Event-driven business agent dynamic construction method and device

Abstract

The invention discloses an event-driven business agent dynamic construction method and equipment, belongs to the technical field of intelligent application, and is used for solving the technical problems that the intelligent transformation cost of a traditional business system is high, and a manually constructed business agent is logically solidified and cannot quickly cope with business requirements and environment changes. The method comprises the following steps: monitoring a service event occurring in a service system, obtaining corresponding service event information and extracting upstream service context information corresponding to the service event; identifying a downstream service of the service event in the service system according to the service event and the upstream service context information, and acquiring service information of the downstream service; based on the business event information and the business information of the downstream business, constructing a business agent for executing the downstream business action, and storing the business agent to a business agent center; and when the new business event is monitored, a corresponding business agent is matched in the business agent center, so that the business agent executes a downstream business action triggered by the business event.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent applications, and particularly to an event-driven method and device for dynamically constructing business agents. Background Art

[0002] In the new development stage of the deep integration of the digital economy and the real economy, the technology of large-scale pre-trained models based on deep learning is evolving at an exponential rate, providing the core driving force for the intelligent transformation of enterprise-level application systems. The systematic defects commonly existing in traditional business systems, such as rigid rules, lagging responses, and decision-making blind spots, are becoming increasingly prominent in the dynamic market environment and personalized service requirements, and have become the key shackles restricting the digital transformation of enterprises.

[0003] The current mainstream business agent construction solutions have significant technical limitations: The first type is the hard-coded implementation method based on the intelligent agent development framework. Although it can achieve basic process automation through technologies such as the BPMN engine and the rule engine, developers need to pre-define all decision-making logics and execution paths in advance. Moreover, when business rules change, a large amount of code modification is required, resulting in an exponential increase in system maintenance costs. The second type is the visual configuration solution relying on the large model platform. Although it shortens the development cycle by dragging and dropping components with low code, its built-in predefined templates and fixed decision tree structures essentially still belong to the static rule assembly mode, with a large system response delay, exposing the fundamental defect of dynamic adaptation ability.

[0004] Both of these two methods rely on manual pre-construction. There are a large number of business agents, with high construction costs, and the logic of the manually constructed business agents is rigid and cannot quickly respond to changes in business requirements and the environment. Summary of the Invention

[0005] Embodiments of the present invention provide an event-driven method and device for dynamically constructing business agents to solve the following technical problems: The intelligent transformation cost of traditional business systems is high, and the logic of the manually constructed business agents is rigid and cannot quickly respond to changes in business requirements and the environment.

[0006] Embodiments of the present invention adopt the following technical solutions: On the one hand, embodiments of the present invention provide an event-driven method for dynamically constructing business agents, and the method includes: Monitoring business events occurring in the business system, obtaining corresponding business event information, and extracting the upstream business context information corresponding to the business events; According to the business events and the upstream business context information, identifying the downstream business of the business events in the business system, and obtaining the business information of the downstream business; Based on the business event information and the business information of the downstream business, construct a business agent for performing downstream business actions and store it in the business agent center; wherein, the business agent at least includes at least one key attribute of user role, target large model, business prompt words, and business tool services; When a new business event is monitored, match the corresponding business agent in the business agent center so that the business agent performs the downstream business actions triggered by the business event.

[0007] In a feasible implementation manner, monitor business events occurring in the business system, obtain the corresponding business event information and extract the upstream business context information corresponding to the business event, specifically including: Monitor business events occurring in the current business system based on a polling mechanism; Obtain the business event information of the business event; wherein, the business event information at least includes event type, event summary, event status, and event source; Understand the semantics of the business event information through natural language processing technology NLP and knowledge graph technology, and identify the upstream business associated with the business event according to the semantics; Extract the business information of the upstream business to obtain the upstream business context information; wherein, the upstream business context information at least includes the business function, business operation, business entity, and operating user role of the upstream business.

[0008] In a feasible implementation manner, according to the business event and the upstream business context information, identify the downstream business of the business event in the business system and obtain the business information of the downstream business, specifically including: Pre-create a domain knowledge graph between business events in the business system; wherein, the domain knowledge graph at least includes business entities, business actions, business services, and business event relationships; According to the business event and the upstream business context information, find the downstream business affected by the business event in the domain knowledge graph; Extract the business information of the downstream business; wherein, the business information of the downstream business at least includes business actions, the user role of the business actions, the business services corresponding to the business actions, and the business entities corresponding to the business actions.

