39 results about "Joint analysis" patented technology

The invention provides a service data security protection method and system for digital enterprise management, and the method comprises the steps: obtaining an access request sequence initiated by a target user at a service operation terminal, generating a dynamic access control strategy vector according to a historical behavior track associated with an access operation identifier, and carrying out the dynamic access control strategy vector; analyzing the dynamic access control strategy vector through a strategy decoder, and generating a real-time access permission instruction; based on the real-time access permission instruction, performing homomorphic encryption mapping on a sensitive data segment in the request context information to generate a ciphertext transmission channel, and performing security parameter synchronization on the ciphertext transmission channel and the cross-department conjoint analysis model; and calling a federated learning framework to carry out multi-modal feature fusion on the real-time operation flow of the service operation terminal, generating an abnormal operation confidence score, triggering a cross-level interception protocol when the abnormal operation confidence score exceeds a dynamic threshold, and rolling back the current session state to a security baseline version. According to the invention, the accuracy and response timeliness of anomaly detection can be improved.

The invention discloses a network security event analysis method and system based on multi-agent decision, and a storage medium, and the method mainly comprises the steps: carrying out the deep investigation and judgment through a deep decision agent based on a reinforcement learning network after an abnormal alarm of a monitoring platform is received, triggering the deep investigation when the calculated deep investigation probability is greater than a threshold value, and carrying out the early warning of the network security event. Calling an LLM-based sub-expert agent to carry out deep analysis aiming at specific features; and according to the analysis result of the sub-expert agent and the risk information provided by the risk information platform, the LLM-based comprehensive risk decision assessment agent performs conjoint analysis decision to generate a network security situation comprehensive assessment report. According to the method, the processing analysis capability of massive and heterogeneous security data and the recognition capability of complex attack paths are remarkably improved, and the discovery capability of unknown threats and the intelligent level of security operation and maintenance are effectively enhanced by introducing reinforcement learning optimization decision and the deep analysis capability of LLM.

The invention provides a dynamic threshold anomaly detection method and system based on deep semantic reconstruction and an isolated forest, and belongs to the technical field of crossing of network security and artificial intelligence. A sliding window segmentation and dynamic zero padding mechanism is adopted to unify texts of any length into standardized segmented data; the method comprises the following steps: constructing a nine-layer depth auto-encoder (four encoding layers, a middle layer and four decoding layers), learning multi-dimensional semantic features of normal data, calculating a reconstruction error matrix of each data segment, extracting error distribution statistics (mean value, variance, skewness and kurtosis), performing unsupervised clustering on the extracted error features by adopting an improved isolated forest algorithm, and obtaining a clustering result; a dynamic detection threshold value is automatically generated, and after semantic reconstruction and feature analysis are carried out on real-time data, whether the real-time data are abnormal or not is judged in combination with the dynamic detection threshold value; a self-adaptive anomaly detection model is constructed through conjoint analysis of a dynamic threshold generation mechanism and deep semantic features, and dependence on artificial experience is reduced while high detection precision is kept.

The invention discloses a flight data safety management method applied to low-altitude economy, and relates to the technical field of data management, and the method comprises the steps: obtaining heterogeneous monitoring data of an infrastructure and IT equipment in real time; calculating the health degree score and the fault risk probability of each logic area in the low-altitude economic flight data center in real time; based on the health degree score and the fault risk probability, root cause analysis is carried out in combination with a pre-constructed operation and maintenance knowledge graph, and an optimal operation and maintenance disposal plan is generated through matching; and converting the generated operation and maintenance disposal plan into a specific automatic instruction. The method has the advantages that through deep fusion and conjoint analysis of multi-modal data, accurate and panoramic perception of the health state of the low-altitude economic flight data center and prospective prediction of the fault risk are realized, and through automatic execution and a closed-loop feedback mechanism, the safety of the low-altitude economic flight data center is improved. A traditional operation and maintenance mode which is passive in response and highly depends on manpower is converted into a closed-loop autonomous mode which is active in early warning, intelligent in decision making and autonomous in execution.

