresponding attributes is reached and the matching coefficient is finalized, the matching coefficient is compared to a threshold. If the matching coefficient is above the threshold, a recommendation is generated indicating a match between the received product information and the stored product information.

Abstract: A transaction model of a general model generates a target transaction vector for a target transaction record. The general model also generates account vectors for accounts. A match score is generated between the account vectors and the transaction vector. The general model selects a first account identifier of an account using the match score. The transaction model also generates historical transaction vectors for historical transaction records. Further, a comparison score is generated between the historical transaction vectors and the target transaction vector. A second account identifier of an historical transaction is selected according to the comparison score. One of the first account identifier and the second account identifier is selected as the account identifier for the transaction record, and the transaction record is stored with the account identifier.

Abstract: A computing system generates a plurality of training data sets for generating the NLP model. The computing system trains a teacher network to extract and classify tokens from a document. The training includes a pre-training stage where the teacher network is trained to classify generic data in the plurality of training data sets and a fine-tuning stage where the teacher network is trained to classify targeted data in the plurality of training data sets. The computing system trains a student network to extract and classify tokens from a document by distilling knowledge learned by the teacher network during the fine-tuning stage from the teacher network to the student network. The computing system outputs the NLP model based on the training. The computing system causes the NLP model to be deployed in a remote computing environment.

Abstract: Systems and methods that may be used to provide policies and protocols for blocking decryption capabilities in symmetric key encryption using a unique protocol in which key derivation may include injecting a random string into each key derivation. For example, a policy may be assigned to each client device indicating whether the client device has been assigned encryption only permission or full access permission to both encrypt and decrypt data. The disclosed protocol prevents client devices with encryption only permission from obtaining keys for decryption.

Abstract: The present disclosure relates to deriving cryptographic keys for use in encrypting data based on a plaintext to be encrypted. An example method generally includes receiving, from a querying device, a request for a cryptographic key. The request generally includes data derived from a plaintext value to be encrypted and an indication of a type of the plaintext value to be encrypted. A cryptographic key is generated based, at least in part, on the derived data and the type of the plaintext value to be encrypted. The key deriver transmits the generated cryptographic key to the querying device.