Intelligent organizational knowledge application method and device for uncertain event response

Abstract

The embodiment of the invention provides an intelligent organizational knowledge application method and device for uncertain event coping, and the method comprises the steps: obtaining experience knowledge for coping with uncertain events from historical cases and simulation cases, constructing a knowledge base system based on the experience knowledge, converting the experience knowledge into nodes and relationships of the knowledge base system, and carrying out the processing of the nodes and relationships of the knowledge base system. And an AI Agent is constructed according to the task chain and the thinking chain so as to be responsible for task execution and decision making. According to the method, by efficiently extracting insights and experience knowledge of business experts and employees under uncertain events and integrating the insights and the experience knowledge into AI agent design, the professionality and reliability of decision making are improved, and then external and intelligent application of the experience knowledge is promoted; therefore, the AI agent can make a more accurate and timely response in a complex and changeful environment.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and specifically to a method and device for intelligently applying organizational knowledge for coping with uncertain events. Background Art

[0002] AI Agent is an autonomous artificial intelligence program. It can perceive the environment (collect information through sensors, data input, etc.) and make decisions or take actions based on these perceptions to achieve specific goals or tasks. With the rapid development of artificial intelligence technology, AI agents are playing an increasingly important role in various fields (such as medical, finance, manufacturing, etc.).

[0003] However, in the case of uncertain event response, the occurrence and outcome of events cannot be determined, and there is also uncertainty in the relationship between events and business activities or other related factors. Traditional AI agents usually rely on predefined rules and limited decision trees, and have difficulty coping with complex and changing environments and tasks. Large Language Model (LLM) provides new possibilities for building smarter and more flexible AI agents with its powerful natural language understanding and generation capabilities. In addition, in such situations, the complementary capabilities of artificial intelligence and humans can be used to effectively cope with such challenges through human-intelligence collaboration.

[0004] However, how to fully tap into the experience and knowledge of business employees and integrate them into the design of AI agents to generate AI agents that can make collaborative decisions with business employees and promote the intelligent application of experience and knowledge is still a problem that needs to be solved urgently. Summary of the invention

[0005] In response to the problems in the prior art, the present application provides a method and device for intelligent application of organizational knowledge for responding to uncertain events, so as to enhance the adaptability of AI agents in dynamic environments, so that AI agents can make more accurate and timely responses in complex and changing environments.

[0006] In order to solve at least one of the above problems, the present application provides the following technical solutions: In a first aspect, the present application provides an intelligent application method of organizational knowledge for dealing with uncertain events, including: Acquire experiential knowledge on coping with uncertain events, and extract key features from the experiential knowledge; the key features include uncertain events, business scenarios, collaborative participants, process steps, and decision logic; Establish a situation model based on the collaborative situation metamodel, associate the uncertain events, business situations, collaborative participants with the situation model, and form decision knowledge nodes; define the relationship between decision knowledge nodes according to the process steps and the decision logic, and build a knowledge base system based on the relationship between the decision knowledge nodes; Searching the knowledge base system to obtain the decision-making knowledge nodes for various uncertain events and the relationships between the decision-making knowledge nodes; generating a task chain and a thinking chain for the AI ​​Agent based on the decision-making knowledge nodes for various uncertain events and the relationships between the decision-making knowledge nodes; generating the AI ​​Agent based on the task chain and the thinking chain; Receive input commands from the user, and parse the input commands through the AI ​​Agent; the AI ​​Agent infers the results of the analysis based on the task chain and the thinking chain to generate decision suggestions, and generates a decision plan for dealing with uncertain events based on the task chain.

[0007] Furthermore, the step of acquiring experience knowledge for dealing with uncertain events includes: Obtain review files through the review guide for coping with uncertain events, and obtain the first-hand experience knowledge of uncertain events from the review files; A large language model is used to parse historical data to obtain second-level empirical knowledge of uncertain events; the historical data includes historical cases and meeting minutes.

[0008] Furthermore, the step of associating the uncertain events, business scenarios, collaborative participants and scenario models to form decision knowledge nodes includes: Using uncertain events as trigger conditions, dynamic response is achieved through event-driven nodes in the situation model; Acquire operational data related to the business scenario and convert the operational data into dynamic constraint variables of the scenario model; Collaborative participants are defined as role nodes in the situation model and their decision-making authority is bound.

[0009] Furthermore, the step of defining the relationship between decision knowledge nodes according to the process steps and the decision logic includes: ‌Decomposing the decision logic into a logical chain between decision conditions and execution actions, and converting the logical chain into causal rules, establishing a deterministic mapping between event triggering conditions and disposal measures, and forming an initial causal rule library; Establish a variable causal relationship template based on the initial causal rule base, allowing the causal strength to be dynamically adjusted according to the business scenario, and support the simulation of causal relationships that may exist in the future by embedding a causal inference engine; It also includes: decomposing the process steps into independent decision knowledge nodes, analyzing the timing constraints between operations in the process steps, extracting process sequence features, and converting the process sequence features into a timing dependency chain between the decision knowledge nodes to obtain a time dependency relationship.

[0010] Furthermore, the step of generating a task chain and a thinking chain for the AI ​​Agent based on the decision knowledge nodes for various uncertain events and the relationship between the decision knowledge nodes includes: Decompose the execution process for uncertain events into a series of subtasks, determine the execution order and dependencies between the subtasks, and generate a task chain for AI Agents; Based on the decision logic for uncertain events and combined with the chain thinking method, we designed the perception thinking chain, memory thinking chain and reasoning thinking chain, and constructed the reasoning path of AI Agent to obtain the thinking chain for AI Agent. For each subtask in the task chain, a set of associated thinking chain nodes is specified to form a mapping relationship table.

[0011] Furthermore, the step of generating an AI Agent based on the task chain and the thought chain includes: According to the subtask types in the task chain, a set of capability dimensions of the AI ​​Agent is defined, and an Agent feature vector is constructed based on the capability dimensions; The LangChain framework is used to build the Agent model architecture, which includes a core reasoning module, a task execution module, a tool calling module and a memory management module; The Agent model is compressed through knowledge distillation technology, and the model version is dynamically selected according to hardware resources to obtain the final AI Agent.

