1522 results about "External data" patented technology

An example operation includes one or more of ingesting profile data of a user from an external data source, identifying a plurality of features of a profile hosted by the external data source based on an execution of an artificial intelligence (AI) model on the ingested profile data of the user from the external data source, identifying a feature from among the plurality of features of the profile hosted by the external data source that is not connected to a local user profile of the user, and displaying a connection request on a user interface, wherein the connection request includes a link to a page associated with the identified feature hosted by the external data source.

The present invention relates to a digital twin processing method and system, and a cloud platform. The method comprises: acquiring production system elements, carrying out abstraction definition and parameterization description on the production system elements by means of digital-to-analog knowledge fusion of a mathematical model and a mechanistic model, so as to construct a digital twin ontology model; analyzing and reconstructing model data to obtain a mapping model of which object variables can be directly accessed and operated by a collective motion control method, so that the model is visualized at the cloud; and using an external data source to drive parameter update and operation matching of the model by means of a motion control method, so as to complete cooperative deployment and synchronous evolution of an actual physical device and the model in the production process on a cloud server. According to the present invention, a model is constructed by means of digital-to-analog knowledge fusion of a mathematical model and a mechanistic model, the model is mapped to achieve motion visualization, model parameter update and operation matching on the cloud are achieved, and then cooperative deployment and synchronous evolution of a physical device and the model are completed.

The invention provides a supply chain sales anomaly detection and root cause analysis system and method fused with a knowledge graph, and the system comprises a demand collection and preprocessing module which is used for connecting an order system, a supply chain system, a customer relationship management system and an external data source, and completing the data cleaning, entity analysis and feature extraction; the supply chain knowledge graph construction module is used for defining an entity type and a relationship type; the real-time anomaly detection module is used for accessing a sales index data stream, performing anomaly detection in combination with lightweight filtering and a graph neural network model, and calculating node and global anomaly scores; and the visual report generation module is used for automatically generating a visual report. According to the method, the dynamic supply chain knowledge graph is constructed, the graph neural network is applied, multi-source heterogeneous data is deeply fused, the complex dependency relationship between entities is effectively captured, the accuracy and timeliness of sales anomaly detection are remarkably improved, automatic positioning of abnormal root causes and evidence chain tracing are achieved, and the analysis efficiency is greatly improved.

A system and method for providing a digital financial advisor using fine-tuned large language models is disclosed. The system includes a data advisor application, data fusion suite advisor engine, human advising engine, a knowledge base and large language model (LLM) fine-tuning engine to create specialized language models (SLMs) that mimic specific human financial advisors. Multiple digital avatars embodying the appearance and communication style of human advisors are generated. The system processes user and client profile data, along with advisor-specific information, to provide personalized financial advice. It handles client queries, escalating complex issues to human advisors when necessary. The digital advisor system communicates with profile datastores, user devices, and external data sources for comprehensive financial analysis. Continuous learning capabilities allow the system to improve its performance based on interactions and feedback, combining AI efficiency with personalized human-like advisory.

The invention discloses a supply chain intelligent management system based on dynamic collaborative optimization, and relates to the technical field of hotel supply chain management, and the system comprises a supply chain intelligent management platform which is in communication connection with the following modules: a multi-modal data sensing module, a multi-modal data processing module and a multi-modal data processing module. The multi-modal data acquisition module is used for acquiring multi-modal data in a hotel through a LoRaWAN + BLE hybrid sensor network, accessing an external data source and acquiring real-time information through an API (Application Program Interface); and the dynamic demand prediction module captures a time sequence trend through bidirectional LSTM based on an ASTGNN model. According to the method, external dynamic data such as social media public opinions and weather are fused, the multi-modal data analysis technology is combined, the accuracy of demand prediction is remarkably improved, the ASTGNN model is utilized, the time sequence trend and the demand association between the branches are analyzed in combination with bidirectional LSTM and GCN, and the feature weight is dynamically adjusted, so that the demand prediction error rate is greatly reduced, and the demand prediction efficiency is improved. A more reliable demand prediction basis is provided for enterprises, and optimization of inventory management and production plans is facilitated.