[0009] In a feasible implementation manner, before constructing a business agent for performing downstream business actions based on the business event information and the business information of the downstream business, the method further includes: Build a large model service center and store several vertical domain large models in the large model service center; wherein, the large model service center can access large model services of multiple manufacturers, multiple domains, and multiple parameter sizes, and each large model service includes at least the following self-description information: model classification, proficient field, model ranking, and semantic description; The large model service center provides a unified execution interface for large model services and dynamically switches the executed vertical domain large models based on the self-description information of the large model services.

[0010] In a feasible implementation manner, based on the business event information and the business information of the downstream business, build a business agent for performing the downstream business actions, which specifically further includes: After obtaining the event summary in the business event information, first search for the business agent of the downstream business that matches the business event in the pre-built business agent center; If a matching business agent is found, directly extract it; if no matching business agent is found, then build a business agent that can perform the downstream business actions of the business event based on a preset program.

[0011] In a feasible implementation manner, building a business agent that can perform the downstream business actions of the business event based on a preset program specifically includes: Extract the large model features required for building the business agent according to the business event information and the business information of the downstream business; wherein, the large model features at least include model classification and the proficient field of the model; Based on the large model features, match the vertical domain large model in the large model service center that has both the matching field and classification and the highest ranking as the target large model; Generate corresponding business prompt words according to the business event information and the business information of the downstream business; Recommend or generate business tool services according to the business service information corresponding to the business actions of the downstream business; Dynamically assemble the user role corresponding to the business event, the target large model, the business prompt words, and the business tool services onto the business agent to obtain a business agent for performing the downstream business actions, and store the business agent entity in the pre-built business agent center; the stored information also includes the business event information and the relationship between the business event and the downstream business.

[0012] In a feasible implementation manner, generating corresponding business prompt words according to the business event information and the business information of the downstream business specifically includes: Dynamically recall the best prompt template from the pre - constructed prompt template center according to the event summary in the business event information and the business information of the downstream business; wherein, the prompt template at least includes the following dynamic parameters: user role, business function, business operation, and business entity. Replace the dynamic parameters in the prompt template with the parameter values extracted from the upstream business context information to obtain a business prompt that conforms to the business event and the downstream business.

[0013] In a feasible implementation manner, recommend or generate a business tool service according to the business service information corresponding to the business action of the downstream business, specifically including: Obtain the business service corresponding to the business action in the business information of the downstream business, and extract the corresponding business service information; wherein, the business service information at least includes service description, service address, service parameter list, and return value. Based on the business service information, match and recommend a business tool service in the pre - constructed agent tool service center. If no match is found, generate a corresponding business tool service based on a preset script and load it into the pre - constructed agent tool service center.

[0014] In a feasible implementation manner, when a new business event is monitored, match the corresponding business agent in the business agent center so that the business agent executes the downstream business action triggered by the business event, specifically including: When a new business event is monitored, match the business agent of the corresponding downstream business in the business agent center based on the upstream business context information of the new business event. Based on the downstream business organization dialogue information, conduct a dialogue with the matched business agent so that the business agent executes the downstream business action triggered by the business event.

[0015] On the other hand, an embodiment of the present invention also provides an event - driven business agent dynamic construction device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions that can be executed by the at least one processor, so that the at least one processor can execute the event - driven business agent dynamic construction method described above.

[0016] Compared with the prior art, the event - driven business agent dynamic construction method and device provided by the embodiment of the present invention have the following beneficial effects: Through the solution for dynamically constructing a business intelligent agent driven by events provided by the embodiments of the present invention, it is possible to directly monitor business events during the business execution process, dynamically extract the upstream business context, identify the downstream business affected by the business event, and construct in real time a business intelligent agent for executing the downstream business actions. Compared with the method of manually constructing an intelligent agent, the construction cost for the transformation and upgrading of the business system is significantly reduced. By dynamically constructing a business intelligent agent at runtime, the flexibility of the business intelligent capabilities is improved, and it can adapt to business requirements and environmental changes agilely. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. In the drawings: Figure 1 It is a flowchart of a method for dynamically constructing an event-driven business intelligent agent provided by an embodiment of the present invention; Figure 2 It is a schematic structural diagram of a device for dynamically constructing an event-driven business intelligent agent provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] To enable those skilled in the art to better understand the technical solutions in the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0019] The embodiments of the present invention provide a method for dynamically constructing an event-driven business intelligent agent. As Figure 1 shown, the method for dynamically constructing an event-driven business intelligent agent specifically includes steps S101 - S104: S101. Monitor the business events occurring in the business system, obtain the corresponding business event information, and extract the upstream business context information corresponding to the business event.