[0012] Furthermore, the AI ​​Agent generates decision suggestions by reasoning about the results of the analysis based on the task chain and the thinking chain, and the step of generating a decision plan for dealing with uncertain events based on the task chain also includes: recording the interaction content between the AI ​​Agent and the user, extracting new decision-making experience from the interaction content, and optimizing the knowledge base system based on the new decision-making experience.

[0013] In a second aspect, the present application provides an intelligent application device for organizational knowledge for dealing with uncertain events, including: A knowledge extraction module is used to obtain experiential knowledge for dealing with uncertain events and extract key features from the experiential knowledge; the key features include uncertain events, business scenarios, collaborative participants, process steps, and decision logic; A knowledge base system construction module is used to establish a situation model based on the collaborative situation metamodel, associate the uncertain events, business situations, collaborative participants with the situation model, and form decision knowledge nodes; define the relationship between decision knowledge nodes according to the process steps and the decision logic, and construct a knowledge base system based on the relationship between the decision knowledge nodes; An AI Agent generation module is used to search the knowledge base system to obtain the decision-making knowledge nodes for various uncertain events and the relationship between the decision-making knowledge nodes; generate a task chain and a thinking chain for the AI ​​Agent based on the decision-making knowledge nodes for various uncertain events and the relationship between the decision-making knowledge nodes; and generate an AI Agent based on the task chain and the thinking chain; The interactive module is used to receive the user's input commands and parse the input commands through the AI ​​Agent; the AI ​​Agent infers the parsed results based on the task chain and the thinking chain to generate decision suggestions, and generates a decision plan for dealing with uncertain events based on the task chain.

[0014] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for intelligently applying organizational knowledge for coping with uncertain events are implemented.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for intelligently applying organizational knowledge for coping with uncertain events.

[0016] In a fifth aspect, the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the method for intelligent application of organizational knowledge for coping with uncertain events.

[0017] It can be seen from the above technical solution that the present application provides a method and device for intelligent application of organizational knowledge for dealing with uncertain events. The method obtains experiential knowledge for dealing with uncertain events from historical cases and simulation cases, and builds a knowledge base system based on the experiential knowledge, converts the experiential knowledge into nodes and relationships of the knowledge base system, and then builds an AI Agent based on its task chain and thinking chain to be responsible for executing tasks and making decisions. By efficiently extracting the insights and experiential knowledge of business experts and employees under uncertain events and integrating them into the design of AI agents, the professionalism and reliability of their decision-making can be improved, thereby promoting the externalization and intelligent application of experiential knowledge, so that AI agents can make more accurate and timely responses in complex and ever-changing environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 This is one of the flow charts of the method for intelligent application of organizational knowledge for dealing with uncertain events in the embodiment of the present application; Figure 2 This is a second flow chart of the method for intelligently applying organizational knowledge for coping with uncertain events in an embodiment of the present application; Figure 3 This is a flowchart of the third method for intelligent application of organizational knowledge for coping with uncertain events in an embodiment of the present application; Figure 4 A structural diagram of an intelligent application device for organizational knowledge for dealing with uncertain events in an embodiment of the present application; Figure 5 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application.

[0020] Reference numerals: Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver program storage unit 9144, antenna 9111, speaker 9131, microphone 9132. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0022] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0023] Taking into account the problems existing in the prior art, this application provides a method and device for intelligent application of organizational knowledge for dealing with uncertain events. The method obtains experiential knowledge for dealing with uncertain events from historical cases and simulation cases, and builds a knowledge base system based on the experiential knowledge, converts the experiential knowledge into nodes and relationships of the knowledge base system, and then builds an AI Agent based on its task chain and thinking chain to be responsible for executing tasks and making decisions. By efficiently extracting and integrating the insights and experiential knowledge of business experts and employees under uncertain events into the design of AI agents, the professionalism and reliability of their decision-making can be improved, thereby promoting the externalization and intelligent application of experiential knowledge, so that AI agents can make more accurate and timely responses in complex and ever-changing environments.

[0024] In order to enhance the adaptability of AI agents in dynamic environments and enable them to respond more accurately and promptly in complex and ever-changing environments, this application provides an embodiment of an intelligent application method of organizational knowledge for dealing with uncertain events, see Figure 1-Figure 3 The intelligent application method of organizational knowledge for dealing with uncertain events specifically includes the following contents: Step S101: Acquire experiential knowledge for dealing with uncertain events, and extract key features from the experiential knowledge; the key features include uncertain events, business scenarios, collaborative participants, process steps, and decision logic.

[0025] Optionally, the experiential knowledge is extracted from simulation cases and historical data within the enterprise. Specifically, a review file is obtained through the review guidance of uncertain event response, and the first experiential knowledge of uncertain events is obtained from the review file; the historical data is parsed using a large language model to obtain the second experiential knowledge of uncertain events; the historical data includes historical cases and meeting minutes.

[0026] Exemplary: (1) Guidance on reviewing and coping with uncertain events Starting from the six dimensions of uncertain events, business scenarios, decision-making goals, collaborative participants, decision-making basis and decision-making behavior, offline workshops are organized to integrate brainstorming and scenario simulations, guide sales experts and employees to review and review the response process of uncertain events, and form a review document for uncertain event response. In this way, the experience of experts and employees in dealing with similar uncertain events is explored, and the response strategies and decision-making processes are identified and extracted. These experiential knowledge will provide support for AI agent generation. Including: (a) Determine the review objectives and framework: Goal setting: The main goal of the review is to explore effective response strategies, experiences, and decision-making logic by reviewing the response process of uncertain events, so as to provide support for the subsequent generation of AI agents.

[0027] Framework design: Organize the review work according to the following six dimensions: Uncertain events: Define the characteristics of events, such as suddenness, complexity, and scope of impact.

[0028] Business context: Identify the business context in which the event occurred.

[0029] Decision-making objectives: Discuss decision-making objectives, such as minimizing losses, improving efficiency, and ensuring rational allocation of resources.

[0030] Collaborative participants: Identify the roles of all parties involved in the incident response process, such as business department heads, management, technical support teams, etc.

[0031] Decision Basis: Identify the data, rules and standards on which the decision-making process is based.

[0032] Decision-making behavior: Review the actual decision-making actions taken during the incident response process and their effects.

[0033] (b) Organising workshops and interactive sessions, which can be done through: Offline workshop organization: Invite business experts, department heads and relevant employees to participate to ensure that the review can cover all aspects of the business perspective.

