1253results about "Knowledge representation" patented technology

A computer system implements a unified framework integrating an adaptive elastic funnel (AEF) with a convergent intelligence fabric (CIF) for flexible and contextualized multi-agent AI and human collaboration at scale. The system provides a universal multi-modal key-value subsystem for sharing partial computations across agents, implements a hybrid greedy / non-greedy placement strategy for dynamic memory management, orchestrates dynamic computational workflows and tensor workflows using hierarchical tensor-fragment scheduling, enables cross-agent orchestration with policy-based privacy preservation, and incorporates quantum-resistant secure memory enclaves. The architecture supports continuous learning without catastrophic forgetting, compositional reasoning across modalities, and secure task execution in distributed environments. This integration enables unprecedented computational efficiency, secure collaboration, and adaptive intelligence in high-dimensional decision-making environments while supporting incremental adoption through modular interfaces.

The invention discloses a credit text analysis method and a credit object auditing method and device, and relates to the technical field of knowledge maps. A specific embodiment of the method comprises the steps of obtaining a target credit card text, wherein the target credit card text comprises at least one target element; a target semantic template corresponding to the target credit document text is obtained, the target semantic template comprises a statement template and an upper sharing hierarchy of the statement template, the statement template comprises at least one target slot position, and the target slot position corresponds to the target element; obtaining a full-link slot position of the target credit card text according to the statement template and the upper sharing hierarchy; and matching the target semantic template with the target credit text to obtain a target slot value corresponding to the full-link slot position. According to the implementation mode, the slot position and the slot value corresponding to the element in the credit card can be automatically obtained, and the related content in the credit card text can be analyzed.

A method includes receiving state inputs pertinent to a system and determining prospective instructions for the system based on at least one of the state inputs. For each prospective instruction, the method includes simulating execution of the prospective instruction to predict at least one corresponding predicted outcome for execution of the prospective instruction and executing a plurality of evaluators. Each evaluator has a corresponding objective and is configured to, for each prospective instruction: evaluate the prospective instruction based on whether the at least one corresponding predicted outcome for execution of the prospective instruction satisfies the corresponding objective of the evaluator; and output an evaluation of the prospective instruction. The method also includes selecting a suggested instruction from the prospective instructions based on the evaluations of the prospective instructions of one or more evaluators, and suggesting execution of the suggested instruction for the system.

Methods, apparatuses, system, devices, and computer program products for AI-driven data aggregation and analysis for entity scoring and assessment are disclosed. In a particular embodiment, a controller generates an entity profile for an entity by iteratively retrieving data related to an entity, the data including structured data and unstructured data, determining a structure for the unstructured data by applying the retrieved data to the AI model, the structure including a plurality of fields, and populating one or more fields of the entity profile with a respective value that is extracted from the retrieved data using the AI model. The controller stores the entity profile in a database comprising a plurality of entity profiles corresponding to different entities. The controller generates, using the AI model, a score for the entity in relation to an attribute based on an AI-driven analysis of the plurality of entity profiles.

Techniques for knowledge discovery based on indirect inference of association are presented. A data management component (DMC) can determine and extract, in a structured format, entities, relationships between entities, and concepts relating thereto in documents, based on analysis of information in the documents and / or keywords relating to concepts, to generate an association inference model. Using artificial intelligence techniques, DMC can embed the entities and relationships to a common representation to generate and train a scoring model that can be used to evaluate and score similarity strength between entities, including entities that do not have a known relationship, and can predict or infer relationships, including indirect relationships, between entities or between concepts. In that regard, DMC or user can evaluate concept-level scores to determine a level of relationship between concepts. DMC can feedback information from the scoring model or evaluation to update the association inference model.

