44 results about "Enterprise level" patented technology

The invention provides an enterprise-level intelligent assistant-oriented multi-modal dialogue management and task execution system establishment method, which is characterized by comprising the following steps of Step 1, multi-modal data acquisition and preprocessing; step 2, carrying out cross-modal information alignment and fusion; 3, multi-mode dialogue management is carried out, and after data fusion is completed, the system enters a dialogue management stage; step 4, multi-modal task execution: in the dialogue management process, when a user initiates a specific task request, the system enters a task execution stage; automatically distributing and executing a plurality of tasks in the enterprise by using the fused multi-modal information; step 5, designing and optimizing a system architecture; and an integrated multi-mode intelligent assistant architecture is designed. By integrating multi-modal data processing and natural language understanding (NLU) technologies, the accuracy and efficiency of the enterprise intelligent assistant in task execution are remarkably improved. The system can process various modal information such as texts, voices and images at the same time, and seamless connection of cross-modal dialogue management and task execution is achieved.

The invention provides an enterprise-level unstructured knowledge governance-oriented method and a storage medium. The method comprises the steps that an initial document is input into a visual language model for layout analysis, and document structure information of a target document is obtained; segmenting the document structure information to generate text slices to be vectorized; metadata extraction is carried out based on the text slices, and structured knowledge content vectors are generated; and performing index configuration according to the knowledge content vectors to form enterprise-level knowledge governance standards. According to the method, the document structure is analyzed through the visual language model, the text slices are generated through the intelligent segmentation algorithm, the knowledge vector is constructed through metadata extraction, and the mixed index configuration is performed, so that the problems of inaccurate layout analysis, logic structure damage and low retrieval efficiency in the traditional technology are effectively solved, and the method has the advantage of improving the knowledge management precision and availability.

The invention discloses an enterprise-level agent system and a data processing method. The system mainly comprises a sensing module, a memory module, a planning module, a Routine agent planning engine, an arrangement engine and an execution module. According to the enterprise-level agent system provided by the embodiment of the invention, short-term memory, long-term memory and enterprise-level memory are combined through a complete closed-loop architecture of the sensing module, the memory module, the planning module and the execution module, and precipitation of personalized and organization-level knowledge is supported; and meanwhile, in combination with a standard business process template provided for an enterprise by a Routine agent planning engine and splitting of a task process by an orchestration engine, role agents are allocated for each sub-task, so that coordination and cross-system orchestration of complex tasks are realized, the construction threshold of intelligent applications is reduced, and the business efficiency is improved.

The invention discloses an enterprise-level three-party dependency package security management and control system and method, and relates to the technical field of software supply chain security, and the enterprise-level three-party dependency package security management and control system comprises the following modules: a timed task scheduling module, an external network package pre-scanning module, a private service package monitoring module, a dependency graph construction module, a product management and control module and a notification display module. By establishing an external network packet pre-scanning mechanism, security scanning is performed and a blocking list is generated before a dependent packet enters an enterprise private server, and introduction of a packet containing high-risk vulnerabilities is blocked from the source; meanwhile, through the continuous monitoring of the private server package and the construction of the dependency relationship graph, the vulnerability discovery and the accurate positioning of the influence range of the stored dependency package are realized; and finally, performing hierarchical management and control on the affected products based on vulnerability levels, and realizing timely transmission and situation visualization of risk information through a notification display module. According to the invention, full-life-cycle safety protection from an external network source to internal products of an enterprise is realized, and the safety management level and risk response efficiency of the dependent package of the enterprise are effectively improved.

The invention relates to an enterprise-level SaaS operation control system based on role layering and authority matrix management. The enterprise-level SaaS operation control system is suitable for complex business scenes of multi-tenant and multi-role cooperation. The system comprises an identity recognition unit, a semantic meta-model subsystem, an authority matrix generation engine, a strategy gray release subsystem, a service instruction mapping module and a unified scheduling and auditing module. Constructing a dynamic permission matrix through a unified semantic token, and complementing permission configuration in combination with context information to form a policy version capable of gray deployment; authority operation is executed in a standard interface form after semantic mapping, log recording and audit control are carried out by a scheduling module, a strategy can be automatically frozen and rolled back when abnormal identification is carried out, and automatic evolution and safe closed-loop management of authority configuration are realized.