[0034] Situational simulation: simulate the situation of an event, let participants play different roles, and experience the process of dealing with uncertain events. Situational simulation includes online simulation and offline simulation. Through situational simulation, employees can better review and understand the decision-making steps and strategies in actual response.

[0035] Brainstorming: Guide experts and employees to brainstorm and come up with what they think are the most effective response strategies, explore potential optimization solutions and improved decision-making processes.

[0036] (c) Summarize and organize coping experience: Document the review process: During the review process, ensure that all discussions and feedback are recorded in detail, especially those key points about decision-making basis, behaviors and results.

[0037] Summarize key experiences: Integrate feedback from business experts and employees to summarize key strategies and best practices for dealing with uncertain events. These experiences will support the subsequent generation of AI agents.

[0038] Extraction of experiential knowledge: During the review process, clarify which is explicit knowledge (such as compiled specifications, standard operating procedures, etc.) and which is tacit knowledge (such as employees' personal experience, intuitive judgment, etc.).

[0039] (d) Form a review document: Based on the key decision points, experience strategies, action steps, etc. extracted during the review process, write a review document for dealing with uncertain events.

[0040] (2) Extraction of empirical knowledge Based on the uncertain event review documents and the explicit knowledge (such as rules and regulations, cases, etc.) and implicit knowledge (such as meeting minutes or recordings, etc.) in the organization, the knowledge of situations, personnel, business behaviors, goals, etc. for dealing with uncertain events is extracted, which further includes the following steps: (a) Explicit knowledge extraction: LLM’s natural language processing capabilities are used to parse the company’s internal order management rules and regulations, historical order case documents, etc. Through text segmentation, entity recognition, and relationship extraction, key decision events (such as order priority setting, production plan adjustment, financial budget adjustment, etc.), decision logic (such as order sorting strategies based on profit margins and delivery time), process steps (such as order receipt, review, and production scheduling processes), etc. are extracted. This information is stored in a structured form in the explicit knowledge base to facilitate subsequent retrieval and application.

[0041] The specific technical solution includes using pre-trained BERT or large language models for text segmentation and sentence boundary detection to perform high-precision paragraph and sentence segmentation on long documents; using large models for named entity recognition to identify key entities such as people, decision-making situations, behaviors, and goals; identifying semantic relationships between entities through the semantic understanding capabilities of large models; and storing the extracted explicit knowledge in non-relational databases such as MongoDB or graph database Neo4j to support subsequent queries and analysis.

[0042] (b) Extraction of implicit knowledge: Use speech recognition technology to convert order management meeting recordings, employee discussion records, etc. into text, and then use LLM technology to perform topic modeling and sentiment analysis to extract the experience, insights and suggestions of business employees and experts in order selection. For example, from a meeting on order priority adjustment, extract employees' experience and suggestions on "flexibly adjusting order priorities to cope with sudden demand" and experts' insights on "optimizing order combinations to maximize profits". This implicit knowledge is stored in the implicit knowledge base in the form of key elements and decision logic, supplementing and enriching the content of the explicit knowledge base.

[0043] The specific technical solution includes using Wav2Vec 2.0 for high-precision speech-to-text conversion. The big model is used for word segmentation, part-of-speech tagging and denoising, and the transcribed text is cleaned and standardized. Through dependency syntactic analysis and semantic role labeling supported by the big model, the key decision-making elements and logical relationships in the text are identified, and the extracted implicit knowledge is finally stored in non-relational databases such as MongoDB or graph database Neo4j to support complex queries and association analysis.

[0044] Step S102: Establish a situation model based on the collaborative situation metamodel, associate the uncertain events, business situations, and collaborative participants with the situation model to form decision knowledge nodes; define the relationship between decision knowledge nodes according to the process steps and the decision logic, and build a knowledge base system based on the relationship between the decision knowledge nodes.

[0045] Optionally, establishing a situation model based on the collaboration situation metamodel includes: Based on the knowledge of uncertain event response review, situational knowledge is extracted to define different order selection decision situation categories and attributes, such as "high-profit order priority selection situation", "urgent order priority selection situation", "order adjustment situation under insufficient resources", etc.

[0046] Based on the collaborative situation meta-model, the corresponding situation model (elements are shown below) is established to call the corresponding decision-making knowledge in different uncertain event response situations. For example, in the "high-profit order priority selection situation", the situation model will include attributes such as order profit margin, production resource availability, and delivery time to provide decision-making references for AI Agent.

[0047] Elements of the situation model: Objective: Describe the business objective of this collaboration scenario.

[0048] Collaborative participants: When conducting business in this scenario, the participation of people inside and outside the department is required. These people are closely related to the scenario or need to provide support.

[0049] Business context: Identify the business context in which the event occurred.

[0050] Goal: Describes the goals that collaborative participants need to accomplish in a collaborative situation.

[0051] Activities: Activities that need to be performed to complete the scenario. These activities may require the participation of scenario participants.

[0052] Capabilities: Describe some of the abilities or knowledge that collaborative participants need to perform activities.

[0053] Resources: Describe the data needed by collaborative participants to perform activities.

[0054] Optionally, in this embodiment, the steps of associating the uncertain events, business scenarios, collaborative participants and situational models to form decision-making knowledge nodes include: using uncertain events as trigger conditions to achieve dynamic response through event-driven nodes in the situational model; obtaining operational data related to the business scenarios, and converting the operational data into dynamic constraint variables of the situational model; defining collaborative participants as role nodes in the situational model and binding their decision-making authority.

[0055] In this embodiment, key features (such as event types, participants, etc.) are essentially discrete abstractions of real elements, but lack the ability to express the following dynamic elements, such as: the real-time status of business scenarios (such as inventory fluctuations, market changes); dynamic adjustment of permissions of collaborative participants (such as temporary increase in approval levels in crisis mode); timing constraints of process steps (such as task A must be started after task B is completed). If nodes are formed directly based on key features, they will become static knowledge fragments and cannot adapt to environmental changes. Therefore, the situational model injects context-awareness into decision-making knowledge nodes through the following mechanisms: dynamic constraint variables (such as real-time inventory levels, supplier risk ratings); event-driven responses (such as automatically activating associated nodes when sensor alarms are triggered); role permission binding (such as differences in operating permissions for participants of different job levels). This dynamic association enables nodes to automatically adjust their logical behavior as the situation changes.