This disclosure provides a method for assessing the health status of bearings. The method includes: receiving multimodal data and prompting information about the bearing; performing feature extraction on the multimodal data using a first set of agents to generate a feature set about the multimodal data; performing a health assessment on the bearing using a second set of agents based on the feature set and prompting information about the multimodal data to generate multiple health assessment results about the bearing; and combining the multiple health assessment results using a third set of agents to generate a health assessment report about the bearing. According to this bearing health status assessment method, users can input multimodal data to perform bearing health status assessments, thereby improving the flexibility of the bearing health status assessment method. Furthermore, the bearing health status assessment method of this disclosure, by leveraging the reasoning capabilities of a large language model, can better understand user questions, thereby providing better maintenance suggestions and other outputs.

One embodiment sets forth a technique for fine-tuning a machine learning model to perform physical reasoning. According to some embodiments, the method can include the steps of obtaining simulation annotations that describe interactions among simulated objects within a physics-based environment and one or more question templates, each question template defining a different parameterized reasoning query; generating, based on the simulation annotations and the one or more question templates, a plurality of question-answer pairs that represent physical reasoning examples; formatting the question-answer pairs into natural-language data compatible with the machine learning model; and fine-tuning the machine learning model based on the natural-language data.

A system generates domain-expert agents for a particular domain. The system generates a hierarchy of topics for the domain by forming structured query inputs and requesting a machine learning-based language model to produce topics and associated sets of search terms. A knowledge store is built for the domain by identifying, for each topic, facts represented symbolically and / or in natural language. For at least one topic, programs comprising instructions for executing domain-specific procedures are generated and stored as symbolic and / or natural language programs. The knowledge store is evaluated to determine an expertise level for a domain-expert agent. If the knowledge store meets a threshold of knowledge, a deployment package is created comprising the knowledge store, and the domain-expert agent is deployed. The deployed agent utilizes the knowledge store together with a machine learning language model to address and solve problems specific to the domain.

Systems, methods, and apparatuses for providing dynamic, prioritized spectrum utilization management. The system includes at least one monitoring sensor, at least one data analysis engine, at least one application, a semantic engine, a programmable rules and policy editor, a tip and cue server, and / or a control panel. The tip and cue server is operable utilize the environmental awareness from the data processed by the at least one data analysis engine in combination with additional information to create actionable data.

Techniques for prompt template optimization with language models are described. In some examples, a prompt template optimization request to optimize a generative artificial intelligence model prompt template is received, the prompt template optimization request including an initial prompt template and an indication of a selected function, the selected function to implement at least a portion of a prompt template optimization workflow. The prompt template optimization workflow is processed with the selected function, the prompt template optimization workflow including one or more iterations of generating, evaluating, and selecting prompt template variants based at least in part on the initial prompt template to yield a final prompt template. The final prompt template is output.

Embodiments described herein provide an AI guardrail system that integrates an agentic compliance evaluation framework with an agentic governance ontology framework to automate AI risk assessment and management. For the compliance evaluation framework, embodiments provide a Digital Risk and Compliance Officer (DRCO) agent built upon an LLM to verify AI compliance in an AI system based on a knowledge base managed by an AI governance ontology. For the governance ontology framework, embodiments provide an AI Governance Ontology (GO) agent built upon two LLMs that generate and update a knowledge base representing concepts, relationships, and metadata related to the governance of AI systems based on AI system generated conversation transcripts.

A computer-implemented method includes: receiving, by a computing device, an input prompt; routing, by the computing device, the input prompt to one or more restructuring routes; restructuring, by the computing device, the input prompt to provide a restructured input prompt to reduce tokens; evaluating, by the computing device, the input prompt to the restructured input prompt to determine that thresholds are met; and sending, by the computing device, the restructured input prompt that meets the thresholds to a large language model.

A data processing system implements performing a root cause analysis that includes identifying a first anomalous signal data predictive of a root cause of a first anomaly in signal data received from a computing system, analyzing the sub-signals of the first anomalous signal data to generate labeled training data, training a gradient boosted tree model using the labeled training data, generating a decision tree based approximating a predictive performance of the gradient boosted tree model, determining insights data predictive of the root cause of the first anomaly based on the gradient boosted tree model and the decision tree, aggregating the insights and analyzing the aggregated insights data to determine a predicted root cause for the first anomaly, determining a confidence level associated with the predicted root cause, and categorizing the predicted root cause into one of a plurality of categories based on the confidence level.

