Abstract: The disclosed computer-implemented method may include determining a taxonomy of an object from its textual description and also standardized attributes of the object from the description and the taxonomy using a language model, according to embodiments. The method may also include building a graph data structure by using the standardized attributes for a node and connecting the node to other nodes using edges for common attributes. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: Example methods and apparatus for managing a network device are described. One example method includes obtaining a static dataset and generating a knowledge corpus based on the static dataset, where the knowledge corpus includes information for managing a network device. The static dataset indicates how to manage the network device, and the static dataset includes data that is generated independent of a device interaction process.

Abstract: Methods, systems, and devices for data management are described. A first data source may be identified to move from a source cloud deployment (that operates in accordance with a first release cadence for updating database schema) to a destination cloud deployment (that operates in accordance with second, different release cadence for updating database schema). The first source database is then migrated from the source cloud deployment to an intermediate cloud deployment hosted on the source cloud deployments that allows the schema of the first data source to remain unchanged for a duration of time that the first data source is hosted on the intermediate cloud deployment. The database schema associated with the first data source is then updated to an updated (most recent) database schema in accordance with the second release cadence of the destination cloud deployment, and migrated from the intermediate cloud deployment to the destination cloud deployment.

Abstract: Systems and methods for generating prompts for a machine learning model based on user-selected source document information are disclosed. Exemplary implementations may: present a user interface, wherein the user interface depicts content from source documents and obtains user input to select items of user-selected information from the content; generate a prompt based on the selected items; present the generated prompt to the user; and/or perform other steps.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for transforming unstructured data from an application running locally on a client device and indexing the data in a local index for use by a generative model. In particular, in one or more embodiments, the disclosed systems receive unstructured local data from an application running locally on a client device and transform the unstructured local data to formatted local data that matches a data format of a connector for the application. The disclosed systems index the formatted local data in a local index and access the formatted local data for use by a generative model to generate a generative response. In some embodiments, the disclosed systems generate additional contextual data from the application running locally on the client device and index the additional contextual data with the formatted local data for use by the generative model.

Abstract: Methods and systems are presented for providing a framework for analyzing graphs that exhibit non-homophilous behavior. Under the framework, a structural analysis and a feature-based analysis will be performed on a sequence of graphs. When performing the feature-based analysis, various features are extracted from each node in the sequence of graphs, and clusters of nodes are identified from each graph based on the features. A set of evolving prototypes is generated to represent evolving characteristics of the clusters of nodes, and a set of persistent prototypes is generated to represent persistent characteristics of the clusters of nodes. Information derived from the structural analysis of the graphs, the set of evolving prototypes, and the set of persistent prototypes are embedded within the nodes of the graphs. The embedded information is then used to classify the nodes.

Abstract: A system for helping an operator to question a help-seeker is provided. The system includes: taking the positive probability of any second reference group included as a cut-off point, and performing a classification test on any first reference group to obtain a classification sensitivity and a classification specificity corresponding to any second reference group; based on the classification sensitivities and the classification specificities corresponding to a plurality of second reference groups amongst any first reference group, obtaining a receiver operating characteristic curve corresponding to any first reference group; according to the receiver operating characteristic curve, determining a target first reference group from the plurality of first reference groups; and according to the distance between a coordinate point corresponding to any second reference group amongst the plurality of second reference groups and a perfect classification coordinate point, determining a target question semantic meaning.

Abstract: A method, a system, and a computer program product for generating an interpretable set of features. One or more search parameters and one or more constraints on one or more search parameters for searching data received from one or more data sources are defined. The data received from one or more data sources is searched using the defined search parameters and constraints. One or more first features are extracted from the searched data. The first features are associated with one or more predictive score values. The searching is repeated in response to receiving a feedback data responsive to the extracted first features. One or more second features resulting from the repeated searching are generated.

Abstract: Aspects of the present disclosure relate to machine learning-based techniques for segmenting entity populations. Embodiments include approximating distributions for outputs generated by a classification model for entities based on ground truth labels that indicate whether an event occurred with respect to the entities. A first distribution may relate to the distribution of classification model scores that indicate the likelihood of the event occurring with respect to entities for which the event did not occur. A second distribution may relate to the distribution of classification model scores that indicate the likelihood of the event occurring with respect to entities for which the event did occur. Based on the distributions, thresholds may be generated based on determining a gradient with respect to the distributions. Entities may then be segmented based on the thresholds, and interventions (such as resource allocations) may be applied based on the segmentation.

Abstract: An apparatus for heuristic re-indexing of stochastic data to optimize data storage and retrieval efficiency is disclosed. The apparatus includes at least processor and a memory communicatively connected to the processor. The memory instructs processor to receive raw data including at least two datasets. The memory instructs the processor to generate one or more associations as a function of a statistical model. The memory instructs the processor to reorganize the raw data as a function of the one or more associations. The memory instructs the processor to store the reorganized raw data in an index structure by implementing an indexing system as a function of the one or more associations, wherein the indexing system is further configured to dynamically adjust the index structure in response to additional raw data.