[0056] Exemplarily, in the knowledge base system, uncertain events are classified into discrete events and continuous events according to their attributes, and the CEP (complex event processing) engine is used to define the event pattern. When an event is triggered, the preset action interface (such as RESTAPI) is called, and the event context parameters are passed; real-time streams such as sensors and transaction systems are connected through Apache Kafka, and batch data such as inventory and orders are synchronized from the ERP system every day, and the JSON Schema constraint variable structure is defined to obtain variable generation rules; policies are defined based on the ABAC (attribute-based access control) model to formulate dynamic permission rules. The binding method is to attach a JWT token to the role node in the situation model, which contains an encrypted permission statement. Exemplarily, when the Agent initiates an operation request, it decrypts the JWT token and extracts the permission list, and dynamically calculates the actual available permissions based on the current business situation parameters (such as risk level, resource type). If the permissions are insufficient, the approval chain route is triggered (such as initiating an authorization request to a higher-level role).

[0057] Optionally, in the mechanism of converting decision logic into causal relationship in this embodiment, it is mainly implemented through structured modeling at the following three levels: 1. Causal Inference Based on Expert Experience 1. Deconstructing decision-making behavior Through the decision records of experts in specific situations (such as emergency response process selection, resource scheduling priority judgment), the logical dependency relationship between decision conditions and execution actions is extracted. For example, in emergency response, the expert's handling steps for the "material shortage" incident imply the causal chain of "resource allocation delay → reduced rescue efficiency".

[0058] 2. Formal expression of rules The emergency plan clauses formulated by experts (such as "when the Class A risk index exceeds the threshold, initiate the Class B response") are converted into If-Then causal rules to establish a deterministic mapping between event triggering conditions and disposal measures. This conversion requires parameterized modeling in combination with the spatiotemporal constraints in the business scenario (such as the scope of disaster impact).

[0059] 2. Data-driven causal discovery 1. Analysis of organizational operation data Use large language models to analyze unstructured data such as historical event reports and communication records to identify the frequently occurring "event feature-handling action-result feedback" triples. For example, extract the causal sequence of "equipment failure → production interruption → emergency repair" from an accident report.

[0060] 2. Construction of causal graph Through the Bayesian network, the probabilistic causal relationship between the behavior of collaborative participants (such as department response speed), process node status (such as approval completion) and event evolution results (such as risk diffusion degree) is quantified. This conversion requires the integration of multi-source signal correlation features in situational awareness.

[0061] 3. Dynamic Adaptation of Context Model 1. Causal chain reconstruction mechanism Establishing a variable causal relationship template in the knowledge base system allows the causal strength to be dynamically adjusted according to business scenario characteristics (such as disaster type and organizational structure). For example, in different scenarios of epidemics and earthquakes, the causal relationship weights of "information lag" and "decision-making errors" will change.

[0062] 2. Counterfactual Reasoning Support By embedding a causal inference engine, it supports the simulation of causal relationships that have not occurred but may exist (such as "if the evacuation process is initiated 2 hours in advance, the casualty rate will be reduced by X%"). This capability relies on the spatiotemporal reasoning module in the situational model.

[0063] In this embodiment, the conversion mechanism achieves the transition from discrete decision-making steps to continuous causal networks by integrating human causal cognition and machine causal discovery, which can improve the predictability and explainability of emergency decisions.

[0064] Optionally, in this embodiment, the structuring method of the time dependency is: (1) Process step breakdown An atomization approach is adopted to parse the flowchart using Business Process Model and Notation (BPMN) to extract key nodes (such as "purchase application approval" and "logistics scheduling") and node attributes (execution role, input / output data, and time threshold).

[0065] (2) Timing Constraint Analysis Identify dependency types: strong sequential dependency (node ​​B must be started strictly after node A is completed, the processing strategy is: serial execution, set checkpoints), weak parallel dependency (node ​​C and node D can be executed at the same time, the processing strategy is: enable multi-threading, resource competition control), conditional trigger dependency (node ​​E is only activated under certain conditions, the processing strategy is: event monitoring + callback mechanism) (3) Time-dependent chain generation Using time Petri net, the process sequence characteristics are transformed into a state transition model with time constraints.

[0066] In this embodiment, the design of the time dependency achieves the optimization of the executability and efficiency of the process, avoids logical deadlocks through timing constraints, and corrects the critical path analysis to facilitate the precise allocation of resources.

[0067] Optionally, in addition to time dependencies, the database system also includes data dependencies (the input data of node B must come from the output result of node A), which can be implemented by defining input / output interfaces through data flow diagrams (DFDs); resource dependencies (multiple nodes compete for the same resource pool (such as equipment, manpower, budget)), which can be implemented through mutexes and priority scheduling algorithms; permission dependencies (node ​​execution requires authorization by a specific role), which can be implemented through RBAC (role-based access control) policy libraries; logical dependencies (execution condition constraints based on business rules), which can be implemented by loading business rules through the Drools rule engine.

[0068] Optionally, this embodiment also includes data service knowledge analysis: combing the data sources inside the enterprise (such as ERP system) and outside (such as market demand forecast data, supplier delivery capacity data) to determine the data support required for order selection decisions. Establish a data pragmatic context model to ensure the accuracy and timeliness of data in the decision-making process. For example, in high-profit order selection decisions, the AI ​​Agent needs to access order profit margin data in the order management system and resource availability data in the production planning system to make reasonable order selection recommendations.

[0069] Optionally, in this embodiment, the construction of the knowledge base system includes: building a knowledge base system by classifying and labeling the knowledge of coping with uncertain events according to requirements (including AI agent requirements of business employees), roles (including business employees, tasks, responsibilities, action guidelines, decision-making styles), tasks and goals (including specific task points and goals), situations (uncertain event decision-making situations), and behaviors (including decision-making methods, knowledge, and knowledge types involved in specific tasks). Specifically: Classification: Classify knowledge according to dimensions such as tasks, roles, situations, and goals. For example, tasks can be classified as "order processing", "production scheduling", "financial review", etc., and situations can be classified as "order urgency", "resource availability", etc.

[0070] Tagging: Each category of knowledge is tagged for quick retrieval and search. For example, under the category of "order processing", tags can include "high-profit orders", "order priority", "inventory shortage", etc.

