Abstract: A system including one or more processors and one or more non-transitory computer-readable media storing computing instructions, that when executed on the one or more processors, cause the one or more processors to execute operations: determining, using a weight determination model of a machine learning architecture, (a) token weights for tokens included in a search query and (b) extraneous tokens from among the tokens based on at least the token weights; comparing the token weights for the tokens to a probability threshold; segregating, based on at least the token weights, the extraneous tokens from a remainder of the tokens; after segregating, identifying first tokens of the remainder of the tokens that satisfy the probability threshold; and identifying, based on the first tokens of the remainder of the tokens that satisfy the probability threshold, search results for the search query. Other embodiments are described.

Abstract: A system includes a memory storing an AI model, and a processor configure to receive a first request to establish a first number of sessions with the database server for a task, determine that a recommended number of sessions for the task has not been provided, determine features associated with the task, collect features associated with a real-time computing environment for the task, determine an optimal number of sessions for the task using the AI model, determine that the optimal number differs from the first number, provide a recommendation for the task with the determined optimal number of sessions for the task, receive a second request to establish a second number of sessions with the database server for the task, determine that the recommended number of sessions for the task has been provided, and establish the second number of sessions with the database server for the task.

Abstract: A hybrid computer architecture a process providing flexible computing resources across a combination of on-premise computing resources and cloud-based computing resources.

Abstract: Systems, methods, and computer-readable storage media for building knowledge graphs which map the aspects of entities, then transforming the graphs into vectors for a similarity comparison. The resulting vectors can be used to identify skills and competencies of individuals, which the system uses to compare those individuals to others. The system can then quantify who would make for a good replacement for a given individual and make an associated recommendation.

Abstract: Computer systems and methods are disclosed for governing use of patient scoped contextual information during execution of reasoning models, agentic systems, and orchestration workflows operating over regulated data. A patient partitioned contextual memory stores memory atoms associated with provenance metadata, temporal validity metadata, and governance attributes. In response to a task request associated with a patient identity, a retrieval engine selects a candidate set of memory atoms for evaluation by a non-bypassable memory gate under an applicable policy snapshot identifier (or policy digest) and one or more integrity criteria. The memory gate determines admissibility and authorized transformation of memory atoms, and a governed context bundle is deterministically assembled from admitted memory atoms subject to a context capacity constraint. Execution is permitted only through a constrained execution interface requiring the governed context bundle and a gate produced governance artifact.

Abstract: A method and a system for auditing websites for compliance expectations are disclosed. A Website Observer Tool (WOT) may capture network message output (NMO) associated with a website; store the NMO in at least one of a storage device or storage platform; transform the NMO to create transformed output; load the transformed output to a database; aggregate at least part of data from the transformed output to create aggregated content in the database; execute a query to select data from the aggregated content; create a first condition for the NMO based on the selected data as a first expectation; and evaluate the NMO compared to the first expectation to monitor at least one of performance expectation of the website, privacy expectation of the website, or tagging expectation of the website.

Abstract: A computerized-method for matching customer queries across diverse digital channels with a response in a multichannel contact center.

Abstract: A method for automatically supplying or locating data over a communication network comprising: configuring a data source to provide source data over the communication network, the source data being identified by a data identifier in a domain of identifiers; receiving a machine-readable description, the machine-readable description specifying at least one predicate that must be satisfied by the data; labeling data identified by the data identifier with labels at least in part indicative of data predicates that hold true for the data identified by the data identifier or that are satisfiable for the data identified by the data identifier; searching for the data based at least in part on the machine-readable description and the labels; obtaining a solution satisfying the predicate based on the search; and supplying the solution over the communication network to a data consumer.

Abstract: Techniques for cross-regional resource linking in a federated data lake management and integration system. The techniques include receiving a query and identifying a referenced database object. First, metadata about this object is obtained from a local data lake metastore. If this metadata includes a cross-regional resource link, the techniques then retrieve second metadata from a remote data lake metastore. Based on this second metadata, a query execution plan is generated. This plan guides a distributed data processing system to execute data processing jobs on data stored in a distributed data storage system. Finally, the data responsive to the query is returned. The techniques efficiently integrate and manage data lake metadata resources across multiple regions.

Abstract: Input data is obtained and converted into a graph. Based on the converted input data, a time series of snapshot graphs is generated by selecting edges whose timestamps are in a given time window for each of a plurality of time windows, each edge having a corresponding attribute. Internal flow probabilities are computed for each snapshot graph of the time series of snapshot graphs and a system is controlled based on the internal flow probabilities.

Abstract: Methods and systems consistent with the disclosure include a method for generating a website, comprising: receiving, by a computer program executed on an electronic device, a request for a website; generating, by the computer program, a formatted request for the website to a microservice hub; identifying, by the microservice hub, one or more template parts based on the formatted request; determining, by the microservice hub, a site specific data to retrieve and sending a call to a site specific database for the site specific data; determining, by the microservice hub, a prompt to query a first machine learning engine based on an internal or external information content of the one or more template parts to fulfill a one or more prompts based on at least one template parts; and rendering, by the microservice hub, the website based on the prompts and the one or more template parts.