Methods, apparatuses, system, devices, and computer program products for automating asset trading based on composite AI-generated data is disclosed that includes a controller receiving from a user, investor-specified risk parameters and portfolio criteria for a portfolio; retrieving from an entity database, composite AI-generated data; based on scoring preferences of the user, utilizing, the scoring data to select from the plurality of entity profiles, a first set of entity profiles corresponding to candidate assets that satisfy the investor-specified risk parameters and the portfolio criteria; for each asset associated with an entity profile in the first set of entity profiles, determining based on the investor-specified risk parameters and the portfolio criteria, an asset allocation for the portfolio; and transmitting instructions to one or more trade fulfillment backends to perform one or more trades in accordance with the determined asset allocations.

Systems and methods are disclosed for processing multidimensional data using machine-learning models to classify an entity, predict metrics, and / or generate transcripts. The method includes receiving multi-dimensional data that include a first value of a first dimension indicating a comparison of an entity with a group of entities, and a second value of a second dimension indicating an attribute of the entity irrespective of the group of entities; inputting the multi-dimensional data to a first machine-learning model to output a prediction value determined based on applying first and second weights to the first and second values, respectively; and upon determining the prediction value satisfies first threshold associated with an adverse event: filtering historical text data associated with the entity to exclude data that satisfies second threshold associated with acceptable operations; and generating, by inputting the filtered historical text data into a second machine-learning model, a transcript of recommended actions.

Techniques for training and using machine learning models for sensitivity detection. A method for sensitivity detection training includes fine-tuning a language model by iteratively applying the language model to prompts and adjusting weights of the language model. The prompts indicate classifications for a set of first resources and characteristics of an entity. The fine-tuned language model is queried with respect to classifications of a set of second resources. The fine-tuned language model is queried using prompts indicating the second classifications and data indicating characteristics of an entity, where outputs of the language model include a sensitivity for each of the second classifications. Training data including the second classifications is labeled based on the sensitivities output by the language model. A sensitivity detection machine learning model is trained using the labeled training data set such that the trained sensitivity detection machine learning model is configured to output sensitivities for resource classifications.

Techniques for building and maintaining model webs are described. In some examples, a model web is built by selecting models for the model web from one or more available model types based on at least one or more of availability, tensor information, and compute type, instantiating synapses between the selected models to form the model web and updating information regarding availability of the selected models of the model web to indicate being in use.

A building management system for monitoring and controlling temperature, pressure, and humidity (TPH) of a building includes one or more processing circuits configured to obtain trend data comprising TPH values of a target area of the building over a time period, compare the trend data against a range of compliant TPH values defined by a compliance standard, obtain a schedule for the target area indicating a scheduled event within the target area or requested TPH settings for the target area, and in response to a determining that the trend data do not satisfy the compliance standard, operate HVAC equipment of the building to affect at least one of the temperature, the pressure, or the humidity of the target area to satisfy the compliance standard based on the schedule for the target area.

Aspects of the disclosure relate to providing a flexible and automated system for automatically detecting when emails include harmful content, flagging the emails, providing interactive reporting functionality, and providing follow-up enforcement actions to protect users. A computing platform may intercept an email in transit to an email server. Subsequently, the computing platform may analyze the email and generate at least one unique link for reporting suspicious content associated with the email. Next, the computing platform may generate an email warning tag comprising text information and the at least one unique link for reporting the suspicious content associated with the email. Then, the computing platform may inject the email warning tag into the email to produce a modified email comprising content from the email and the email warning tag, and may send the modified email to the email server.