[0071] Step S103: Search the knowledge base system to obtain the decision-making knowledge nodes for various uncertain events and the relationship between the decision-making knowledge nodes; generate a task chain and a thinking chain for the AI ​​Agent based on the decision-making knowledge nodes for various uncertain events and the relationship between the decision-making knowledge nodes; generate the AI ​​Agent based on the task chain and the thinking chain.

[0072] Optionally, in this embodiment, task chain design: according to a specific decision-making scenario, relevant decision nodes and processes are retrieved from the knowledge base, and a task chain that the AI ​​Agent needs to execute when processing the decision is designed. The task chain includes steps such as planning, decision-making, and action to ensure that the AI ​​Agent can systematically complete the task of coping with uncertain events. Thinking chain construction: Based on the knowledge base, key decision scenarios and decision logic are retrieved, combined with the Chain-of-Thought (CoT) method, perception, memory, and reasoning thinking chains are designed to construct the reasoning path of the AI ​​Agent. Improve the transparency and explainability of AI Agent's decisions.

[0073] Optionally, in this embodiment, the method for generating a task chain includes: (1) Subtask decomposition method Input the process step attributes in the decision knowledge node (such as "procurement approval → risk assessment → plan execution"), each subtask must meet the single responsibility principle (such as "procurement approval" can be divided into "form verification" and "budget verification"), and mark the dependencies, and output the task chain structure represented by DAG (directed acyclic graph).

[0074] (2) Execution order and dependency definition Extract timestamp data (such as the average duration of historical tasks) from the process sequence characteristics of the knowledge base, calculate the critical path (CPM algorithm), determine the sequence of subtasks that cannot be delayed, and classify dependencies (strong sequential dependency, resource competition dependency, conditional trigger dependency).

[0075] The methods for generating thought chains include: (1) Implementation of chain thinking method Perception thinking chain design: define a unified data model (Apache Avro Schema), including fields: event type, sensor ID, timestamp, value, implement data input standardization, and define anomaly detection rules.

[0076] Memory thinking chain design: Use hybrid retrieval (vector + keyword) as the knowledge retrieval strategy, and set the knowledge validity period (if the case has not been used for more than 2 years, the similarity weight will be reduced by 50%).

[0077] Reasoning thinking chain design: Causal reasoning engine: Integrate the DoWhy library to perform four-step causal inference: modeling causal graphs (based on causal relationships in the knowledge base), identifying estimators (such as ATE average treatment effects), estimating causal effects, and refutation verification (by adding unobserved confounding factor sensitivity tests).

[0078] (2) Reasoning Path Construction Path priority calculation: Dynamic sorting based on the confidence and causal strength of knowledge nodes.

[0079] Multi-path parallel exploration: Monte Carlo Tree Search (MCTS) is used to explore the top-3 high-weight paths within a limited time, and finally select the one with the highest comprehensive score.

[0080] The binding mechanism between task chain and thought chain includes: (1) Mapping relationship definition Establish an index relationship between the task chain node and the thinking chain node, and convert the logical conditions in the thinking chain (such as "inventory safety factor < 0.5") into the execution trigger conditions of the task chain.

[0081] (2) Real-time collaborative control When a subtask in a task chain fails to execute, it can trace back to the corresponding reasoning node in the thinking chain, start the backup reasoning path (such as downgrading from "causal reasoning" to "rule matching"), generate a new task chain branch and reschedule.

[0082] Optionally, the generation of task chains and thinking chains also includes: experts' correction operations on AI decision results (such as manually adjusting the priority of solutions) are transmitted back to the knowledge base in real time, triggering incremental learning of task chains and thinking chains.

[0083] Optionally, the step of generating an AI Agent based on the task chain and the thought chain includes: Agent capability profile modeling: According to the subtask types in the task chain, define the set of AI Agent capability dimensions (including: perception dimension: data collection frequency (per second / batch processing), multi-source data type (text / sensor / image), reasoning dimension: logic chain depth (single-step / multi-step backtracking), uncertainty processing method (probability threshold / fuzzy logic), execution dimension: action response speed (real-time / delayed), operation authority level (read-only / control instructions)), and build the Agent feature vector based on the capability dimension.

[0084] Modular Agent component assembly: The LangChain framework is used to build a pluggable Agent architecture, including: Core reasoning module: relying on the reasoning logic path in the thinking chain, injecting domain knowledge based on the GPT-4 model, using LoRA (Low-Rank Adaptation) fine-tuning technology, and superimposing the rank decomposition matrix on the pre-trained weights; Task execution module: defining the Toolset tool set for each subtask, including: Data query tool: accessing the knowledge base through the SQL / NoSQL connector, logical verification tool: calling the causal inference engine to verify the rationality of the decision chain, external interaction tool: encapsulating enterprise API (such as ERP system work order creation interface); Tool calling module: by interacting with external tools, calling search engines, database query tools, mathematical operation tools, code executors and other tools for task execution; Memory management module: designing a hierarchical memory structure: short-term memory: using the LRU elimination strategy to cache the most recent interaction context, long-term memory: persisting the verified decision plan to the knowledge base.

[0085] Lightweight deployment optimization: compress the Agent model through knowledge distillation technology, including: teacher model selection: retain the full version of GPT-4 as the teacher model, student model construction: adopt the TinyBERT architecture and design the distillation loss function. Dynamically select the model version according to hardware resources during deployment.

[0086] In this embodiment, by involving business experts and employees in the task chain and decision design, the implicit knowledge of business experts and employees is embedded in the task chain and thinking chain, thereby enhancing the professionalism and adaptability of the AI ​​Agent in specific decision-making situations.

[0087] In this stage, we first retrieve relevant knowledge from the knowledge base and design the task chain and thinking chain. First, according to the specific decision-making scenario, we retrieve the relevant decision nodes and processes from the knowledge base, and design the task chain that the AI ​​Agent needs to execute when processing the decision. The task chain includes steps such as planning, decision-making, and action to ensure that the AI ​​Agent can systematically complete the task of coping with uncertain events. In addition, based on the knowledge base retrieval of key decision scenarios and decision logic, combined with the Chain-of-Thought (CoT) method, we design the perception, memory, and reasoning thinking chains to construct the reasoning path of the AI ​​Agent. Improve the transparency and explainability of the AI ​​Agent's decisions. Finally, through the participation of business experts and employees in the task chain and decision design, the implicit knowledge of business experts and employees is embedded in the task chain and thinking chain, enhancing the professionalism and adaptability of the AI ​​Agent in specific decision-making scenarios.