Abstract: An apparatus for optimal zone strategy selection, wherein the apparatus comprises a processor and a memory configuring the processor to receive user data; classify the user data to a plurality of target classes; generate a user goal as a function of the plurality of target classes; generate a plurality of zone strategies based on the user goal as a function of identifying a zone strategy score for each of the plurality of zone strategies; determine follow-through data as a function of ranking the plurality of zone strategy scores and the zone strategies; populate a user interface data structure comprising a visual representation of the zone strategies and the follow-through data; and transmit the user interface data structure to a display device communicatively connected to the at least a processor to display the visual representation of the zone strategies and the follow-through data using a graphical user interface (GUI).

Abstract: In some examples, a data management system processes snapshots of a distributed file system, the distributed file system having files, each file comprising multiple data chunks. The data management system performs operations including storing file-to-chunk mapping in file system metadata; creating, for each chunk, a chunk generation ID by associating each chunk with a file system generation ID; in a next-generation snapshot of the distributed file system, incrementing, for all chunks in the next-generation snapshot, the respective chunk generation IDs; and taking a snapshot of the file system metadata and storing an updated file-to-chunk mapping associated with the next-generation snapshot.

Abstract: A system and method for creating a remediation plan with in-context learning (ICL) to a large language model (LLM) involves accessing pieces of content from a corpora and calculating similarity scores between selected content and the corpora. If the similarity score exceeds a set threshold, static metadata—such as authorship, file name, file size, last updated date, or source location—is retrieved. Dynamic metadata, including word count, sentence count, grammar correctness, number of citations, or layout classification, is also generated. Based on static and dynamic metadata, a remediation action is determined, which may involve merging content, prioritizing one piece over another, or ignoring certain content. A remediation prompt is then created and sent to the LLM to generate a preferred output. The process continues by accessing additional unprocessed content and repeating until a time limit expires, computational resources are consumed, or all content has been processed.

Abstract: A recommendation system includes a content similarity analyzer configured to determine a first set of content item identifiers similar to a set of viewed content items based on respective similarity scores and add them to a first list. A similar user content extraction module identifies a set of similar user identifiers from a user similarity database; obtains, based on respective viewing histories of the set, a second set of content item identifiers; and adds them to the first list. The recommendation system includes a content filter configured to select a subset of content item identifiers from the first list based on the corresponding similarity scores between content item identifiers of the first list and a viewing history. The content filter is configured to transmit the subset of content item identifiers for display of the corresponding content items via a web portal on a user interface of a first user device.

Abstract: Certain aspects of the disclosure provide techniques for categorization by a device. An example method includes receiving input information regarding a plurality of classification targets and a plurality of categories for classification of the plurality of classification targets; generating a plurality of embeddings for the input information using a first model, the plurality of embeddings including: a set of first embeddings associated with the plurality of classification targets, and a set of second embeddings associated with the plurality of categories; determining that a similarity score for the set of first embeddings and the set of second embeddings fails to satisfy a threshold; generating, based on the similarity score failing to satisfy the threshold and using a graph neural network (GNN), a classification of the plurality of classification targets in accordance with the plurality of categories; and outputting information regarding the classification.

Abstract: A device for using message identifiers for Publish/subscribe messaging deduplication is described. The system may fetch one or more sets of data records from a data source, and each data record is associated with a message identifier. The system may store the one or more sets of data records in a data file, which is associated with a metadata comprising the message identifier, a file path and a row number for each data record. The system may determine whether one or more of the data records are duplicated based on the associated message identifiers. In response to determining that the one or more data records are duplicated, the system may generate a second metadata comprising the file paths and row numbers associated with the duplicated data records.

Abstract: A processor-implemented method is disclosed.

Abstract: Embodiments of the present disclosure are directed to monitoring and responding to social media content according to one embodiment of the present disclosure. Monitoring and responding to social media content can comprise reading content of a social media source and identifying a sub-set of content from the content of the social media source for response based on a model defining content for which a response is to be prepared. One or more natural language responses to the identified sub-set of content can be generated based on a model defining responses and a knowledge base of information. Each natural language response can comprise factual information from the knowledge base of information. The generated one or more natural language responses can be submitted to the social media source in response to the identified sub-set of content.

Abstract: Disclosed herein are systems and method for data chunking. In one aspect, a method includes consecutively scanning each respective byte in a byte stream of data. The method includes in response to detecting a first amount of adjacent bytes with values arranged in a decreasing order, marking a cut-off point of a data chunk comprising at least the adjacent bytes. The method includes in response to detecting a second amount of bytes with values arranged in an increasing order, executing a jump mechanism that skips scanning of a fixed amount of bytes after the bytes with values arranged in the increasing order. The method includes subsequent to scanning the byte stream in entirety, outputting a plurality of cut-off points of identified data chunks.

Abstract: A KID's computer system is configured by PARENT-controlled software, giving each PARENT the power and responsibility to control the social media content being presented to their KID. The PARENT prescribes the criteria for action to be taken; and the PARENT prescribes the action to be taken by the KID's computer system. The PARENT-prescribed criteria comprises a list of BAAE-type states, e.g., emotional state. If one of these states is induced in the KID when they accessed the content from a Social Media Platform, then it triggers a PARENT-prescribed action. The list of possible PARENT-prescribed actions comprises at least the following: blocking the content, informing the KID, the PARENT, the Social Media Platform, School Authorities, Law Enforcement, Health Authorities of the KID's state, and archiving the event-data. The archived event-data is analyzed to discover harmful behavior patterns, e.g., lack of self-control, exhibited by the KID vis-à-vis social media content.