Iterative method, device and equipment for co-creating intelligent agent, and readable storage medium

Abstract

The application discloses an industry-technology co-creation intelligent agent iteration method and device, equipment and a readable storage medium, relates to the technical field of intelligent agent iteration, and is applied to an industry-technology co-creation intelligent agent. The industry-technology co-creation intelligent agent comprises a change factor unit for debugging by a business personnel. Change rule data received by the change factor unit is acquired, and a to-be-debugged environment is constructed according to the change rule data. Full-amount historical data of the industry-technology co-creation intelligent agent is acquired, and an iteration evaluation result is determined in the to-be-debugged environment according to the full-amount historical data. When the iteration evaluation result meets a preset iteration condition, the industry-technology co-creation intelligent agent is iterated according to the change rule data. The application provides a new intelligent agent iteration mode, and realizes intelligent agent iteration without the participation of a developer.

Description

Technical Field

[0001] This application relates to the field of intelligent agent iteration technology, and in particular to an intelligent agent iteration method, apparatus, device and readable storage medium created through industry-technology collaboration. Background Technology

[0002] As intelligent agents are increasingly used in various fields, users are placing higher demands on the iteration methods of intelligent agents.

[0003] Traditional agent iteration relies on iteration guidelines provided by business personnel. Developers then iterate the agent based on these guidelines, meaning the entire iteration process and its effects are determined by the developers. This approach has limitations, as it necessitates developer involvement to complete the iteration. In other words, the requirement for developer participation restricts the scope of agent iteration.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this application is to provide a collaborative intelligent agent iteration method, apparatus, device, and readable storage medium, aiming to provide a new intelligent agent iteration approach that enables intelligent agent iteration without the involvement of developers.

[0006] To achieve the above objectives, this application provides an iterative method for business-technology co-creation of intelligent agents. This method is applied to a business-technology co-creation intelligent agent, which includes a change factor unit debugged by business personnel. The iterative method for business-technology co-creation of intelligent agents includes: The change rule data received by the change factor unit is obtained, and the debugging environment is constructed based on the change rule data; Obtain the full historical data of the intelligent agent created through industry-technology collaboration, and determine the iterative evaluation results based on the full historical data in the environment to be debugged; When the iterative evaluation results meet the preset iterative conditions, the intelligent agent of the industry-technology co-creation is iterated according to the change rule data.

[0007] In one embodiment, the collaborative intelligent agent further includes a fixed logic unit for developers to debug, and the step of constructing the debugging environment based on the change rule data includes: Acquire the solidified logic data of the solidified logic unit, and determine the initial debugging environment based on the change rule data and the solidified logic data, and determine the debugging verification result of the preset debugging data in the initial debugging environment; When the debugging verification result is that the debugging verification is successful, the initial debugging environment is used as the environment to be debugged.

[0008] In one embodiment, the step of determining the iterative evaluation result based on the full historical data in the environment to be debugged includes: For each evaluation data point in the full historical data, the output result data is determined in the environment to be debugged, and the target application scenario corresponding to the environment to be debugged is determined. The iterative evaluation results are determined based on the output data and the target application scenario.

[0009] In one embodiment, the step of determining the iterative evaluation result based on the output result data and the target application scenario includes: When the target application scenario is a first application scenario based on accuracy, the evaluation accuracy between the output result data and the preset standard result is determined, and the evaluation accuracy is used as the iterative evaluation result; When the target application scenario is a second application scenario based on the success rate, the evaluation success rate between the output result data and the preset success result is determined, and the evaluation success rate is used as the iterative evaluation result; When the target application scenario is a third application scenario based on the accuracy rate, the evaluation accuracy rate between the output result data and the preset correct result is determined, and the evaluation accuracy rate is used as the iterative evaluation result.

[0010] In one embodiment, the preset iteration conditions include an accuracy threshold, a success rate threshold, and a correctness threshold. Following the step of determining the iteration evaluation result based on the full historical data in the debugging environment, the method includes: If the evaluation accuracy in the iterative evaluation results is greater than the accuracy threshold, then the iterative evaluation results of the first application scenario are determined to meet the preset iteration conditions. If the success rate in the iterative evaluation results is greater than the success rate threshold, then the iterative evaluation results of the second application scenario are determined to meet the preset iteration conditions. If the evaluation accuracy rate in the iterative evaluation results is greater than the accuracy rate threshold, then the iterative evaluation results of the third application scenario are determined to meet the preset iteration conditions.