[0088] Step S104: receiving the user's input command, and parsing the input command through the AI ​​Agent; the AI ​​Agent infers the parsed result based on the task chain and the thinking chain to generate a decision suggestion, and generates a decision plan for dealing with uncertain events based on the task chain.

[0089] Optionally, in this embodiment, the AI ​​Agent includes a perception module, a reasoning module, a decision module and an execution module, so that the Agent can perceive the business environment, make reasoning decisions, and perform corresponding operations. The modules interact through interfaces to ensure the flexibility and scalability of the system. Among them, the perception module identifies the current business environment and user input based on the perception thinking chain, including functions such as natural language understanding, context recognition and data collection. The reasoning module performs logical reasoning and information analysis based on the reasoning thinking chain and the task chain to generate preliminary decision suggestions. The decision module generates a specific decision plan based on the task chain, combining the comprehensive reasoning results and business needs. The execution module is based on the task chain: executes the decision plan, such as generating reports, sending notifications or calling external system interfaces for operations (for example, calling a code interpreter for related analysis tasks). At this stage, the system integration capabilities of the LangChain framework are used to build the overall architecture of the AI ​​Agent. The entire system interacts with data and calls functions through a unified interface to ensure the efficient operation and stability of the AI ​​Agent.

[0090] In this embodiment, the multi-round natural language interaction mechanism between AI Agent and business personnel optimizes the decision-making process and results through continuous feedback and learning. For example, business personnel can adjust the order priority criteria through dialogue, and AI Agent regenerates the order selection plan based on the new criteria and provides corresponding analysis and reasons. By recording and analyzing the data during the interaction process, the decision model and reasoning logic of AI Agent are continuously optimized to ensure that it can meet the personalized needs of business personnel.

[0091] In the order selection decision-making process, business staff and AI Agent make collaborative decisions through dialogue, and record all interaction content, including dialogue text, decision-making process, and feedback. Use natural language understanding technology to analyze interaction logs and extract new experiences, improvement suggestions, and uncovered knowledge points in the decision-making process of business staff. Retrain and optimize AI Agent based on experience knowledge and business needs of business staff, and update the knowledge base. For example, analyze common decision-making deviations in the order selection process, optimize AI Agent's decision rules, reduce the impact of human factors, and improve the scientificity and consistency of decision-making.

[0092] In order to enhance the adaptability of AI agents in dynamic environments and enable AI agents to make more accurate and timely responses in complex and ever-changing environments, the present application provides an embodiment of an organizational knowledge intelligent application device for uncertain event response that implements all or part of the content of the organizational knowledge intelligent application method for uncertain event response, see Figure 4 The intelligent application device for organizational knowledge for dealing with uncertain events specifically includes the following contents: The knowledge extraction module 10 is used to obtain the empirical knowledge for dealing with uncertain events and extract key features from the empirical knowledge; the key features include uncertain events, business scenarios, collaborative participants, process steps and decision logic; The knowledge base system construction module 20 is used to establish a situation model based on the collaborative situation metamodel, associate the uncertain events, business situations, collaborative participants with the situation model, and form decision knowledge nodes; define the relationship between decision knowledge nodes according to the process steps and the decision logic, and construct a knowledge base system based on the relationship between the decision knowledge nodes; The AI ​​Agent generation module 30 is used to search the knowledge base system to obtain the decision-making knowledge nodes for various uncertain events and the relationship between the decision-making knowledge nodes; generate a task chain and a thinking chain for the AI ​​Agent based on the decision-making knowledge nodes for various uncertain events and the relationship between the decision-making knowledge nodes; and generate the AI ​​Agent based on the task chain and the thinking chain; The interaction module 40 is used to receive the user's input command and parse the input command through the AI ​​Agent; the AI ​​Agent infers the parsed result based on the task chain and the thinking chain to generate a decision suggestion, and generates a decision plan for dealing with uncertain events based on the task chain.

[0093] From the above description, it can be seen that the organizational knowledge intelligent application device for dealing with uncertain events provided by the embodiment of the present application obtains experiential knowledge for dealing with uncertain events from historical cases and simulation cases, and builds a knowledge base system based on the experiential knowledge, converts the experiential knowledge into nodes and relationships of the knowledge base system, and then builds an AI Agent based on its task chain and thinking chain to be responsible for executing tasks and making decisions. By efficiently extracting the insights and experiential knowledge of business experts and employees under uncertain events and integrating them into the design of AI agents, the professionalism and reliability of their decision-making are improved, thereby promoting the externalization and intelligent application of experiential knowledge, so that AI agents can make more accurate and timely responses in complex and ever-changing environments.

[0094] From the hardware level, in order to achieve precise lighting adjustment control, break through the limitations of traditional fixed-mode dimming, and provide a comprehensive technical solution for intelligent lighting systems, the present application provides an embodiment of an electronic device for implementing all or part of the content of the intelligent application method of organizational knowledge for dealing with uncertain events, and the electronic device specifically includes the following content: Processor, memory, communication interface and bus; wherein the processor, memory and communication interface communicate with each other through the bus; the communication interface is used to realize information transmission between the intelligent application device of organizational knowledge for dealing with uncertain events and related devices such as core business systems, user terminals and related databases; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., but the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiment of the intelligent application method of organizational knowledge for dealing with uncertain events and the embodiment of the intelligent application device of organizational knowledge for dealing with uncertain events in the embodiment, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0095] It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0096] In practical applications, part of the method for intelligent application of organizational knowledge for dealing with uncertain events can be executed on the electronic device side as described above, or all operations can be completed in the client device. The selection can be made based on the processing capability of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor.