Disclosed herein are embodiments of systems, methods, and products comprises an analytic server, which automatically manages appointment scheduling. The analytic server receives a customer request to schedule an appointment. The analytic server determines the required data from both customer and service provider for making the appointment. The analytic server retrieves customer data comprising requested service attributes, user preferences, users attributes from internal database and external data source. The analytic server retrieves service providers' data comprising provider service attributes, providers' attributes from internal database and external data sources. The analytic server accesses external data source by web crawling various websites. The analytic server executes an artificial intelligence model to predict user preferences and needs. The analytic server determines potential service providers best matching the customer's input or predicted preferences. The analytic server generates an appointment for each matching service provider and transmits an electronic message comprising the appointments to customer device.

This invention provides an intelligent operation and maintenance question-answering method, system, and device based on a large model. The method includes: performing a hybrid retrieval on a user question based on a preset knowledge base to determine the set of text fragments most relevant to the user question; inputting the user question and the set of text fragments into a large model for decomposition, generating a set of independent sub-questions; identifying the semantic associations and logical relationships between the sub-questions in the sub-question set, generating a causal graph; performing a secondary hybrid retrieval on each sub-question to determine the subset of text fragments most relevant to the sub-question, forming a mapping relationship set; inputting the mapping relationship set and the causal graph together into the large model for correction, determining a corrected set of sub-questions; and inputting the corrected set of sub-questions, the causal graph, and the user question together into the large model for reasoning, generating an operation and maintenance answer that conforms to the topological order of the causal graph. This invention ensures the coherence and reliability of the question decomposition and correction process, thereby improving the accuracy and completeness of complex operation and maintenance question-answering.

This invention relates to a method, apparatus, and medium for automatically generating test cases by integrating multi-source human programming experience. The method includes: a preparation step: based on a target programming problem P, searching for existing correct human programming code for the target programming problem P, as well as human error submission code and human correction code for each error submission; a multi-dimensional analysis step: obtaining a first analysis result generated by a large language model; a difference analysis step: obtaining a second analysis result generated by the large language model; and a test case generation step: based on the first and second analysis results, generating executable test script construction prompts to obtain a test script generated by the large language model, and running the script to generate test cases for the programming problem P. Compared with existing technologies, this invention has advantages such as significantly improving the discriminative power of test cases.

A method, apparatus, non-transitory computer readable medium, and system for data classification include obtaining input data and a policy graph. The policy graph includes a decision node indicating a machine learning classifier. Embodiments then generate, using the machine learning classifier, a classification result based on the input data and the decision node. Subsequently, embodiments generate a decision label for the input data based on the classification result.

The application relates to a knowledge-driven multi-agent cooperative reactor scheme demonstration system and a design method, medium and equipment thereof, the system comprising: a knowledge internalization unit for converting unstructured and semi-structured documents into a dynamic knowledge source which can be queried and utilized by a model in real time; a knowledge structuring unit for extracting core entities, relationships and attributes from the knowledge source and constructing a professional knowledge graph of a reactor design field; and a knowledge application unit for constructing intelligent agents and a collaborative working mechanism of multi-professional intelligent agents based on the knowledge source and the knowledge graph, and constructing an intelligent question and answer interface. Through automatic knowledge management and intelligent agent collaborative work, the application significantly reduces the time and effort of manual intervention and improves the efficiency of reactor scheme demonstration.

A system and a method for data protection assessment using threat modelling and adaptive optimization is disclosed. The system (100) comprising a processor (105) and memory (110) with instructions to receive a data access request (345) from entities (120) via authenticated digital interface (125) including structured and unstructured data categories (350), metadata parameters (355), initiating contextual analysis to establish baseline compliance parameters (360). The system analyses data category and metadata using adaptive classification logic and correlation metrics across stored data lineage, distinguishing privacy -critical from non- sensitive data. Threat modelling (130) evaluates correlations among privacy attributes (365), metadata, operational parameters, and regulatory requirements to identify vulnerabilities (375) and exposure points. A composite risk index (135) is derived, forming a structured risk profile (140) with quantified scores (380). Compliance recommendations (150) are generated, safeguard parameters (155) are derived, authorization validated by a compliance authority (160), and a compliance trace (388) is recorded.