[0020] Specifically, monitor the business events occurring in the current business system based on a polling mechanism. Then obtain the business event information of the business event; wherein, the business event information at least includes the event type, event summary, event status, and event source.

[0021] Further, the semantics of business event information are understood through natural language processing technology (NLP) and knowledge graph technology, and the upstream businesses associated with the business event are identified based on the semantics. The business information of the upstream businesses is extracted to obtain upstream business context information; wherein, the upstream business context information at least includes the business functions, business operations, business entities, and operating user roles of the upstream businesses.

[0022] As a feasible implementation, information such as the event type, event summary, event status, and event source of the business event is obtained; NLP and knowledge graph technology are used to understand the semantics of the business event, and the upstream business functions, business operations, business entities, and operating user roles associated with the business event are identified to form the upstream business context.

[0023] S102. According to the business event and the upstream business context information, identify the downstream businesses of the business event in the business system and obtain the business information of the downstream businesses.

[0024] Specifically, a domain knowledge graph between business events is pre-created in the business system. According to the business event and the upstream business context information, search for the downstream businesses affected by this business event in the domain knowledge graph. Among them, the domain knowledge graph at least includes business entities, business actions, business services, and business event relationships.

[0025] Further, the business information of the downstream businesses is extracted; wherein, the business information of the downstream businesses at least includes business actions, the user roles of the business actions, the business services corresponding to the business actions, and the business entities corresponding to the business actions.

[0026] As a feasible implementation, the information of the downstream businesses includes information such as business actions, the user roles of the business actions, the business services corresponding to the business actions, and the business entities corresponding to the business actions. The business information of the downstream businesses is obtained through the knowledge reasoning service of the graph database.

[0027] In one embodiment, if the business event is that order 123 is successfully paid, then the upstream context information of this business event includes: business function: payment; business operation: transfer; business entity: the commodity details included in order 123; operating user role: the order-placing user information (user ID, delivery address, etc.). After the monitoring service or event processing service receives the event, using the key identifier in the event (such as the order ID), by calling the APIs of relevant business systems (order service API to query order details, user service API to query user information), querying the shared database, or querying the cache, etc., actively obtain the data required to construct the complete upstream context. Then the downstream businesses of this business event are: notifying the warehouse system to deduct the inventory of the corresponding commodities, notifying the logistics system to generate a shipping order for ordinary express delivery, and notifying the customer service system to update the order status.

[0028] S103. Based on the business event information and the business information of the downstream business, construct a business agent for executing the downstream business actions and store it in the business agent center; wherein, the business agent includes at least one of the key attributes such as user role, target large model, business prompt words, and business tool services.

[0029] Specifically, to achieve the automatic construction of business agents, the present invention pre-constructs a plurality of auxiliary construction centers, including a large model service center, a prompt word template center, an agent tool service center, and a business agent center.

[0030] First, construct a large model service center and store several vertical domain large models in the large model service center; wherein, the large model service center can access large model services of multiple manufacturers, multiple fields, and multiple parameter sizes, and each large model service includes at least the following self-description information: model classification, good at fields, model ranking, and semantic description; the large model service center provides a unified execution interface for the large model service and dynamically switches the executed vertical domain large model based on the self-description information of the large model service.

[0031] Then, construct the prompt word templates for each agent and store them in the prompt word template center. Construct the business tool services for each agent and store them in the agent tool service center. The business agent center is used to save the business agents constructed each time a business event occurs.

[0032] After the above preparations, obtain the event summary in the business event information, and first search for the business agent of the downstream business that matches the business event in the pre-constructed business agent center. If a matching business agent is found, directly extract it; if no matching business agent is found, construct a business agent that can execute the downstream business actions of the business event based on a preset program.

[0033] The specific process of constructing the business agent is as follows: First, according to the business event information and the business information of the downstream business, extract the large model features required for constructing the business agent; wherein, the large model features include at least model classification and the fields that the model is good at. Based on the large model features, match the vertical domain large model in the large model service center that has both a matching field and classification and the highest ranking as the target large model.

[0034] Further, according to the business event information and the business information of the downstream business, corresponding business prompt words are generated. According to the business service information corresponding to the business actions of the downstream business, a business tool service is constructed. Then, the user role, target large model, business prompt words, and business tool service corresponding to the business event are dynamically assembled onto the business agent to obtain a business agent for executing the downstream business actions, and the business agent entity is stored in the pre-constructed business agent center; the stored information also includes the business event information and the relationship between the business event and the downstream business.