The invention relates to the technical field of computers, discloses an astronomical sensitive data security processing method, device and system based on privacy calculation, and a storage medium, and aims to solve the problems that in astronomical data cross-mechanism joint analysis, original data is easy to leak, the cooperation efficiency is low, and a traditional desensitization or encryption method is difficult to consider security and availability at the same time. The method specifically comprises the following steps: each participant deploys a private computing agent node locally, and original data is not out of a domain; the task coordination center issues an analysis task; the proxy node extracts and preprocesses local data, and loads a corresponding secure multi-party computing protocol template; according to the method, share segmentation is carried out on multi-modal data such as images, spectrums and star catalogues by adopting addition homomorphic secret sharing, and distribution is carried out through a national secret SM4 encryption channel; and multiple parties cooperatively execute task-oriented security calculation in an encrypted state, and the result is aggregated and returned with the minimum information amount. The method has the effect of realizing high-precision safe collaborative analysis which is available and invisible.

The invention provides an intelligent security and protection monitoring system and method based on a large model, and relates to the field of intelligent security and protection, and the method comprises the steps: firstly receiving video stream data of a monitored region, transmitting the video stream data to an edge end for data preprocessing, and then transmitting the data to a large model analysis module; a conjoint analysis framework fusing target object action semantic time sequence transmission information and background time sequence information is constructed, an intelligent monitoring system based on a large model is utilized to obtain a security and protection monitoring result, finally, an alarm level is determined, and an alarm signal is generated and stored. Thus, the problems that a traditional method is low in data utilization rate and single in feature extraction are solved, and the adaptation algorithm is quickly generated under the small sample even zero sample condition through a mechanism of using while learning, so that the real-time performance and adaptability of security monitoring are remarkably improved, and technical support is provided for accurate early warning and efficient response in a dynamic environment.

The invention discloses a tunnel portal slope risk identification and analysis system based on artificial intelligence, and relates to the field of data processing. Wherein a risk factor investigation and acquisition module, a portal risk simulation processing module, a tunnel risk simulation processing module, a risk factor conjoint analysis module and a slope risk identification and analysis module are arranged; the risk factor investigation and acquisition module is used for acquiring data information of a tunnel construction area; the hole risk simulation processing module is used for constructing a slope damage model according to the obtained data information; the tunnel risk simulation processing module is used for constructing a support failure model according to the obtained data information; the risk factor conjoint analysis module is used for generating a risk influence database; the slope risk identification and analysis module is used for acquiring corresponding risk identification data; according to the invention, the accuracy in the risk identification and analysis process is improved to a certain extent.

The application provides a business data security protection method and system for digital enterprise management, which comprises the following steps: obtaining an access request sequence initiated by a target user on a business operation terminal, generating a dynamic access control policy vector according to a historical behavior track associated with an access operation identifier, and analyzing the dynamic access control policy vector through a policy decoder to generate a real-time access permission instruction; based on the real-time access permission instruction, homomorphic encryption mapping is performed on sensitive data segments in the request context information to generate a ciphertext transmission channel, and the ciphertext transmission channel is synchronized with a security parameter of a cross-department joint analysis model; a federal learning framework is called to perform multi-modal feature fusion on a real-time operation flow of the business operation terminal, an abnormal operation confidence score is generated, and when the score exceeds a dynamic threshold, a cross-level interception protocol is triggered and the current session state is rolled back to a safe baseline version. The application can improve the accuracy and response timeliness of anomaly detection.

The application belongs to the field of data security, and particularly relates to a time-sequencing encryption storage and expiration pulverization method of external device monitoring data, comprising: obtaining an external device monitoring data stream, segmenting according to a preset time window and deriving a unique encryption key for each segment; encrypting the corresponding window data with the key and storing to obtain a time-sequencing encryption data block; in response to a joint analysis request, distributing a specified window key to a calculation party according to a preset authorization strategy for it to perform secure multi-party computation on the window data block; simultaneously, monitoring the data block life cycle in real time, automatically erasing the data and destroying the corresponding key when expired, generating a pulverization verifiable evidence, and uploading to a storage chain constructed by a general data platform root key and each calculation party key combined with a blockchain; the application guarantees data storage security and joint analysis compliance, avoids residual key risk, ensures pulverization operation traceability, and improves multi-party cooperation credibility.