The application discloses a knowledge graph construction method and device of a promotion path, wherein the method comprises the following steps: step S1, combing core concepts related to the promotion path and semantic relations thereof, constructing a basic knowledge framework, and creating an ontology model based on the basic knowledge framework; step S2, setting mapping rules between ontology data and a graph database, and mapping the ontology model to the graph database; step S3, collecting multi-source information, performing knowledge extraction on the multi-source information to obtain knowledge triples and constructing entities, and constructing a knowledge graph by comprehensively combining the ontology model and the entities; and step S4, performing retrieval and correlation analysis based on the knowledge graph, forming promotion path instances, dynamically updating the promotion path instances to the knowledge graph, and obtaining a promotion path knowledge graph. The promotion path knowledge graph is obtained based on dynamic updating of the knowledge graph.

The application provides a peak detection and qualitative method for full two-dimensional mass spectrum data based on a multi-modal model and related equipment, and comprises the following steps: performing supervised fine-tuning training on a pre-trained multi-modal large language model by using a mixed data set, performing reinforcement learning training on an initial peak detection and qualitative model by using a reinforcement learning fine-tuning data set, inputting preprocessed multi-modal input data into the peak detection and qualitative model to perform data extraction, and obtaining a corresponding data extraction result; by combining the multi-modal large language model and the reinforcement learning fine-tuning training, the method can accurately detect target peaks when processing full two-dimensional mass spectrum data, and can efficiently analyze compounds; the use of a low-rank synthetic data set helps to enhance the accuracy of model data processing; in the reinforcement learning fine-tuning stage, the structure generation strategy of the model in unknown compound analysis is optimized, so that the model can cope with complex situations of unknown compounds and optimize the structure generation capability.

This application relates to a method and system for intelligent policy analysis and risk monitoring based on a family graph. The method includes: acquiring multi-source policy data and performing structured processing; constructing a family policy relationship graph based on the relationships between policyholders and insured persons in the dataset; acquiring a predefined risk scenario library, containing multiple preset risk scenarios and corresponding coverage types for each scenario; calculating the total coverage amount based on the family graph by identifying related policies for each risk scenario; acquiring a risk exposure benchmark value, comparing the total coverage amount with the benchmark value, and generating a quantitative result of the coverage status; combining a pre-built insurance product knowledge graph, generating coverage optimization suggestions based on the quantitative results; and re-executing the calculation and generation operations in response to user lifecycle events or periodic assessment instructions, and generating risk warning notifications based on the analysis of result changes. This application achieves intelligent analysis of family policies, dynamic risk monitoring, and coverage optimization, improving risk management efficiency.

The application discloses a kind of federal positioning method based on individualized patch and dynamic fusion, belong to indoor positioning and federal learning technical field.When applying federal learning to indoor Wi-Fi fingerprint positioning of Internet of Things, global model is often difficult to capture local data feature preference, local training cost is high and poor adaptability alone, the present application is by setting individualized patch and dynamic parameter fusion mechanism, reduce the influence of above-mentioned problem on positioning performance.The algorithm configures private position decoder as individualized patch for each local client, extracts local data features in training and uploads to server, server calculates the similarity of data distribution between clients to generate combination weight, for local clients with similar data distribution, individualized patch can selectively linearly combine with the decoder of the local model with higher relevance, avoid local model overfitting, improve the positioning performance of global model in local scene.

The application discloses an industrial internet of things optimization system and method fusing digital twinning and federated learning, belongs to the technical field of industrial internet of things, and aims to solve the problems of physical mapping inconsistency, model drift and low collaborative efficiency under resource limitation. The system comprises a physical entity layer, a digital twinning mapping layer, a distributed federated learning layer and a cross-domain resource scheduling layer. The digital twinning layer ensures consistency between virtual and real; the federated learning layer realizes model collaboration by using physical constraint alignment and adaptive aggregation; and the scheduling layer optimizes calculation, communication and energy resources. The application realizes deep fusion of mechanism and data driving, suppresses model drift, improves resource utilization and system adaptive capacity while ensuring privacy.