[0011] In one embodiment, before the step of obtaining the change rule data received by the change factor unit, the following steps are included: Iterative object data is generated in response to iterative requirements, wherein the iterative requirements include version iteration and business changes; Obtain the changed data and initial data from the iterative object data, and use the changed data and initial data as change rule data.

[0012] In one embodiment, prior to the step of acquiring the change rule data received by the change factor unit, the collaborative innovation agent iteration method further includes: The initial demand data is obtained and an initial production environment for the collaborative intelligent agent is created based on the initial demand data. The initial production environment includes a formal production environment for production and an iterative production environment for iteration. The iterative production environment and the formal production environment are isolated from each other. The iterative production environment is used to acquire the change rule data received by the change factor unit, and to iterate the formal production environment based on the change rule data when the iterative evaluation result meets the preset iterative conditions.

[0013] Furthermore, to achieve the above objectives, this application also provides an iterative device for business-technology co-creation intelligent agents. This iterative device is applied to a business-technology co-creation intelligent agent, which includes a change factor unit for debugging by business personnel and an iteration controller. The iteration controller is connected to the change factor unit and includes: The environment construction module is used to acquire the change rule data received by the change factor unit and construct the environment to be debugged based on the change rule data. The evaluation and processing module is used to acquire the full historical data of the intelligent agent created by the industry-technology co-creation, and to determine the iterative evaluation results based on the full historical data in the environment to be debugged. The iterative update module is used to iterate the intelligent agent of industry-technology co-creation according to the change rule data when the iterative evaluation result meets the preset iterative conditions.

[0014] Furthermore, to achieve the above objectives, this application also provides an intelligent agent iteration device for industry-technology co-creation, the intelligent agent iteration device for industry-technology co-creation comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the intelligent agent iteration method for industry-technology co-creation described above.

[0015] In addition, to achieve the above objectives, this application also provides a readable storage medium, which is a computer-readable storage medium, storing an intelligent agent iteration program for collaborative innovation, wherein when the intelligent agent iteration program for collaborative innovation is executed by a processor, it implements the steps of the intelligent agent iteration method for collaborative innovation as described above.

[0016] This application provides an iterative method for a business-technology co-created intelligent agent. The method is applied to an intelligent agent co-created by business personnel, which includes a change factor unit for debugging. The method acquires change rule data received by the change factor unit and constructs a debugging environment based on this data. It also acquires all historical data of the intelligent agent and determines the iterative evaluation result within the debugging environment based on this historical data. When the iterative evaluation result meets preset iteration conditions, the intelligent agent is iterated based on the change rule data. This iterative method utilizes change rule data received by the change factor unit (i.e., data from the change factor unit debugged by business personnel, which can provide business input) for iteration. It primarily constructs a debugging environment based on the change rule data, then uses all historical data to evaluate the environment and determine the iterative evaluation result. Finally, when the iterative evaluation result meets preset iteration conditions, the intelligent agent is iterated based on the change rule data, thus completing the iteration without developer intervention. This business-technology co-creation intelligent agent iteration method designs an intelligent agent co-created by business and technology, and then designs change factor units for business personnel to debug. This allows business personnel to input change rule data based on the change factor units, verify the change rule data, and iterate the business-technology co-created intelligent agent based on the change rule data after verification. The entire process does not require the participation of developers, thus providing a new way of intelligent agent iteration that achieves intelligent agent iteration without the need for developer intervention. Attached Figure Description

[0017] Figure 1 This is a schematic flowchart of the first embodiment of the intelligent agent iteration method co-created by industry and technology in this application; Figure 2 A schematic diagram of the overall architecture of the intelligent agent co-created by the industry and technology in this application; Figure 3 A schematic diagram of the iterative update logic of the intelligent agent iterative method co-created by industry and technology in this application; Figure 4 This is a schematic diagram of the module for identifying the controller in this application; Figure 5 This is a schematic diagram of the intelligent agent iteration device co-created by industry and technology in this application.

[0018] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings.

[0019] Explanation of icon numbers: 1001. Processing device; 1002. Read-only memory; 1003. Storage device; 1004. Random access memory; 1005. Bus; 1006. Input / output interface; 1007. Input device; 1008. Output device; 1009. Communication device. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0021] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0022] A common approach to iterating on intelligent agents involves developers building the agent based on iteration guidelines provided by business personnel, and then deploying the completed agent in a real-world environment. This approach often results in developers having limited understanding of the actual business logic, with the development process relying more on theoretical results and requirements from the iteration guidelines. Therefore, it demands high accuracy from the iteration guidelines provided by business personnel and requires developers to possess certain relevant knowledge, ultimately impacting the effectiveness and efficiency of the entire iterative update process.