The invention relates to the technical field of privacy computing, and discloses a data circulation and security protection dual-guarantee privacy computing conjoint analysis system. According to the system, by introducing real-time data acquisition, dynamic storage and privacy computing technologies, the efficiency and accuracy of data analysis are improved, and compared with a traditional static data analysis system, the system can flexibly adapt to constantly changing data environments, reflect changes of user behavior modes, market conditions and new data types in time, and improve the user experience. The real-time performance of data is ensured through the dynamic response capability, an enterprise can make a more accurate decision based on latest information, the privacy calculation module effectively protects the privacy of a user through an encryption algorithm and an enterprise tagging technology, high-quality enterprise portraits and credit risk assessment are provided during deep analysis, and the user experience is improved. Therefore, an enterprise can optimize resource allocation, reduce risks and enhance market competitiveness while guaranteeing compliance, and finally higher operation efficiency and higher decision-making ability are achieved.

The invention discloses a passive intelligent vulnerability detection system and method based on Web2 and Web3 fusion, and belongs to the field of block chains, and the passive intelligent vulnerability detection system and method based on Web2 and Web3 fusion comprises a data acquisition layer, a processing analysis layer, a joint analysis module, an intelligent decision-making layer and a response processing layer. The vulnerability detection accuracy is remarkably improved, the real-time response capability of the system is greatly improved, the average detection response time is shortened from the minute level to the millisecond level, threats can be found and handled in time at the early stage of occurrence, and the safety risk is effectively reduced.

The application relates to the technical field of electric digital data processing, and particularly discloses a product information risk assessment method and system based on a large model, which comprises the following steps: simultaneously scanning source code and products by a scanner to generate a source dependency list and a product component list, performing main name normalization and alignment through a large model, locking a warehouse and a version according to the source dependency list, analyzing original code by a parser to generate a cross-language path logic group, combining the product component list to correct a real package version, converging into a unified intermediate representation, performing reachability and joint analysis on the unified intermediate representation to obtain a risk candidate path set and complete path scoring, finally, according to a preset weight and a running period evidence, modifying the score, screening out a risk path, matching a disposal strategy, and generating executable strategy code by the large model for risk control of a gateway or a service side.

This invention discloses a method, system, and electronic device for personnel operation assessment based on video temporal joint analysis, relating to the field of power safety technology. This application involves acquiring video from the work site and extracting frames; extracting the sequence of key human points from each frame; dynamically defining risk areas for the head, torso, and hands based on these key points; calculating hand movement speed, distance from the risk area, and movement consistency characteristics; triggering refined behavior recognition when hand movement characteristics indicate a tendency to enter the risk area; extracting multi-scale temporal features before and after the trigger; and judging specific risk behaviors using a pre-trained action recognition model. This invention, through a two-level mechanism of "trend triggering + refined recognition," achieves early warning and accurate identification of risk behaviors such as removing helmets and gloves, effectively improving the real-time performance and preventative capabilities of work safety monitoring. This application realizes a shift from post-event identification to pre-event warning, resulting in better safety monitoring performance.

The invention discloses an application dynamic configuration optimization method and device, electronic equipment and a storage medium, and the method comprises the steps: in the running process of a target application program based on a Spring Boot framework, monitoring resource use data, and capturing running abnormity; when the resource use exceeds the limit or is abnormal, collecting related target log data in a preset time window, and generating structured first input data; when a user initiates a configuration optimization query, acquiring context data during current operation from a context object of a target application program, and generating structured second input data; a query instruction, first input data and second input data are input into an artificial intelligence analysis model, joint analysis is carried out in combination with historical configuration cases and log data, the problem can be accurately positioned in combination with runtime context and log data, the troubleshooting efficiency is improved, and dependence on artificial experience is reduced; the configuration parameters of the target Bean are dynamically updated through the configuration refresher, so that the configuration change takes effect under the condition that the target application program is not restarted.