[0035] As a feasible implementation, generating corresponding business prompt words according to the business event information and the business information of the downstream business specifically includes: dynamically recalling the best prompt word template from the pre-constructed prompt word template center according to the event summary in the business event information and the business information of the downstream business; wherein, the prompt word template at least includes the following dynamic parameters: user role, business function, business operation, and business entity. Replace the dynamic parameters in the prompt word template with the parameter values extracted from the upstream business context information to obtain business prompt words that conform to the business event and the downstream business.

[0036] As a feasible implementation, recommending or generating a business tool service according to the business service information corresponding to the business actions of the downstream business specifically includes: obtaining the business service corresponding to the business action in the business information of the downstream business and extracting the corresponding business service information; wherein, the business service information at least includes a service description, a service address, a service parameter list, and a return value. Then, based on the business service information, match and recommend a business tool service in the pre-constructed agent tool service center. If no match is found, generate a corresponding business tool service based on a preset script and load it into the pre-constructed agent tool service center.

[0037] S104. When a new business event is monitored, match the corresponding business agent in the business agent center so that the business agent executes the downstream business actions triggered by the business event.

[0038] Specifically, when a new business event is monitored, match the business agent of the corresponding downstream business in the business agent center based on the upstream business context information of the new business event. Then, based on the downstream business organization dialogue information, have a dialogue with the matched business agent so that the business agent executes the downstream business actions triggered by the business event.

[0039] In one embodiment, when a new business event is monitored and the business event is: Payment for Order 124 is successful. Then extract the upstream business context information of the business event: Business function: Payment; Business operation: Transfer; Business entity: Product details included in Order 124; Operating user role: Placing user information (user ID, delivery address, etc.). According to the above information, match the business agent of the corresponding downstream business in the business agent center. Since the same type of event has occurred before, the business agent center can match the business agent of the corresponding downstream business. At this time, organize the conversation information according to the downstream business actions of the business event and have a conversation with the matched business agent, so that the business agent can execute the downstream business actions of the business event based on the target large model, business prompt words, and business tool services assembled by it.

[0040] In addition, an embodiment of the present invention also provides an event-driven business agent dynamic construction device, as Figure 2 shown, the event-driven business agent dynamic construction device specifically includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, so that the at least one processor can execute: Monitor business events occurring in the business system, obtain the corresponding business event information, and extract the upstream business context information corresponding to the business event; According to the business event and the upstream business context information, identify the downstream business of the business event in the business system and obtain the business information of the downstream business; Based on the business event information and the business information of the downstream business, construct a business agent for executing downstream business actions and store it in the business agent center; wherein, the business agent at least includes at least one key attribute of user role, target large model, business prompt words, and business tool services; When a new business event is monitored, match the corresponding business agent in the business agent center, so that the business agent executes the downstream business actions triggered by the business event.

[0041] The systems, devices, modules, or units illustrated in the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0042] For the convenience of description, when describing the above device, it is divided into various units according to functions and described separately. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0043] Those skilled in the art should understand that the embodiments of this specification can be provided as a method, a system, or a computer program product. Therefore, the embodiments of this specification can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of this specification can 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.) that contain computer-usable program code.

[0044] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, commodity or device including the said element.

[0045] This specification can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification can also be practiced in a distributed computing environment, where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0046] Each embodiment in this specification is described in a progressive manner. The same or similar parts between each embodiment can be referred to each other, and the key point of each embodiment is to illustrate the differences from other embodiments. In particular, for the embodiments of the device, equipment, and non-volatile computer storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments.

[0047] The above description has been made of specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the acts or steps recited in the claims can be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require the particular order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0048] The above is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, various changes and modifications can be made to one or more embodiments of this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of one or more embodiments of this specification shall be included within the scope of the claims of this specification.

Claims

1. An event-driven method for dynamically constructing business agents, characterized in that, The method includes: Listening for business events occurring in the business system, obtaining corresponding business event information, and extracting the upstream business context information corresponding to the business event; Identifying the downstream business of the business event in the business system according to the business event and the upstream business context information, and obtaining the business information of the downstream business; Based on the business event information and the business information of the downstream business, constructing a business agent for executing the downstream business action and storing it in the business agent center; wherein, the business agent at least includes at least one key attribute of user role, target large model, business prompt words, and business tool service; When a new business event is monitored, matching the corresponding business agent in the business agent center to enable the business agent to execute the downstream business action triggered by the business event.