The invention discloses an enterprise-level MCP service calling method and device, equipment, a medium and a product, and relates to the field of enterprise-level artificial intelligence security services, and the method comprises the steps that an LLM interaction module generates an MCP tool calling list according to a user service request and a user token, and signs the token and the list; the MCP module verifies the signature; the authentication module obtains a field white list according to the token and the list; the MCP module signs the token, the list and the field white list; the service module verifies the signature and sends a result to the MCP module; the MCP module filters and desensitizes the result to obtain a security result, and signs the security result; the LLM interaction module verifies the signature, and inputs a security result and a user service request into a large language model to obtain a natural language reply, the application can realize the problems of dynamic minimum permission, trusted transmission and sensitive data compliance and desensitization, and meets enterprise-level permission control requirements.

The invention discloses an enterprise-level knowledge base information query system and method based on HMAC, and relates to the technical field of information security, the system comprises a client, an authentication module, a large language interaction module and a knowledge base module; the client is used for acquiring an account, a password and a retrieval request of a user; the authentication module is connected with the client and is used for receiving the account and the password and generating a user token; the large language interaction module is connected with the client and used for receiving the retrieval request, generating a query plan and signing the query plan and the user token to obtain a first signature packet; the knowledge base module is connected with the client and is used for receiving the first signature packet and signing again after verification is passed to obtain a second signature packet; and the big language interaction module is also used for receiving the second signature packet, and generating a query result through a second big language model in the big language interaction module after the verification is passed. According to the invention, the security of information transmission between the client and the knowledge base can be improved.

The invention discloses a method and a device for automatically skipping repeated approvers based on a global monitor, and aims to solve the problems of low efficiency, poor experience and the like caused by repeated approval in an enterprise-level approval process. According to the method, a repeated approver detection monitor is automatically injected into all user tasks in a process definition analysis stage by inheriting a BPMN analysis monitor of a process engine, and a process definition file does not need to be modified. When the task is created, the monitor reads the repeated skipping configuration defined by the process, judges whether the scene is a multi-person approval or countersigning scene or not, queries a historical task table to detect whether approvers are repeated or not, and automatically completes the task and sets a skipping mark variable if the approvers are repeated. The method has the advantages of zero invasion, flexible configuration, scene compatibility, traceability, high reliability and the like, and the approval efficiency and experience are effectively improved.

The application provides a component security vulnerability processing method and device, equipment and a storage medium. It relates to the technical field of application security. The method comprises the following steps: based on a full amount warehouse of component risk knowledge, marking the known risk enterprise-level components existing in the enterprise, the version range of the enterprise-level components, and the related applications having a dependency relationship with the enterprise-level components in the pre-constructed component panoramic view, the component panoramic view recording all components in the enterprise and the applications having a dependency relationship with each component; determining the risk level of the enterprise-level components based on the use of the enterprise-level components in each related application; determining the risk processing strategy for the enterprise-level components based on the risk level; and processing the enterprise-level components published in the enterprise component warehouse based on the risk processing strategy. The method of the application solves the problem of poor rectification effect when rectifying the enterprise-level components having security vulnerabilities.

The invention discloses an enterprise-level form generation method and system supporting low-code development, and relates to the technical field of enterprise form generation, and the method comprises the following steps: building a semantic security construction layer, analyzing an original character string logic expression generated by a low-code logic composer, analyzing the logic expression into a plurality of structured semantic nodes, and storing the structured semantic nodes in a database; the action range of the variables is limited based on the structured semantic nodes, so that the logic generation process is expanded in the closed action range; and constructing a dynamic action domain isolation container based on the generated structured semantic nodes, and distributing each structured semantic node to an independent context space. According to the method, closed operation of logic generation and execution is realized through semantic security construction and dynamic action range isolation, variable pollution is prevented, and semantic stability is ensured; through cooperation of namespace control and a limited execution environment, logic operation forms a safe closed loop, and the operation stability and safety of the low-code form are improved.