The invention relates to the technical field of mine safety, in particular to a mine intelligent self-rescuer real-time evaluation system based on multiple modes, and aims to improve the dynamism, scene adaptability and regionalization management level of mine self-rescuer evaluation. The system comprises a multi-modal data acquisition module, a sensor data calibration module, an intelligent self-rescuer evaluation module and a regional joint analysis module. The multi-modal data acquisition module integrates physical, chemical and visual three-dimensional data; the sensor data calibration module corrects errors caused by environmental interference; the intelligent self-rescuer evaluation module innovatively combines a geographic position and environmental parameters to realize dynamic evaluation of the self-rescuer; the area joint analysis module divides similar sub-areas through K-means clustering, constructs a function exertion capability index and a miner wearing performance index, and quantifies the overall performance of the area. The system breaks through the limitation of traditional static evaluation, and provides comprehensive technical support for precise maintenance and regionalization optimization of the mine self-rescuer.

The invention relates to the technical field of decision support, in particular to a water affair operation data decision support system, which comprises a joint analysis unit used for acquiring historical full-dimension operation data and historical water affair decision knowledge reference data of a water affair system in a target decision period; performing data normalization on the historical full-dimension operation data to obtain a pipe network parameter time sequence and a capacity data time sequence, and performing joint analysis on the historical full-dimension operation data and the historical water decision knowledge reference data to obtain a plurality of corresponding historical decision triples; and the feature extraction unit is used for carrying out associated feature extraction on the pipe network parameter time sequence and the capacity data time sequence. According to the method, the historical decision triple is associated with knowledge nodes such as the pipe network and the productivity, so that the system can construct a comprehensive decision knowledge graph, the knowledge graph integrates the mutual relation between the pipe network and water plant equipment, and a decision maker is helped to better understand and manage a complex network of a water affair system.

The invention discloses a table-based inspection report generation method and related products, and the table-based inspection report generation method comprises a table structure analysis stage which comprises the steps of identifying a header row of a table and judging the column type of each column, and the column type comprises an inspection object and an inspection item; a derivative inspection item generation stage including identifying a combination of inspection items having a conjoint analysis value, and generating a derivative column based on the combination and adding the derivative column to a table; the inspection content analysis stage comprises inspection background analysis, table column analysis and table row analysis of the table; a report unit generation stage including matching the analysis result of the inspection content analysis stage according to the report unit selected by the user to generate report unit content; and the report output stage comprises the steps of adjusting the content of the report unit according to user requirements and outputting a final inspection report. According to the method, large-scale table data can be efficiently processed in a limited context window, table structure semantics can be accurately understood, and report generation quality is remarkably improved.

The invention provides a data sharing security decision-making method and system applying artificial intelligence, and the method comprises the steps: receiving data sharing task description information and an associated talent original data set, inputting the data sharing task description information and the associated talent original data set into a context-aware mapping network, generating a task data association strength mapping table, and carrying out the strategy matching with a preset basic security label set, performing conflict coordination on the data sharing demand and the security constraint condition, generating a fine-grained data protection guidance document, performing structured desensitization conversion operation on corresponding data units in the talent original data set to obtain an intermediate data form, and storing the intermediate data form into a database; and generating a matched access verification rule set and a transmission encryption rule set according to the access authority rule and the transmission encryption rule, deploying the access verification rule set, the transmission encryption rule set and the controlled cooperative computing environment to perform data sharing conjoint analysis, performing authority correction and encryption compliance check on the data access behavior, and outputting a sharing analysis result after the conjoint analysis operation is completed. According to the invention, the compliance and reliability of the shared analysis result report are guaranteed.