2. The method for dynamically constructing an event-driven business agent according to claim 1, wherein Listening for business events occurring in the business system, obtaining corresponding business event information, and extracting the upstream business context information corresponding to the business event specifically includes: Listening for business events occurring in the current business system based on a polling mechanism; Obtaining the business event information of the business event; wherein, the business event information at least includes event type, event summary, event status, and event source; Understanding the semantics of the business event information through natural language processing technology NLP and knowledge graph technology, and identifying the upstream business associated with the business event according to the semantics; Extracting the business information of the upstream business to obtain the upstream business context information; wherein, the upstream business context information at least includes the business function, business operation, business entity, and operating user role of the upstream business.

3. The method for dynamically constructing an event-driven business agent according to claim 1, characterized in that Identifying the downstream business of the business event in the business system according to the business event and the upstream business context information, and obtaining the business information of the downstream business specifically includes: Pre-creating a domain knowledge graph between business events in the business system; wherein, the domain knowledge graph at least includes business entities, business actions, business services, and business event relationships; Searching for the downstream business affected by the business event in the domain knowledge graph according to the business event and the upstream business context information; Extracting the business information of the downstream business; wherein, the business information of the downstream business at least includes business actions, the user role of the business actions, the business services corresponding to the business actions, and the business entities corresponding to the business actions.

4. A method for dynamically constructing an event-driven business agent according to claim 1, characterized in that, Before constructing a business agent for executing the downstream business action based on the business event information and the business information of the downstream business, the method further includes: Constructing a large model service center and storing a number of vertical domain large models in the large model service center; wherein, the large model service center can access large model services of multiple manufacturers, multiple domains, and multiple parameter sizes, and each large model service at least includes the following self-description information: model classification, good at fields, model ranking, and semantic description; The large model service center provides a unified execution interface for large model services and dynamically switches the executed vertical domain large models based on the self-description information of the large model services.

5. The method for dynamically constructing an event-driven business agent according to claim 1, wherein Construct a business agent for performing downstream business actions based on the business event information and the business information of the downstream business. Specifically, it further includes: After obtaining the event summary in the business event information, first search for the business agent of the downstream business that matches the business event in the pre-constructed business agent center; If a matching business agent is found, directly extract it. If no matching business agent is found, construct a business agent that can perform the downstream business actions of the business event based on a preset program.

6. An event-driven business agent dynamic construction method according to claim 5, characterized in that, Construct a business agent that can perform the downstream business actions of the business event based on a preset program. Specifically, it includes: Extract the large model features required for constructing the business agent according to the business event information and the business information of the downstream business; among them, the large model features at least include model classification and the fields that the model is good at; Based on the large model features, match the vertical domain large model with the same field and classification and the highest ranking in the large model service center as the target large model; Generate corresponding business prompt words according to the business event information and the business information of the downstream business; Recommend or generate business tool services according to the business service information corresponding to the business actions of the downstream business; Dynamically assemble the user role corresponding to the business event, the target large model, the business prompt words, and the business tool services onto the business agent to obtain a business agent for performing downstream business actions, and store the business agent entity in the pre-constructed business agent center; the stored information also includes the business event information and the relationship between the business event and the downstream business.

7. An event-driven business agent dynamic construction method according to claim 6, characterized in that Generate corresponding business prompt words according to the business event information and the business information of the downstream business. Specifically, it includes: Dynamically recall the best prompt word template from the pre-constructed prompt word template center according to the event summary in the business event information and the business information of the downstream business; among them, the prompt word template at least contains the following dynamic parameters: user role, business function, business operation, and business entity; Replace the dynamic parameters in the prompt word template with the parameter values extracted from the upstream business context information to obtain business prompt words that conform to the business event and the downstream business.

8. An event-driven business agent dynamic construction method according to claim 6, characterized in that Recommend or generate business tool services according to the business service information corresponding to the business actions of the downstream business. Specifically, it includes: Obtain the business service corresponding to the business action in the business information of the downstream business and extract the corresponding business service information; among them, the business service information at least includes service description, service address, service parameter list, and return value; Based on the business service information, match the recommended business tool service in the pre-constructed agent tool service center; If no match is found, generate the corresponding business tool service based on a preset script and load it into the pre-constructed agent tool service center.

9. An event-driven business agent dynamic construction method according to claim 1, characterized in that When a new business event is monitored, match the corresponding business agent in the business agent center so that the business agent performs the downstream business actions triggered by the business event. Specifically, it includes: When a new service event is monitored, a business agent of a corresponding downstream service is matched in the business agent center based on the upstream service context information of the new service event; Based on the downstream service organization dialogue information, a dialogue is carried out with the matched business agent so that the business agent executes the downstream service actions triggered by the service event.

10. An event-driven business intelligent agent dynamic construction device, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, so that the at least one processor can execute an event-driven business agent dynamic construction method according to any one of claims 1-9.

Images