Systems and methods for detecting and mitigating prompt injection attacks on a generative LLM are disclosed. A deployment scenario is considered, in which the generative LLM supports a task automation function. Prompts are received and interpreted by the generative LLM, and outputs from the generative LLM are used to trigger automation actions. The prompts are constructed based on a combination of user input and external data and are, therefore, vulnerable to prompt injection attacks though manipulation of the external data. To mitigate this risk, a separate discriminative classification, decoupled from the generative LLM, engine is configured to identify malicious prompts, and filter out any malicious prompts before they reach the generative LLM.

The invention discloses a food safety sampling inspection data verification method and system and a computer storage medium thereof, particularly relates to the technical field of food safety supervision informatization, and is used for solving the problem of false consistency caused by silent failure when an external data source fails in an existing verification system. Field-level binding is realized by establishing a dependency mapping table of a sampling inspection system and an external data source; constructing a service response causal graph model, and dynamically marking a failure data source based on the deviation between the anti-fact response and the actual response; analyzing the verification field set and positioning a key verification field; degrading the weight of the key field to be below a threshold value and activating a manual auditing flag bit; scanning a preset coupling relationship between fields, freezing associated fields which do not trigger rules, verifying and constructing a conflict topology; when the check engine is executed, dynamically skipping the weight reduction field, suspending the freezing field, and generating a to-be-rechecked report in combination with the artificial flag bit and the topological graph; error data transmission is blocked from the source, and the food safety risk missed judgment rate is remarkably reduced.

The invention provides a semantic enhancement and dynamic completion method and system for a power market data graph. The method comprises the following steps: acquiring target entity information and external data of a power market data graph by responding to a data query instruction; converting external data into structured semantic tags, then traversing the topological structure, establishing a mapping relationship between the tags and entity attributes, and combining query instruction influence logic to calculate topological integrity indexes and mark missing entities or correlate to form a missing tag set; establishing attribute mapping based on a target entity and external data, generating a semantic vector fusing cross-domain features, and projecting the semantic vector to a topological space according to a missing mark set to form cross-domain feature topological nodes and connections; embedding a missing position to reconstruct a topological structure, and checking logic connectivity; the dynamic conduction chain is analyzed, and a feedback interface containing the semantic relation and the completion structure is generated to respond to query. According to the method, conduction logic closed-loop verification of the missing elements in the power market data graph is realized.

A distributed generative artificial intelligence (AI) reasoning and action platform that utilizes a cloud-based computing architecture for neuro-symbolic reasoning. The platform comprises systems for distributed computation, curation, marketplace integration, and context management. A distributed computational graph (DCG) orchestrates complex workflows for building and deploying generative AI models, incorporating expert judgment and external data sources. A context computing system aggregates contextual data, while a curation system provides curated responses from trained models. Marketplaces offer data, algorithms, and expert judgment for purchase or integration. The platform enables enterprises to construct user-defined workflows and incorporate trained models into their business processes, leveraging enterprise-specific knowledge. The platform facilitates flexible and scalable integration of machine learning models into software applications, supported by a dynamic and adaptive DCG architecture.

Systems, methods, and devices to detect unauthorized third-party connections within a network infrastructure, such as by analyzing network traffic data using fuzzy matching and machine learning techniques. One aspect includes receiving network traffic data comprising records of communication events involving network identifiers, determining communication relationships between network entities identified by the network identifiers, accessing entity identifiers associated with known third-party systems, and determining associations between the network identifiers and the entity identifiers using a fuzzy matching process. Other aspects include identifying communication relationships involving the third-party systems based on the associations and detecting unregistered or unknown third-party connections within the network infrastructure. Further aspects include normalizing identifiers, computing string similarity metrics, assigning confidence scores, incorporating external data sources, building network association patterns, comparing current patterns to baseline patterns to detect anomalies, and updating security policies or firewall rules in response to detected anomalies. Additional aspects are provided.