[0097] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0098] Figure 5 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Figure 5 As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Figure 5 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0099] In one embodiment, the function of the intelligent application method of organizational knowledge for dealing with uncertain events can be integrated into the central processing unit 9100. The central processing unit 9100 can be configured to perform the following control: Step S101: Acquire experiential knowledge for dealing with uncertain events, and extract key features from the experiential knowledge; the key features include uncertain events, business scenarios, collaborative participants, process steps, and decision logic; Step S102: establishing a situation model based on the collaborative situation metamodel, associating the uncertain events, business situations, collaborative participants with the situation model to form decision knowledge nodes; defining the relationship between decision knowledge nodes according to the process steps and the decision logic, and constructing a knowledge base system based on the relationship between the decision knowledge nodes; Step S103: searching the knowledge base system to obtain the decision-making knowledge nodes for various uncertain events and the relationships between the decision-making knowledge nodes; generating a task chain and a thinking chain for the AI ​​Agent based on the decision-making knowledge nodes for various uncertain events and the relationships between the decision-making knowledge nodes; generating the AI ​​Agent based on the task chain and the thinking chain; Step S104: receiving the user's input command, and parsing the input command through the AI ​​Agent; the AI ​​Agent infers the parsed result based on the task chain and the thinking chain to generate a decision suggestion, and generates a decision plan for dealing with uncertain events based on the task chain.

[0100] From the above description, it can be seen that the electronic device provided in the embodiment of the present application obtains experiential knowledge for dealing with uncertain events from historical cases and simulation cases, and builds a knowledge base system based on the experiential knowledge, converts the experiential knowledge into nodes and relationships of the knowledge base system, and then builds an AI Agent based on its task chain and thinking chain to be responsible for executing tasks and making decisions. By efficiently extracting and integrating the insights and experiential knowledge of business experts and employees under uncertain events into the design of AI agents, the professionalism and reliability of their decision-making are improved, thereby promoting the externalization and intelligent application of experiential knowledge, so that AI agents can make more accurate and timely responses in complex and ever-changing environments.

[0101] In another embodiment, the intelligent application device of organizational knowledge for responding to uncertain events can be configured separately from the central processing unit 9100. For example, the intelligent application device of organizational knowledge for responding to uncertain events can be configured as a chip connected to the central processing unit 9100, and the function of the intelligent application method of organizational knowledge for responding to uncertain events can be realized through the control of the central processing unit.

[0102] like Figure 5 As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Figure 5 In addition, the electronic device 9600 may also include Figure 5 For components not shown, reference may be made to the prior art.

[0103] like Figure 5 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device. The central processing unit 9100 receives input and controls the operation of various components of the electronic device 9600.

[0104] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0105] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0106] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0107] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0108] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0109] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module. The communication module 9110 (transmitter / receiver) is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0110] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps of the method for intelligently applying organizational knowledge for dealing with uncertain events in the above embodiments, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all the steps of the method for intelligently applying organizational knowledge for dealing with uncertain events in the above embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented: Step S101: Acquire experiential knowledge for dealing with uncertain events, and extract key features from the experiential knowledge; the key features include uncertain events, business scenarios, collaborative participants, process steps, and decision logic; Step S102: establishing a situation model based on the collaborative situation metamodel, associating the uncertain events, business situations, collaborative participants with the situation model to form decision knowledge nodes; defining the relationship between decision knowledge nodes according to the process steps and the decision logic, and constructing a knowledge base system based on the relationship between the decision knowledge nodes; Step S103: searching the knowledge base system to obtain the decision-making knowledge nodes for various uncertain events and the relationships between the decision-making knowledge nodes; generating a task chain and a thinking chain for the AI ​​Agent based on the decision-making knowledge nodes for various uncertain events and the relationships between the decision-making knowledge nodes; generating the AI ​​Agent based on the task chain and the thinking chain; Step S104: receiving the user's input command, and parsing the input command through the AI ​​Agent; the AI ​​Agent infers the parsed result based on the task chain and the thinking chain to generate a decision suggestion, and generates a decision plan for dealing with uncertain events based on the task chain.

[0111] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application obtains experiential knowledge for dealing with uncertain events from historical cases and simulation cases, and builds a knowledge base system based on the experiential knowledge, converts the experiential knowledge into nodes and relationships of the knowledge base system, and then builds an AI Agent based on its task chain and thinking chain to be responsible for executing tasks and making decisions. By efficiently extracting and integrating the insights and experiential knowledge of business experts and employees under uncertain events into the design of AI agents, the professionalism and reliability of their decision-making are improved, thereby promoting the externalization and intelligent application of experiential knowledge, so that AI agents can make more accurate and timely responses in complex and ever-changing environments.

[0112] The embodiments of the present application also provide a computer program product capable of implementing all the steps of the method for intelligently applying organizational knowledge for dealing with uncertain events in the above embodiments, where the execution subject is a server or a client. When the computer program / instruction is executed by a processor, the steps of the method for intelligently applying organizational knowledge for dealing with uncertain events are implemented. For example, the computer program / instruction implements the following steps: Step S101: Acquire experiential knowledge for dealing with uncertain events, and extract key features from the experiential knowledge; the key features include uncertain events, business scenarios, collaborative participants, process steps, and decision logic; Step S102: establishing a situation model based on the collaborative situation metamodel, associating the uncertain events, business situations, collaborative participants with the situation model to form decision knowledge nodes; defining the relationship between decision knowledge nodes according to the process steps and the decision logic, and constructing a knowledge base system based on the relationship between the decision knowledge nodes; Step S103: searching the knowledge base system to obtain the decision-making knowledge nodes for various uncertain events and the relationships between the decision-making knowledge nodes; generating a task chain and a thinking chain for the AI ​​Agent based on the decision-making knowledge nodes for various uncertain events and the relationships between the decision-making knowledge nodes; generating the AI ​​Agent based on the task chain and the thinking chain; Step S104: receiving the user's input command, and parsing the input command through the AI ​​Agent; the AI ​​Agent infers the parsed result based on the task chain and the thinking chain to generate a decision suggestion, and generates a decision plan for dealing with uncertain events based on the task chain.

[0113] From the above description, it can be seen that the computer program product provided by the embodiment of the present application obtains experiential knowledge for dealing with uncertain events from historical cases and simulation cases, and builds a knowledge base system based on the experiential knowledge, converts the experiential knowledge into nodes and relationships of the knowledge base system, and then builds an AI Agent based on its task chain and thinking chain to be responsible for executing tasks and making decisions. By efficiently extracting and integrating the insights and experiential knowledge of business experts and employees under uncertain events into the design of AI agents, the professionalism and reliability of their decision-making are improved, thereby promoting the externalization and intelligent application of experiential knowledge, so that AI agents can make more accurate and timely responses in complex and ever-changing environments.