[0023] To address the shortcomings of the aforementioned collaborative business-technology agent iteration methods, this application proposes a collaborative business-technology agent iteration method. The solution of this application's embodiments is as follows: iteration is performed using change rule data received by the change factor unit (i.e., change factor units from which business personnel perform debugging, allowing for business input). This primarily involves constructing a debugging environment based on the change rule data, then using full historical data to evaluate the environment and determine the iteration evaluation results. Finally, when the iteration evaluation results meet preset iteration conditions, the collaborative business-technology agent is iterated based on the change rule data, thus enabling iteration without developer intervention. This collaborative business-technology agent iteration method designs a collaborative business-technology agent and then designs change factor units for business personnel to debug, allowing business personnel to input change rule data based on the change factor units, verify the change rule data, and iterate the collaborative business-technology agent based on the change rule data after verification. The entire process can be completed without developer intervention, providing a new agent iteration method that achieves agent iteration without developer involvement.

[0024] It should be noted that the executing entity in this embodiment can be a computing server with data processing, network communication, and program execution functions, such as an iterative controller for intelligent agent iteration in industry-technology co-creation. The following description uses an iterative controller (such as a common controller, cloud processor, etc.) as an example to illustrate this embodiment and the subsequent embodiments.

[0025] Based on this, embodiments of this application provide an iterative method for intelligent agents through industry-technology collaboration, referring to... Figure 1 , Figure 1 This is a schematic flowchart of the first embodiment of the intelligent agent iteration method co-created by the industry and technology of this application.

[0026] Reference Figure 1 This application provides an iterative method for business-technology co-creation of intelligent agents. This method is applied to a business-technology co-created intelligent agent, which includes change factor units debugged by business personnel. The iterative method for business-technology co-created intelligent agents includes: Step S10: Obtain the change rule data received by the change factor unit, and construct the debugging environment based on the change rule data; In this embodiment, the business-technology co-creation agent iteration method is applied to the business-technology co-created agent. A change factor unit is designed within the business-technology co-created agent for business personnel to debug. This allows business personnel to debug, evaluate, and release the entire agent, thus completing the iteration of the agent without the need for developer intervention. When an iterative update is needed, the change rule data received by the business personnel in the change factor unit is retrieved. That is, when the business carried by the agent changes, business personnel can verify their new ideas in real time within the change factor unit. For example, when a financial product is launched, if business personnel want to extract more product attribute information from the product description through the agent, they can perform online debugging within the corresponding business-technology co-created agent, completing preliminary feasibility verification without relying on developers. Among them, change rule data refers to the change rules of the business, such as changing the extraction object of the extraction instruction to A, or changing the initial display object from B to C. Of course, it can also be other rule changes, which will not be explained one by one here. That is, the mapping rules, extraction rules, initial display rules, and the correspondence rules between parent and child pages in the intelligent body, etc. Then, the debugging environment will be constructed based on the change rule data. That is, the change rule data is input into the intelligent body of business and technology co-creation. At this time, the implementation logic inside the intelligent body of business and technology co-creation is related to the change rule data, and thus the environment inside the entire intelligent body of business and technology co-creation is determined to be the debugging environment. For example, the intelligent agent in the industry-technology co-creation is a simple environment where a D command controls the display of content C. However, the change rule data adds the call and display relationship of the D command to content B and C. Then, the internal logic of the environment to be debugged is that the call and display relationship of the D command is increased to content B and C. Thus, when the user inputs the D command, if the displayed content is determined to be content B and C, then it is determined that this is the environment to be debugged. Conversely, if it is determined that when the user inputs the D command, only content B is displayed, then it is determined that the change rule data cannot meet the user's needs, and the environment to be debugged cannot be generated. That is, the user needs to adjust the change rule data again to meet the conditions for generating the environment to be debugged. This can ensure the accuracy of the change rule data input by the user, and thus ensure the accuracy of the entire intelligent agent iteration.

[0027] Step S20: Obtain the full historical data of the intelligent agent co-created by industry and technology, and determine the iterative evaluation results in the debugging environment based on the full historical data; Step S30: When the iterative evaluation results meet the preset iterative conditions, the intelligent agent of industry-technology co-creation is iterated according to the change rule data.