A colorectal lesion multi-modal classification method based on pathological attention and multi-instance learning constructs an efficient automatic diagnosis model by fusing visual features and a textual prototype defined by pathology experts. The method comprises the following steps: collecting histopathological image data, segmenting the histopathological image data into standardized image blocks, and extracting visual features by using a pre-training feature extraction network after color standardization and noise processing; a multi-instance learning framework and a pathological attention mechanism are combined, feature space distribution of a text prototype is adjusted in a self-adaptive mode through a dynamic prototype optimization module, and optimization targets of visual clustering and cross-modal semantic alignment are balanced by adopting a gradient perception double-loss dynamic weighting strategy; and after the model is trained in stages, the generalization performance is verified in an external data set. According to the method, the classification precision is remarkably improved, the method can adapt to dyeing difference and tissue heterogeneity without pixel-level labeling, the accuracy rate in cross-center verification is superior to that of an existing reference model, and the efficiency of pathological diagnosis is greatly improved.

The invention relates to the technical field of intelligent pianos, in particular to a piano automatic playing driving method and system, and the method comprises the steps: obtaining first external data which is used for describing the position change of keys at first playing time; converting the first external data into discrete data, wherein the discrete data is associated with a time sequence label; the discrete data is sent to a main control module, and the main control module forwards the discrete data to a control unit of the exercise device; monitoring second external data of the key at the first moment, and searching a corresponding discrete position through time sequence label association according to the first moment; and calculating a difference value between the second external data and the discrete position, and outputting an adjusting current value through the difference value by adopting a closed-loop control algorithm. The invention provides a lossless built-in driving technology and a driving method, and based on the driving method, a low-cost and miniaturized exerciser can be adopted, so that automatic playing of the piano keys is realized under the condition that the original framework of the piano is not influenced.

The invention belongs to the technical field of network security early warning, and discloses a network security threat information early warning method and system based on big data, and the method comprises the steps: collecting and integrating internal data and external data related to network security based on the big data, carrying out the natural language processing and analysis of unstructured threat information, and constructing a threat entity association relationship through a knowledge graph; constructing an adaptive threat model based on network security threat information of big data and deep learning, and using a convolutional neural network CNN to identify an abnormal mode in encrypted traffic; establishing a user-device behavior baseline through a long short-term memory (LSTM) network; predicting a threat diffusion path in combination with a graph neural network GNN, developing a threat risk quantitative model, and calculating an information risk value by comprehensively considering an attack success rate, an asset value, an influence range value and a response delay; and performing risk prediction on the network security threat information based on the calculated information risk value in combination with calculation and analysis of the information risk attenuation factor.

The invention relates to a vertical domain relation extraction method and device, electronic equipment and a storage medium, and the method comprises the steps: constructing a knowledge base of a target vertical domain, carrying out the entity relation extraction of a target text through combining with the knowledge base of the target vertical domain, and obtaining a first entity relation set; performing entity relationship extraction on the target text by utilizing a trained entity relationship extraction model to obtain a second entity relationship set, and performing correction processing on the second entity relationship set by utilizing a knowledge base to obtain a fourth entity relationship set, and screening out a target entity relationship set from the first entity relationship set and the fourth entity relationship set in combination with the knowledge base of the target vertical field. By introducing the vertical field knowledge base, external data support is provided for a large model, the defects of the prior art in the aspects of long-tail relation processing and vertical field reasoning capability are overcome, and meanwhile, the filtering capability of noise samples is enhanced and the extraction quality and efficiency are improved by combining a combined comparison screening mechanism, a knowledge base automatic updating mechanism and a health management mechanism.

The invention belongs to the technical field of remote sensing inversion, and aims to provide an automatic remote sensing inversion method and system, electronic equipment and a program product. The method comprises the following steps: acquiring full-modal remote sensing data and multi-field external data; performing space-time alignment processing on the multi-field external data to obtain multi-field space-time aligned external data in space-time matching with the fused remote sensing feature data; performing cross-modal feature fusion processing on the full-modal remote sensing data to obtain fused remote sensing feature data; inputting the fused remote sensing feature data into a preset remote sensing inversion model to obtain an initial remote sensing inversion result; and performing local correction on the initial remote sensing inversion result according to the external data after multi-field space-time alignment to obtain a final remote sensing inversion result. According to the method, high-precision and automatic inversion of the surface physical parameters in the remote sensing image can be realized, and the method is suitable for application in a complex environment.