The invention relates to the technical field of artificial intelligence, and is used for solving the problems that an existing employee digital duplication system often neglects effective collection and analysis of task performance data of employees in a unit period and lacks a screening mechanism based on historical behaviors and task performance; the problems of high response delay, task omission and overtime completion of partial clones in task execution due to employees of a construction source are solved; the enterprise-level employee digital duplication system based on artificial intelligence comprises a server, and the server is further connected with a data acquisition module, a difficulty judgment module, a response analysis module, a reliability analysis module, a planning module and an effect verification module. Quantitative analysis is carried out through the process complexity, the employee response timeliness and the task reliability degree, and the intelligence and the accuracy of the digital duplication system are improved in combination with a judgment rule and an interval classification mechanism.

The invention discloses a group insurance underwriting method based on artificial intelligence. The method comprises the steps that S1, enterprise level and employee level multi-source data information submitted by an insurance buying enterprise is acquired and preprocessed; s2, constructing a composite risk feature set of an enterprise level and an employee level based on the data preprocessed in the S1; s3, performing correlation evaluation on the data preprocessed in the S1 based on the composite risk feature set constructed in the S2, screening out high-correlation features higher than a preset threshold value, and generating feature vectors according to the high-correlation features; s4, performing primary review on the feature vectors generated in the S3, and screening insurance policies which are obviously compliant or illegal; s5, constructing a multi-model fusion engine for risk prediction by adopting a Stacking integrated learning architecture; and S6, carrying out risk scoring on the insurance policy screened and compliant in the S4 by adopting the multi-model fusion engine constructed in the S5, and outputting a final underwriting conclusion. According to the method, multi-dimensional analysis is carried out on the underwriting data, and the risk assessment report is generated regularly, so that risk grading is carried out on insurance enterprises and customers.

The invention relates to the technical field of financial risk control, in particular to an enterprise-level risk base line management and control system and method based on a dynamic rule engine and multi-level workflow. The system is used for realizing real-time / batch dual-mode scanning of customer risk portraits, automatic hierarchical control of black and grey lists, full-process digital examination and approval of special case services and list transfer, and multi-dimensional visual analysis of risk data.

The embodiment of the application discloses a kind of clothing data processing method and equipment based on enterprise-level operating system, belong to clothing data processing technical field, solve clothing industry data entry link, due to the higher error rate of manual operation, affect the accuracy and completeness of industry data problem.The entity data recognition and format uniformity of clothing basic data are carried out by prepositioning multimodal pre-training model;Clothing basic data is input into different data modules, and the knowledge graph corresponding to entity data and different data modules respectively, to detect error data for clothing basic data;The repair value prediction of detected error data is carried out by data prediction filling model, and the clothing data to be fused is generated by repaired data;The feature vector corresponding to the clothing data to be fused and historical clothing data respectively is associated degree processing, and constructs clothing association relationship network;Based on clothing association relationship network, the clothing data to be fused is fused to the corresponding position of system.

The invention provides an enterprise-level data integration processing method which is applied to a data integration processing system. The method comprises the steps of obtaining original business data of various heterogeneous data sources, and performing standardization processing to generate initial standardized data assets; carrying out ETL processing on the initial assets and storing the initial assets in a unified storage layer to generate intermediate standardized data assets; performing data management on the intermediate standardized data assets to generate target standardized data assets; unified scheduling and fault-tolerant control are carried out on execution tasks depended on by target assets through visual DAG arrangement, and continuous and stable generation of the execution tasks is guaranteed; and finally, performing data analysis and visual display based on the target standardized data assets, and generating an analysis result interface. The standardized and automatic management of the full life cycle of the data is realized, the problems of difficult integration of the multi-source heterogeneous data, low processing efficiency, uncontrollable quality, high operation and maintenance cost and the like are effectively solved, and the efficiency, stability, accuracy and collaboration of data processing are improved.