[0028] In this embodiment, once business personnel determine the debugging environment for the collaboratively created intelligent agent, they can further assess its performance in production operations. For example, with a financial risk rating intelligent agent (a collaboratively created agent), after business personnel adjust the business rules for assessing product risk ratings (change rule data) and update the intelligent agent to obtain the debugging environment, they will create an evaluation task in the agent's platform. This task will then run batch tests on historical products, using the updated intelligent agent to assess the risk ratings of those products, and ultimately provide the success rate and accuracy of the evaluation. This allows business personnel to understand the effectiveness of the intelligent agent in production operations in advance. This means that by acquiring the full historical data of the intelligent agent in the industry-technology co-creation, the full historical data refers to all the operational data in the actual production environment. Taking the debugging environment where the displayed content is determined to be B or C when the D command is input as an example, the full historical data can be different data when the D command is input in scenario F, when the D command is input in product G, and when the D command is input in the intelligent agent state J. Then, the iterative evaluation results under different data can be determined based on the verification results under different data, that is, the evaluation results of the change rule data, which can be the ratio of accuracy, success rate and precision, or other indicators, which are not limited here.

[0029] Furthermore, when the iterative evaluation results meet the preset iterative conditions, it is determined that the change rule data meets the needs of actual application. Therefore, the collaboratively created intelligent agent will be iterated directly based on the change rule data. This means that the collaboratively created intelligent agent in actual production will be iteratively updated using the change rule data to obtain a new intelligent agent. For example, after release review, the parameters of the new version of the intelligent agent will be automatically synchronized from the verification environment to the production environment, allowing the intelligent agent to take effect in the production environment. This allows business personnel to implement iterative updates of the intelligent agent, providing a new iterative update approach. For example, refer to... Figure 2 , Figure 2This is a schematic diagram of the overall architecture of the intelligent agent created through industry-technology co-creation in this application. The application layer of the intelligent agent's middle platform designs the intelligent agent debugging, creates a production environment to be debugged, and then generates the iterative evaluation results for the environment to be debugged during the intelligent agent evaluation. Finally, when the results meet the requirements, the updated rules are applied to the actual production intelligent agents through intelligent agent distribution. For example, the intelligent agent verification environment includes task processing, models, and databases. The corresponding change rule data can be the changed task processing rules, model algorithms, and data in the database. Then, the environment is verified based on the changed rule data, and after successful verification, it is applied to the formal production environment. The entire iterative update process is based on the core concept of "industry-technology co-creation," decoupling business and technical development. Technical developers can focus more on the implementation and integration of intelligent agent technology, while business personnel can also independently debug, evaluate, and release versions of the intelligent agent. This allows both business personnel and developers to focus on their respective areas of expertise to improve the efficiency of value transformation of requirements. It also provides an intelligent agent evaluation function. Business personnel can create batch evaluation tasks for intelligent agents in the intelligent agent platform. The system will automatically calculate the accuracy and success rate of the evaluation tasks. In this way, through evaluation, business personnel can understand the expected effect of the intelligent agent in advance and reduce the probability of secondary iteration optimization.

[0030] Furthermore, the intelligent agent for industry-technology co-creation also includes fixed logic units for developers to debug, and the steps of building the environment to be debugged based on change rule data, including: Step S11: Obtain the solidified logic data of the solidified logic unit, and determine the initial debugging environment based on the change rule data and the solidified logic data, and determine the debugging verification result of the preset debugging data in the initial debugging environment; Step S12: When the debugging verification result is that the debugging verification is successful, the initial debugging environment is used as the environment to be debugged.