The invention discloses an intelligent pipe network topological structure restoration analysis system, and relates to the technical field of intelligent pipe network management, the intelligent pipe network topological structure restoration analysis system comprises a design management platform, and the design management platform is in communication connection with a topological structure data acquisition module, a topological structure analysis module, a pipe network fault diagnosis module, a restoration scheme generation module and a pipe network operation monitoring module. According to the method, the pipe network topological structure model is constructed by integrating various sensors and external data sources, branches, loops or special-shaped connection structures in a complex pipe network are accurately recognized, the judgment capacity of the pipe network connection relation and the flow direction is improved, and the fault diagnosis efficiency is improved by combining the pipe network topological structure model and the flow direction analysis result and utilizing the pre-constructed fault diagnosis model. According to the method, the pipe network fault can be quickly and accurately diagnosed, the fault type and position can be quickly identified by integrating the multi-source data, extracting the fault features and matching the fault features with the preset fault mode, the misdiagnosis rate is reduced, and the fault diagnosis efficiency and accuracy are greatly improved.

The invention provides an industrial carbon footprint dynamic prediction method fusing multi-source data, which comprises the following steps: step 1, collecting multi-source data, core data including electricity consumption, converted standard coal consumption, key industrial process product yield and carbon emission, and three types of extended data including new energy consumption data, process upgrading data and external data; 2, performing dynamic modeling according to the multi-source data obtained in the step 1, including dynamic carbon emission factor calculation and a process-environment coupling equation; 3, an industrial carbon-energy-economy multi-objective optimization model of dynamic coupling new energy consumption is constructed, minimization of carbon emission and operation cost is taken as an objective function, and constraint conditions comprise production requirements, energy supply and policy compliance; and 4, performing optimization solution on the target function of the multi-target optimization model by adopting an NSGA-II algorithm, and outputting a carbon emission-cost tradeoff curve.

The invention provides a food information tracing method and system based on Internet data, and relates to the technical field of industrial Internet, and the method comprises the following steps: generating an initial batch identifier of a to-be-produced product according to a product production plan, obtaining real-time external data of an associated region, and storing the real-time external data in a database; when it is judged that a batch adjustment event occurs in the production process of the to-be-produced product, based on real-time state data and real-time external data of a production line, dynamic batch decision is carried out to generate at least two sub-batch identifiers, and a tree-shaped association relationship is established between the sub-batch identifiers and the initial batch identifier; and receiving sub-batch detailed data of each sub-batch identifier, uploading the sub-batch detailed data to a block chain network, storing the batch topological relation in a block chain main chain, storing the sub-batch detailed data in a corresponding sub-chain node, responding to a tracing request of a user terminal, and according to the tree-shaped association relation, tracing the sub-batch data of each sub-batch identifier. And retrieving the full life cycle data of the corresponding target batch in the block chain network. The method overcomes the problem of tracing information splitting caused by batch splitting.

The invention discloses a fire risk accurate prediction method and system based on a multi-source data fusion knowledge graph, and belongs to the technical field of machine learning, and the method comprises the following steps: accessing a multi-source heterogeneous data set, integrating sensor data, geographic information, historical data and external data, and obtaining a multi-source heterogeneous data set; a full-size forest three-dimensional model is constructed by means of FDS and SolidWorks, combustion simulation is carried out, environmental parameters are set carefully to ensure that the result is accurate, and a foundation is laid for construction of a multi-source heterogeneous data set; the method comprises the steps of data acquisition, data preprocessing, noise data cleaning, space-time alignment and feature engineering, multi-source heterogeneous data acquisition and preprocessing are carried out, and forest fire related data covering structured, semi-structured and non-structured types are acquired from a multi-source heterogeneous data set. The accuracy and timeliness of fire early warning are remarkably improved, effective fusion of forest fire multi-source heterogeneous data can be achieved, and transparency and traceability in the data fusion process are ensured.

The invention provides a cellular network flow space-time prediction method and system based on a hybrid neural network, and belongs to the technical field of intelligent communication. The method adopts a layered deep neural network architecture, and comprises a data embedding layer, a space-time coding layer, a feature fusion layer and an output layer. The data embedding layer maps a historical traffic sequence, cross-domain external data and metadata into high-dimensional features; the space-time coding layer is used for respectively fusing one-dimensional causal convolution and a Mama neural network to extract multi-scale time features and densely connecting convolution and a multi-head attention mechanism to capture multi-scale space features through time and space modeling branches; the feature fusion layer realizes adaptive weighted fusion of spatial-temporal features, cross-domain features and metadata features by using a gating fusion mechanism; and the output layer performs linear transformation on the fusion features to generate a final prediction result. According to the method, the spatial-temporal dynamic capture of the service traffic is accurate, the prediction curve is highly fit with the true value, and the accurate prediction of the multi-service traffic of the cellular network is realized.