The application provides a data unloading method and device for enterprise-level data, an electronic device and a storage medium. The application relates to data processing technology, and is particularly suitable for an enterprise-level data unloading scenario. The method comprises the following steps: acquiring a source data group in a source data layer of a data warehouse, so as to determine an enterprise-level data dictionary followed by the source data group; then, acquiring an application data version of a business component in a business system; and according to the application data version and the enterprise-level data dictionary, unloading enterprise-level data to the source data group, so that the data types of the business component and the source data group are consistent. The application can solve the problems that the data unloading transition in the prior art depends on specific databases and the business component is not connected with the enterprise-level data dictionary.

The invention relates to the technical field of artificial intelligence, in particular to an enterprise-level permission implicit risk identification method and system based on a large language model, and the method comprises the steps: collecting permission setting data from one or more target information systems; uniformly converting the permission setting data in different formats into a structured data set comprising preset core permission elements; inputting the structured data set into a large language model subjected to security domain knowledge fine adjustment, and executing conversion from codes to texts to generate permission description; executing a relationship extraction task by using a large language model subjected to security domain knowledge fine adjustment, extracting permission entities and relationships from the permission description, and constructing a global permission knowledge graph; and visualizing the global authority knowledge graph to generate an authority topological graph, traversing the knowledge graph in combination with a built-in risk rule base to identify hidden risks, and generating a semantic alarm. According to the invention, intelligent, transparent and visual enterprise-level authority management is realized, and various authority hidden dangers can be found in time.

This application provides an enterprise-level model validation method and an enterprise-level model validation device. The method involves acquiring user requirement information and the previous version of the data model, parsing the user requirement information to obtain the parsing results, generating the current data model based on the parsing results (including model documents and their implementation attributes), performing a difference analysis between the previous version and the current data model to generate difference analysis results (including structural and field changes), generating an incremental difference SQL script based on the difference analysis results, executing the incremental difference SQL script in the target database environment, and performing a consistency check between the table structure of the target database after execution and the current data model to obtain the validation results. This application solves the problems of low efficiency and low accuracy in existing model validation methods.

The invention discloses an enterprise-level application operation and maintenance risk management and control method and device, a program and a storage medium. The method comprises the following steps: acquiring a plurality of risk characteristics of a target risk event; determining the risk type of the target risk event; determining a risk coping strategy for the target risk event according to the risk field and a preset risk coping strategy set; under the condition that the risk field is the known risk, obtaining operation environment data of the operation and maintenance carrier; determining a target diagnosis script according to the operation environment data, the risk coping strategy and a preset diagnosis script library; determining a target solution for the target risk event based on the target diagnosis script and a preset solution set; sending the target solution to the front end, and in response to a received confirmation instruction sent by the front end, generating a risk disposal script according to the target solution; and controlling the operation and maintenance carrier to operate the risk disposal script, and outputting a risk disposal result. According to the method, the risk blind area in enterprise-level application operation and maintenance can be reduced.

The application discloses an enterprise-level MEC authentication system hierarchical deployment method and device, wherein the method comprises the following steps: deploying a first-level system at an operator to provide SMF configuration, enterprise-level authentication system configuration, signaling firewall configuration and Radius message forwarding function; deploying a second-level system at an enterprise to provide APN configuration, user account configuration, client configuration and authentication service function; the two-level systems work in cooperation and interact through the Radius message forwarding function. The method and device realize two-level control, the first-level system realizes operator network access equipment management and security policy, and the second-level system realizes enterprise internal control, effectively solving the complexity and security problems of operator and enterprise user configuration.