[0114] It should be understood by those skilled in the art that embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0115] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0116] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0117] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0118] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. An intelligent application method of organizational knowledge for dealing with uncertain events, characterized in that: The method comprises: Acquire experiential knowledge on coping with uncertain events, and extract key features from the experiential knowledge; the key features include uncertain events, business scenarios, collaborative participants, process steps, and decision logic; Establish a situation model based on the collaborative situation metamodel, associate the uncertain events, business situations, collaborative participants with the situation model, and form decision knowledge nodes; define the relationship between decision knowledge nodes according to the process steps and the decision logic, and build a knowledge base system based on the relationship between the decision knowledge nodes; Searching the knowledge base system to obtain the decision-making knowledge nodes for various uncertain events and the relationship between the decision-making knowledge nodes; generating a task chain and a thinking chain for the AI ​​Agent based on the decision-making knowledge nodes for various uncertain events and the relationship between the decision-making knowledge nodes; generating an AI Agent based on the task chain and the thinking chain; Receive input commands from the user, and parse the input commands through the AI ​​Agent; the AI ​​Agent infers the results of the analysis based on the task chain and the thinking chain to generate decision suggestions, and generates a decision plan for dealing with uncertain events based on the task chain.

2. The method for intelligent application of organizational knowledge for dealing with uncertain events according to claim 1 is characterized in that: The steps of acquiring experience knowledge for dealing with uncertain events include: Obtain review files through the review guide for coping with uncertain events, and obtain the first-hand experience knowledge of uncertain events from the review files; A large language model is used to parse historical data to obtain second-level empirical knowledge of uncertain events; the historical data includes historical cases and meeting minutes.

3. The method for intelligent application of organizational knowledge for dealing with uncertain events according to claim 1 is characterized in that: The step of associating the uncertain events, business scenarios, collaborative participants and scenario models to form decision knowledge nodes includes: Using uncertain events as trigger conditions, dynamic response is achieved through event-driven nodes in the situation model; Acquire operational data related to the business scenario and convert the operational data into dynamic constraint variables of the scenario model; Collaborative participants are defined as role nodes in the situation model and their decision-making authority is bound.

4. The method for intelligent application of organizational knowledge for dealing with uncertain events according to claim 1 is characterized in that: The step of defining the relationship between decision knowledge nodes according to the process steps and the decision logic comprises: ‌Decomposing the decision logic into a logical chain between decision conditions and execution actions, and converting the logical chain into causal rules, establishing a deterministic mapping between event triggering conditions and disposal measures, and forming an initial causal rule library; Establish a variable causal relationship template based on the initial causal rule base, allowing the causal strength to be dynamically adjusted according to the business scenario, and support the simulation of causal relationships that may exist in the future by embedding a causal inference engine; It also includes: decomposing the process steps into independent decision knowledge nodes, analyzing the timing constraints between operations in the process steps, extracting process sequence features, and converting the process sequence features into a timing dependency chain between the decision knowledge nodes to obtain a time dependency relationship.

5. The method for intelligent application of organizational knowledge for dealing with uncertain events according to claim 1 is characterized in that: The step of generating a task chain and a thinking chain for the AI ​​Agent based on the decision knowledge nodes for various uncertain events and the relationship between the decision knowledge nodes includes: Decompose the execution process for uncertain events into a series of subtasks, determine the execution order and dependencies between the subtasks, and generate a task chain for AI Agents; Based on the decision logic for uncertain events and combined with the chain thinking method, we designed the perception thinking chain, memory thinking chain and reasoning thinking chain, and constructed the reasoning path of AI Agent to obtain the thinking chain for AI Agent. For each subtask in the task chain, a set of associated thinking chain nodes is specified to form a mapping relationship table.

6. The method for intelligent application of organizational knowledge for dealing with uncertain events according to claim 1 is characterized in that: The step of generating an AI Agent based on the task chain and the thought chain includes: According to the subtask types in the task chain, a set of capability dimensions of the AI ​​Agent is defined, and an Agent feature vector is constructed based on the capability dimensions; The LangChain framework is used to build the Agent model architecture, which includes a core reasoning module, a task execution module, a tool calling module and a memory management module; The Agent model is compressed through knowledge distillation technology, and the model version is dynamically selected according to hardware resources to obtain the final AI Agent.

7. The method for intelligent application of organizational knowledge for dealing with uncertain events according to claim 1 is characterized in that: The AI ​​Agent generates decision suggestions by reasoning about the results of the analysis based on the task chain and the thinking chain, and generates a decision plan for dealing with uncertain events based on the task chain. The step also includes: recording the interaction content between the AI ​​Agent and the user, extracting new decision-making experience from the interaction content, and optimizing the knowledge base system based on the new decision-making experience.

8. An intelligent application device for organizational knowledge for dealing with uncertain events, characterized in that: The device comprises: A knowledge extraction module is used to obtain experiential knowledge for dealing with uncertain events and extract key features from the experiential knowledge; the key features include uncertain events, business scenarios, collaborative participants, process steps, and decision logic; A knowledge base system construction module is used to establish a situation model based on the collaborative situation metamodel, associate the uncertain events, business situations, collaborative participants with the situation model, and form decision knowledge nodes; define the relationship between decision knowledge nodes according to the process steps and the decision logic, and construct a knowledge base system based on the relationship between the decision knowledge nodes; An AI Agent generation module is used to search the knowledge base system to obtain the decision-making knowledge nodes for various uncertain events and the relationship between the decision-making knowledge nodes; generate a task chain and a thinking chain for the AI ​​Agent based on the decision-making knowledge nodes for various uncertain events and the relationship between the decision-making knowledge nodes; and generate an AI Agent based on the task chain and the thinking chain; The interactive module is used to receive the user's input command and parse the input command through the AI ​​Agent; the AI ​​Agent infers the parsed result based on the task chain and the thinking chain to generate a decision suggestion, and generates a decision plan for dealing with uncertain events based on the task chain.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method for intelligent application of organizational knowledge for dealing with uncertain events as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for intelligently applying organizational knowledge for coping with uncertain events as described in any one of claims 1 to 7 are implemented.

Images