Systems and methods described herein enable adaptive, threshold-based modification of node maps representing ontologies, knowledge graphs, or code development pipelines using generative artificial intelligence. The disclosed platform can retrieve a node map and generate one or more candidate perturbations that modify nodes or relationships within the node map. The disclosed platform can evaluate the effect of the perturbations by comparing respective outputs against ground-truth data. Perturbations can be automatically determined based on changes in external datasets, compliance policies, or operational requirements. The perturbations can be implemented when a computed perturbation quality value satisfies a threshold quality criterion. As such, the system enables efficient, policy-compliant evolution of relational system architectures in dynamic environments.

The invention discloses an unmanned aerial vehicle dynamic environment perception response system based on an end side LLM, and belongs to the technical field of artificial intelligence and unmanned aerial vehicle application, and the system comprises a visual perception module which collects image external data in real time and outputs identification information; the warning identification module is used for generating a structured label according to the identification information; the LLM reasoning and decision-making module is used for reasoning a pre-deployed LLM model and giving threat assessment and decision-making content; the local knowledge enhancement module pre-constructs a local knowledge base, distinguishes categories of different labels and gives corresponding solutions; the bidirectional network communication module is used for realizing double-end communication between the unmanned aerial vehicle end and the ground end; the visualization and interaction module is used for displaying the image identification information in real time and displaying a target evaluation result; according to the unmanned aerial vehicle dynamic environment perception response system based on the end-side LLM, through fusion of the end-side LLM and visual perception, the high delay problem of cloud dependence is solved, the real-time response capability in a dynamic scene is remarkably improved, and the computing power and bandwidth requirements are reduced.

The invention discloses a balance scheduling system for energy management and a balance scheduling method thereof, and relates to the technical field of energy management scheduling, and the balance scheduling system comprises an energy data acquisition module, an energy demand trend analysis module, an energy supply and demand dynamic prediction module, an energy balance scheduling decision module, a scheduling execution module and a performance monitoring module. After the energy supply and demand dynamic prediction module integrates real-time external factor data of the energy data acquisition module and a long-term historical rule of the energy demand trend analysis module, a reference prediction baseline based on a long-term trend is generated, and then a dynamic correction value is superposed to respond to short-term disturbance. And meanwhile, through an error traceability mechanism for separating a trend baseline from external correction, the module performance can be optimized in a targeted manner, if an error is derived from long-term trend deviation, a trend analysis model is calibrated, and if external data is lagged, priority adjustment of a data acquisition module is triggered, so that system self-diagnosis and prediction precision closed-loop improvement are realized.

The invention belongs to the field of data processing, and particularly discloses a method and system for automatically generating a document according to a template and structured data, and the method comprises the steps: processing an archive template file according to a document object model analysis method, extracting the position, the type and the associated tag of a placeholder, and obtaining a placeholder metadata set; according to the placeholder metadata set, traversing a document hierarchical nesting relationship by adopting a depth-first search algorithm to obtain a hierarchical placeholder mapping table; obtaining the structured data of the document repair record through an external data source, judging whether the structured data field corresponds to the tag in the mapping table according to the hierarchical placeholder mapping table, and if so, generating a matching pair list from the field to the placeholder to obtain a preliminary matching data set; the objective of the invention is to solve the problem of a business scene in which complex nested data and file template placeholders cannot be accurately matched in a document repair record in the prior art.

The invention particularly relates to an intelligent security and protection dynamic risk assessment and prevention and control system based on multi-source data fusion, and the system comprises a data collection module which collects corresponding data and carries out the preprocessing of the data; the data processing module is used for analyzing the data to obtain a risk assessment coefficient; the risk assessment module is used for obtaining a risk level based on the risk assessment coefficient and making a corresponding prevention and control strategy; and a prevention and control decision optimization module. According to the invention, the data acquisition module integrates sensor data, service system data and external data, and realizes omnibearing data coverage of a security scene; real-time data collected by a visual sensor, an environment sensor and the like are combined with business system data of access control, vehicle management and the like, then external information of meteorology, public security, public opinions and the like is fused, feature extraction and weight distribution are carried out by using technologies of deep learning, an entropy weight method and the like through a data processing module, and finally an accurate risk assessment coefficient is obtained. More risk factors are comprehensively considered, and potential safety hazards are effectively identified.