The invention provides a fault code unified generation method based on an enterprise-level pedigree tree, and relates to the field of complex electronic product fault diagnosis, and the method comprises the steps: constructing an enterprise-level product pedigree tree which is a hierarchical structure with a plurality of levels, and distributing a unique enterprise-level code for each unit in the pedigree tree; a public information coding unit for generating a fault code based on the enterprise-level product pedigree tree; establishing a fault information set of a corresponding product, allocating a unique fault level code to each piece of fault element information in the fault information set, and generating a private information coding unit of a fault code based on the fault level code; and combining the public information coding unit and the private information coding unit to generate a complete fault code. By constructing an enterprise-level product pedigree tree system, fault code design is divided into a public information part and a private information part, a fault type information unit and a fault information set unit are recorded, and unification, standardization, hierarchy and traceability of fault codes are achieved.

The invention discloses an enterprise-level remote office method combined with an intelligent security mechanism, and belongs to the field of network security. The method comprises the following steps: after a user completes two-factor authentication through a Web browser, selecting a remote computer allocated by an administrator to initiate connection; and establishing WebRTC connection, coding screen data of the controlled end into a video stream, and transmitting the video stream to the control end through P2P direct connection or a TURN server. And the control end records keyboard and mouse operation instructions in real time and stores the keyboard and mouse operation instructions in the Redis database. The stored instructions are scanned through a detection model of the automatic encoder, and malicious scripts are identified. And an instruction is encrypted through AES-256, and the encrypted instruction is transmitted to a controlled end through an RTC Data Channel or TURN server. And after the controlled end decrypts the instruction, the RobotGo module analyzes and executes the operation. According to the method disclosed by the invention, the safety, the management efficiency and the use convenience of remote office are remarkably improved.

The embodiment of the invention discloses an enterprise-level network equipment abnormal flow control method and device, equipment and a medium. According to one specific embodiment, the method comprises the steps that mirror image network flow data corresponding to preset enterprise-level network equipment are received from a preset mirror image port, and the mirror image network flow data comprise inflow sub-data sequences; performing abnormal primary filtering detection processing on the inflow sub-data sequence to obtain primary filtering abnormal detection information; based on the preliminary filtering anomaly detection information, performing time sequence hidden change self-adaptive protection processing on the inflow sub-data sequence to obtain dynamic anomaly sensing information; and performing global abnormal traffic inbound control processing on the preset enterprise-level network equipment based on the preliminary filtering anomaly detection information and the dynamic anomaly sensing information. According to the embodiment, the potential threat risk of enterprise-level network equipment is reduced.

The present application relates to resource scheduling and flow control technology field, specifically relates to a kind of enterprise level construction resource's quota and flow limiting method and device, by obtaining the preset construction resource quota according to its belonging team or project;Based on the system load data collected in real time, the best concurrent task number allowed by system is calculated and smoothly adjusted by dynamic flow limiting algorithm;According to multidimensional priority rating rules, the request is sorted;Abnormal construction behavior is identified and managed using risk control strategy;Finally, the quota, flow limiting, priority and risk control results are comprehensively scheduled to make unified scheduling decision.The quota management, dynamic flow limiting, intelligent scheduling and risk control management are integrated into a unified framework, realizing the fine, automated and intelligent management of construction resources, solving the problem of uncontrollable, easy overload and key task guarantee of enterprise-level construction platform resources.

The application relates to an artificial intelligence technology and discloses a data intelligent processing method based on an enterprise-level administrative organization tree, which comprises the following steps: extracting features of obtained enterprise information, performing hierarchical classification, obtaining hierarchical features, constructing an administrative organization tree of the target enterprise according to the hierarchical features, obtaining an initial resource framework and an initial billing framework, performing module configuration on the initial resource framework and the initial billing framework, optimizing resource cost by using a generated cost algorithm, visually marking the resource cost, establishing a correlation relationship, rendering an overview interface according to the correlation relationship, and performing detail display on the overview page in response to a browsing instruction of a user. In addition, the application also relates to a blockchain technology, and a data list can be stored in a node of the blockchain. The application further provides a data intelligent processing device and equipment based on the enterprise-level administrative organization tree. The application can improve the efficiency of data management.