The present application relates to the field of big data, and provides a big data task exception detection and repair method, system and medium.The big data task exception detection and repair method comprises the following steps: obtaining task information of a big data task with an abnormal task state, wherein the task information comprises a task log; obtaining execution code of the big data task via a decompilation tool, and analyzing the execution code and the task log via an exception analysis module to identify an exception problem; in response to the exception problem being a data source exception and a code exception, analyzing an abnormal code segment, data source information and an exception type via a joint analysis module to determine an exception root cause, wherein the abnormal code segment is identified by the exception analysis module, the data source information is determined by the execution code, and the exception type is determined according to the task log; and triggering abnormal code repair and / or data processing optimization according to the exception root cause.

The invention discloses a streaming data time window trigger scheduling method for multi-data item conjoint analysis, an operation object of the method is composed of a logic rule list of business application, and each rule is composed of used data items and time window parameters; updating the latest time of the data item; searching a logic rule list using the data item, and then circularly traversing the logic rule list; calculating the minimum value of the latest time of the data item list used by the logic rule; calculating and updating the trigger time of the logic rule; and judging whether a trigger condition is met or not, if the trigger condition is met, triggering analysis and diagnosis, updating time window parameters and then continuing the next cycle, and if the trigger condition is not met, directly continuing the next cycle. According to the method, for each rule, the original data of the corresponding data item can be obtained when the triggering condition is met, the problems of data disorder, delay and repetition are solved, and the reliability of equipment analysis and diagnosis results is improved.

The invention discloses a multi-modal data analysis method and device, equipment and a storage medium, and relates to the technical field of data analysis, and the method comprises the steps: obtaining a data analysis task and structured data and unstructured data of a to-be-analyzed document, performing cross-modal joint analysis and fusion on the structured data and the unstructured data through a cross-modal agent analysis engine to generate multi-modal analysis data; based on the data analysis task, performing dynamic context anchoring and fusion on the multi-modal analysis data and the financial ontology knowledge graph to obtain multi-modal context fusion information; and performing difference identification and traceability based on the multi-modal context fusion information and the financial ontology knowledge graph, and generating an interactive difference traceability graph and a data difference analysis report. According to the method, structured and unstructured data are integrated, data islands are broken, the financial semantic understanding precision is improved, and difference sources are traced, so that the comprehensiveness, timeliness and credibility of multi-modal data analysis are improved.

This application provides a large-scale model-based intelligent security monitoring system and method, which relates to the field of intelligent security. It first receives the video stream data of the monitored area and transmits it to the edge end for data preprocessing. It then transmits it to the large-scale model analysis module. By constructing a joint analysis framework that integrates the semantic temporal transmission information of the target object's action and the background temporal information, and using the intelligent monitoring system based on the large model to obtain the security monitoring results, it finally determines the alarm level, generates an alarm signal and stores it. In this way, it solves the problems of low data utilization and single feature extraction in traditional methods, and through the mechanism of learning while using, it quickly generates an adaptation algorithm under small sample or even zero sample conditions, thereby significantly improving the real-time and adaptability of security monitoring, and providing technical support for accurate early warning and efficient response in dynamic environments.

The invention provides an intelligent routing inspection detection method and system based on a large model and artificial cooperation, and solves the problems of high false alarm rate, poor interpretability, lack of sustainable evolution ability and the like in the prior art. The method comprises the steps of obtaining multi-source heterogeneous inspection data and performing standardized preprocessing, performing multi-modal data conjoint analysis through an initial inspection large model to generate a detection report, performing shunting processing based on confidence score, directly outputting a high-confidence result, and pushing a low-confidence result to an artificial collaborative verification platform for correction. And finally, optimizing the model based on a verification result.

The invention discloses a distribution box fault diagnosis system based on the Internet of Things, and relates to the technical field. The distribution box fault diagnosis system based on the Internet of Things comprises a time sequence data acquisition unit used for acquiring operation state time sequence data of a distribution box, distribution box upstream power supply equipment and distribution box downstream power consumption equipment based on a set time period; the conjoint analysis unit is used for performing conjoint analysis on the operation state time sequence data based on a pre-trained operation state conjoint analysis model to obtain operation state characteristics of the distribution box; the comprehensive analysis unit is used for analyzing the operation state index of the distribution box based on the operation state characteristics, and the fault diagnosis unit is used for judging whether the distribution box breaks down or not based on the operation state index. The sensing granularity and the diagnosis accuracy of the operation state of the distribution box can be effectively improved.