The invention provides a digital employee robot performance intelligent evaluation method and system, and relates to the technical field of data processing, and the method comprises the steps: carrying out the feature extraction of user interaction contents in all scenes; performing fluctuation detection, identifying a time period of abnormal fluctuation, and extracting user request content and robot response content in the time period; restoring the current session path based on a preset business process template, performing path consistency comparison in combination with a response rule of the robot under the path, and generating an abnormal node set if the path is interrupted or the response is offset; extracting a user request format, a robot response behavior and an external data calling state corresponding to each abnormal node, performing abnormal source analysis, calculating user input abnormity, and identifying an attribution probability of abnormity and data interface abnormity; according to the invention, autonomy and accuracy of robot performance intelligent evaluation are improved.

The invention discloses an enterprise information visual management system and method based on data sharing, and belongs to the technical field of enterprise information management.The method comprises the steps that by connecting business systems in an enterprise, user behavior data and business data are collected and preprocessed, and an internal data set is constructed; acquiring data from an external data source, and constructing an external data set through semantic analysis, screening and classification; internal and external data are integrated, a dynamic cleaning rule is formulated, multi-source fusion is realized through data association, data are stored in a classified manner according to business requirements, and a data warehouse with a layered architecture is constructed; meanwhile, a personalized data mart is generated based on user roles and permissions, user demands are automatically predicted, and analysis indexes and dimensions are pre-generated; and finally, constructing a visual analysis platform, generating a preliminary recommendation scheme in combination with the pre-generated content and the user operation track, fusing user-defined setting, and outputting a final visual result in real time, thereby realizing efficient management and visual presentation of enterprise information.

The invention discloses an electric power spot market electric charge settlement system and method fused with a block chain technology, and the system comprises a core settlement layer which is used for storing the electricity price and user identity data, and executing an electric charge settlement rule through a built-in intelligent contract module; the high-frequency transaction processing layer is used for processing a transaction process through an under-chain caching technology and uploading data in the transaction process to the core settlement module; an external data interface is integrated on one side of the high-frequency transaction processing module and used for acquiring electric quantity and load data in real time, and a cross-chain gateway is arranged on the other side of the high-frequency transaction processing module and used for achieving synchronous settlement of electric charge and capital flow. According to the method, the efficiency and transparency defects of a traditional mode are overcome, the performance and energy consumption bottleneck of the block chain technology in an electric power scene is broken through, high efficiency, safety, expandability and environmental protection benefits are achieved, and a credible, real-time and compliant digital settlement infrastructure is provided for an electric power spot market.

Methods, apparatuses, system, devices, and computer program products for customized integrated entity analysis using an AI model are disclosed. In a particular embodiment, customized AI-powered analysis of an entity includes a controller creating a custom attribute for a user to analyze an entity and processing a first set of external data related to the created custom attribute and the entity including structured data and unstructured data retrieved from a first set of structured data sources and unstructured data sources. In this embodiment, the controller augments an AI model using the processed first set of external data and generates using the augmented AI model, one or more metrics for assessing the custom attribute in relation to the entity. The controller also presents to the user the generated one or more metrics for assessing the custom attribute in relation to the entity.

Methods and corresponding systems and apparatuses for configuring user access to data stored in and / or accessed through an external computer system are described. Access permissions can be configured through defining a permission set relative to a proxy entity and assigning the permission set to one or more users. A proxy entity is a local representation of an external data entity. A proxy entity can be a virtual entity in that the proxy entity does not contain underlying data. A proxy entity can, however, include metadata describing its corresponding external data entity. A computer system maintaining a proxy entity can store mapping information linking the proxy entity to an external data entity. The mapping information and the permission set can be used to determine an access permission relative to the external data entity and to communicate this access permission to the external computer system so that access can be provided accordingly.