Abstract: The present disclosure involves systems, software, and computer implemented methods for improved design and implementation of data access metrics for automated physical database design. An example method includes identifying a database workload for which index advisor access counters are to be tracked. Each SQL statement in the database workload is executed. For each SQL statement, attribute sets are determined for which a selection predicate filters a result for an SQL statement. An output cardinality of each selection predicate is determined. A logarithmic counter for an attribute set corresponding to the selection predicate is determined based on the output cardinality of the selection predicate. The determined logarithmic counter is incremented. Respective values for logarithmic counters of the determined attributes are provided to an index advisor. The index advisor determines attribute sets for which to propose an index based on the logarithmic counters of the respective attribute sets.

Abstract: A database system is operable to execute a first to generate the machine learning model from a training set of rows based on accessing and processing the training set of rows via a plurality of operators. system configuration data tracking a plurality of objects of a database system is updated to further track the machine learning model as a corresponding first object tracked via the system configuration data. A query output for a second query indicating applying of the machine learning model via execution of the second query based on applying the machine learning model to a set of rows in accordance with at least one property of the corresponding first object based on accessing the system configuration data.

Abstract: Disclosed are implementations that include a method for detecting anomalous data, including converting a set of data values representative of a multi-dimensional item into a nodes-and-edges graph representation of the item, applying a graph convolution process to the graph representation to generate a transformed graph representation for the item comprising a resultant transformed configuration of the nodes and edges representing the item, and determining, based on the transformed configuration, a probability that the item is anomalous. Another example method includes receiving input data at a neural network circuit comprising a plurality of node layers, with each of the plurality of node layers comprising respective one or more nodes, with the neural network circuit further comprising adjustable weighted connections connecting at least some nodes in different layers of the plurality of node layers. The method further includes removing one or more of the weighted connections at one or more time instances.

Abstract: A method and system for collecting, matching and reporting evidence related to a collision event captured on a first collision detection device associated with a vehicular transportation system. A first incident report of the collision event communicated from the first collision detection device to the cloud-based application. The first incident report includes image data, location data, and temporal data of the collision event. A collision event database of the cloud-based application is searched to locate a second incident report of the collision event. The first incident report is compiled with the second incident report when the collision event of the first incident report matches the recorded collision event of the second incident report. A collision evidence report is generated based on the compiled first and second incident reports.

Abstract: The present disclosure is generally directed to methods, apparatus, and computer-readable media (transitory and non-transitory) for learning to automatically navigate interactive web documents and/or websites. More particularly, various approaches are presented for training various deep Q network (DQN) agents to perform various tasks associated with reinforcement learning, including hierarchical reinforcement learning, in challenging web navigation environments with sparse rewards and large state and action spaces. These agents include a web navigation agent that can use learned value function(s) to automatically navigate through interactive web documents, as well as a training agent, referred to herein as a “meta-trainer,” that can be trained to generate synthetic training examples. Some approaches described herein may be implemented when expert demonstrations are available. Other approaches described herein may be implemented when expert demonstrations are not available.

Abstract: Embodiments relate to a system and method to plan preparation of a meal that may include more than one dish. Online recipes are obtained from a database based on criteria set by the user. Recipes are analyzed for time and effort requirements at each step. If a meal includes more than one dish, a merge of the timelines is performed to determine if it is feasible to prepare all dishes concurrently. Suppliers that can supply all ingredients required by the recipes that meet all user criteria are recommended to the user.

Abstract: The present disclosure is related to framework for automatically and efficiently finding machine learning (ML) architectures that are optimized to one or more specified performance metrics and/or hardware platforms. This framework provides ML architectures that are applicable to specified ML domains and are optimized for specified hardware platforms in significantly less time than could be done manually and in less time than existing ML model searching techniques. Furthermore, a user interface is provided that allows a user to search for different ML architectures based on modified search parameters, such as different hardware platform aspects and/or performance metrics. Other embodiments may be described and/or claimed.

Abstract: A search engine of a data exchange may receive a query comprising a set of search terms, retrieve a plurality of data listings based on the search terms of the query, compare a first embedding generated by a large language model (LLM) from the search query to second embeddings generated by the LLM for each of the plurality of data listings to determine a respective relevance for each of the plurality of data listings to the search query, and rank the plurality of data listings based on the respective relevance for each of the plurality of data listings to the search query.

Abstract: An orchestration platform leverages an in-memory database for guaranteed synchronous processing of requests from origin devices. The in-memory database is created on a per-transaction basis and localized to a synchronous orchestration engine, which is a part of an orchestration platform. The in-memory database is serialized and stored in memory. To update the in-memory database, the in-memory database is deserialized and one or more database operations can be performed on the in-memory database. The updated in-memory database is communicated to a return to caller service which deserializes and extracts data from the updated in-memory database. The data is stored in a disk-based database that is accessible by an asynchronous orchestration engine. At this point, the updated in-memory database no longer exists.