The technical scheme of the present application provides a metadata management method applied to enterprise-level data governance.Another technical scheme of the present application provides a metadata management platform, which comprises a meta-model management module, a metadata query module, a metadata API module, a metadata analysis module and an operation log module.The technical scheme disclosed by the present application structures metadata definition into specific information, so that the metadata of each system can be structured, expressed and recorded.In addition, the present application defines the relationship between meta-models, and the description of the relationship can clearly link different systems, describe the relationship between systems, and help prevent the circular association, repeated association or wrong association between systems.

The invention provides an enterprise-level intelligent scheduling agent system and method, a medium, a terminal and a program product, and the system comprises an agent application market construction module which is used for constructing a workflow on an agent construction platform according to employees, and disclosing an agent set for collective use; the workflow intelligent scheduling module obtains and judges a task processing request, and if the task is judged to be a simple task, the task is processed and a processing result is returned; if the task is judged to be a complex task, searching an agent application mart to match a corresponding agent, and executing a corresponding workflow according to the matched agent to return a processing result and a full execution path; the multi-platform interaction module is used for the employees to interact on a plurality of platforms; and the health office module triggers a health reminding strategy based on a preset detection mechanism and generates a health data report in a set period. According to the invention, efficient office experience can be provided for employees, and collaborative optimization of health and work is realized.

The invention provides an enterprise-level full-volume data intelligent search method and system. The method comprises the following steps: acquiring total data from a plurality of heterogeneous data sources of an enterprise operation platform, and performing dynamic semantic association analysis on the total data based on enterprise business rules and domain terms to generate a semantic association framework; after a natural language search request input by a user is received, the natural language search request is analyzed into a semantic search mode, then the semantic search mode and a semantic association framework are combined and analyzed to obtain a target data retrieval sequence, finally data search is executed in stages according to the sequence, and all search results are integrated to generate enterprise-level search responses. According to the technical scheme provided by the invention, a search behavior is converted into intelligent retrieval which can adapt to enterprise data ecology and accurately interpret business intentions, and the accuracy, correlation and integrity of retrieval in an enterprise complex data environment are also improved.

The invention discloses an enterprise-level multi-source abnormal behavior monitoring method and system, and particularly relates to the field of enterprise-level multi-source abnormal behavior monitoring, and the method comprises the steps: an end-side device executes a process structure analysis operation, and executes an expected behavior set construction operation according to a function module, an interface calling relation and a process rule deployed by the end-side device; generating an end-side expected behavior set; and the end-side equipment executes matching calculation operation on the actual behavior events collected in the operation cycle, and executes item-by-item comparison operation on the actual behavior events and the end-side expected behavior set. An expected behavior sequence composed of a process node, a function module and interface calling is constructed on an end side, double-layer comparison of node coverage and interface coverage is performed on an actual behavior event in an operation cycle, and cross-end-side missing alignment, path expansion and link judgment calculation of a cloud end are combined, so that a real behavior event is obtained. And computable recognition of key behaviors which should appear but are actually missing in the business process is realized.

A system and methods for enterprise hierarchical persistent cognitive machines (PCMs) that extends mobile-optimized multi-stage language model processing with organizational hierarchy awareness and enterprise-specific adaptations. The system comprises a CEO-PCM at the executive level coordinating multiple functional domain PCMs corresponding to organizational departments. A supervisory network layer dynamically adapts underlying language models to enterprise-specific dialect, terminology, and communication patterns through automated analysis of organizational knowledge sources. User prompts are routed through enterprise hierarchical pathways based on authority levels and sensitivity classification, enabling appropriate escalation and cross-functional coordination. The CEO-PCM synthesizes insights from multiple domain PCMs to generate enterprise-wide strategic intelligence while domain-specific compliance modules enforce regulatory requirements during cognitive processing. The system automatically adapts to organizational changes by redistributing knowledge and reconfiguring PCM architecture, while maintaining persistent cognitive state across enterprise operations through comprehensive state preservation and restoration mechanisms that ensure organizational knowledge continuity.