The application discloses an application preheating terminal system based on artificial intelligence and belongs to the technical field of application preheating. In the system, a full-amount data analysis module collects application use data of all authorized users as full-amount data, carries out joint analysis on the full-amount data based on CNN and LSTM, obtains general use habits and general use trends for all authorized users, and generates general preheating strategies; a personalized data training module trains a personalized preheating model according to the use data of the user on the application, the personalized preheating model optimizes the use data of the user and predicts the use habits of the user on the application; and a full-amount data and personalized data fusion module fuses the general preheating strategies and the predicted use habits of the user on the application to preheat the application. The application combines large-scale global data and personalized end-side models, so that the scheduling of preheating resources is more accurate.

A method and apparatus for A / B testing data analysis based on a large-model intelligent agent are disclosed. The method includes: displaying an intelligent interactive page associated with the intelligent agent, which performs joint analysis of different data indicators using a large model; in response to inputting at least two data indicators on the intelligent interactive page, identifying at least two data indicators as target data indicators to be jointly analyzed; and in response to a joint analysis request triggered on the intelligent interactive page for the target data indicators, displaying the data analysis results for the joint analysis request. This allows for automatic data analysis through relatively simple input operations, simplifying the data analysis process and improving efficiency. Furthermore, since it supports the input of data indicators and the joint analysis of different data indicators, it allows for customized, targeted analysis of different data indicators according to user needs, thereby improving the flexibility of data analysis and the user experience.

The application provides a building comprehensive management and control system and method based on artificial intelligence and multi-source data fusion, all subsystem data are fused and processed, cross-field and cross-system joint analysis and decision are realized, and the information island problem is solved.The application combines structure safety and environmental safety data, uses AI to perform risk assessment and prediction, changes passive alarm to active safety prevention and control, and significantly improves the building essential safety level.The application realizes on-demand supply and dynamic optimization based on the AI energy consumption model of multi-dimensional data, and achieves deep energy saving and consumption reduction.The application automatically creates a safe, healthy and comfortable building internal environment by sensing personnel activities and learning preferences, and meets the personalized human settlement environment demand.The application links with the smart city through a standard interface, makes the building an intelligent node responding to the city operation and contributing to the city data, and improves the fine level and overall efficiency of city management.

The embodiment of the invention relates to a virtual DCS safety operator for accident detection and response. A method for security accident detection in a cloud-local DCS in industrial process automation includes monitoring IT-related and OT-related data at a production process and a containerized DCS associated therewith. The method further includes conjointly analyzing first data indicating first monitoring data from the monitoring of the IT-related data and second data indicating second monitoring data from the monitoring of the OT-related data. The conjoint analysis is based on associating at least a portion of the first data with at least a portion of the second data and / or based on associating at least a portion of the second data with at least a portion of the first data. The method further includes, based on the conjoint analysis, detecting a safety accident in consideration of a predetermined safety accident detection rule; and responding to the detected safety accident based on a result of the detection so as to process the detected safety accident under the condition that a preset safety accident response rule is considered.

The application discloses a tunnel portal slope risk identification and analysis system based on artificial intelligence, relates to the field of data processing, and comprises a slope risk identification system, wherein a risk factor investigation and collection module, a portal risk simulation processing module, a tunnel risk simulation processing module, a risk factor joint analysis module and a slope risk identification and analysis module are arranged; the risk factor investigation and collection module is used for collecting data information of a tunnel construction area; the portal risk simulation processing module is used for constructing a slope damage model according to obtained data information; the tunnel risk simulation processing module is used for constructing a support damage model according to obtained data information; the risk factor joint analysis module is used for generating a risk influence database; and the slope risk identification and analysis module is used for obtaining corresponding risk identification data; and the application improves the accuracy in the risk identification and analysis process to a certain extent.