[0031] In this embodiment, before iteratively updating the intelligent agent co-created by business and technology, the "fixed logic" (such as prompt word template format, assembling prompt words by grabbing product information, compression of large images, model interaction, formatting and verification of returned results, etc.) of the intelligent agent is separated from the "variable factors" (such as prompt word content, document selection, model selection, etc.). The "fixed logic" refers to the relevant data and logical relationships within the fixed logic unit, which is the fixed logic data. The "variable factors" refer to the relevant data and logical relationships within the variable factor unit, which can input change rule data. This provides business personnel with an optimization interface that can adjust the "variable factors". Business personnel can complete the intelligent agent effect debugging by adjusting the variable factor content, and can complete the iterative update of the intelligent agent. After the user inputs change rule data, the initial debugging environment is determined based on the change rule data and the fixed logic data of the fixed logic unit. For example, in a multi-layer mapping relationship, the fixed logic data is that instruction 1 opens page 2 and extracts the E1 parameter from page 2, and the change factor is defined as extracting parameters from page 2. Then, the change rule data can be instruction 1 corresponding to extracting E1 and E2 parameters. The initial debugging environment is then instruction 1 opening page 2 and extracting E1 and E2 parameters from page 2. After determining the initial debugging environment, the debugging verification result of the preset debugging data in the initial debugging environment is determined. If the debugging verification result is successful, the initial debugging environment is used as the environment to be debugged. That is, the initial debugging environment is subjected to the most routine initial verification. For example, if the preset debugging data is instruction 1, it is determined whether instruction 1 can extract the E1 and E2 parameters from page 2. If it can be extracted, it is the environment to be debugged. Otherwise, it is determined that the change rule data design is unreasonable or cannot achieve the goal, and no further processing is performed, or the user is prompted that the change rule data is unreasonable or cannot achieve the goal. Of course, the preset debugging data can also be any data that can be selected to verify whether the changed rule data is reasonable or whether it has achieved the target, depending on the actual situation. For example, if the instruction correspondence is modified, then it is the instruction; if the numerical mapping relationship is modified, then it is the numerical value. There is no limitation here. That is, the whole process supports business personnel to debug the intelligent agent online based on their needs, and can complete the initial feasibility verification without relying on developers.

[0032] For example, refer to Figure 3 , Figure 3This diagram illustrates the iterative update logic of the collaborative innovation method for intelligent agents in this application. Each intelligent agent consists of "fixed logic" and "variable factors," primarily placeholders. Modifying these placeholders alters the final rules, affecting prompts and determining different large models. Ultimately, this changes information extraction, enabling iterative updates of the intelligent agent. For example, during the entire intelligent agent integration process, developers encapsulate the fixed logic (such as prompt template formatting, assembling prompts from product information, compressing large images, model interaction, formatting and validating returned results, etc.) and provide business personnel with an interface for adjusting variable factors (such as prompt content, document selection, model selection, etc.). Finally, business personnel debug the intelligent agent by adjusting prompt content, changing documents, and switching models, ensuring no interference and improving iteration efficiency through a collaborative innovation model.

[0033] This embodiment provides an iterative method for a business-technology co-created intelligent agent. The method applies to a business-technology co-created intelligent agent, which includes a change factor unit for debugging by business personnel. The method acquires change rule data received by the change factor unit and constructs a debugging environment based on this data. It also acquires all historical data of the business-technology co-created intelligent agent and determines the iterative evaluation result within the debugging environment based on this historical data. When the iterative evaluation result meets preset iteration conditions, the business-technology co-created intelligent agent is iterated based on the change rule data. This iterative method uses change rule data received by the change factor unit (i.e., from the change factor unit debugged by business personnel, which can provide business input) for iteration. It primarily constructs a debugging environment based on the change rule data, then uses all historical data to evaluate within this environment to determine the iterative evaluation result. Finally, when the iterative evaluation result meets preset iteration conditions, the business-technology co-created intelligent agent is iterated based on the change rule data. This allows for iteration without the need for developer intervention. This business-technology co-creation agent iteration method designs a business-technology co-created agent and then designs change factor units for business personnel to debug. This allows business personnel to input change rule data based on the change factor units, verify the change rule data, and then iterate the business-technology co-created agent based on the change rule data after verification. In other words, the entire process can be completed without the involvement of developers, thus providing a new way to iterate agents without the need for developers.

[0034] Furthermore, based on the first embodiment of this application described above, a second embodiment of the intelligent agent iteration method of this application is proposed. In this embodiment, the step of determining the iteration evaluation result based on the full historical data in the debugging environment includes: Step S21: For each evaluation data in the full historical data, determine the output result data in the environment to be debugged, and determine the target application scenario corresponding to the environment to be debugged. Step S22: Determine the iterative evaluation results based on the output data and the target application scenario.

[0035] In this embodiment, after determining the environment to be debugged, all historical data is used to evaluate it. Specifically, each evaluation data point from the entire historical dataset is used to determine the output data within the environment. Each evaluation data point can be data from different scenarios, different products, or different agent states. This allows for evaluation tailored to different needs in actual operating environments, ensuring accuracy. An initial test is also conducted based on the initial debugging environment to avoid resource waste. Furthermore, the target application scenario corresponding to the environment to be debugged is determined, specifying the application scenario. Based on the output data and the target application scenario, iterative evaluation results are then determined to complete the normal evaluation process.

[0036] Furthermore, the steps for determining the iterative evaluation results based on the output data and the target application scenario include: Step S221: When the target application scenario is the first application scenario based on accuracy, determine the evaluation accuracy between the output result data and the preset standard result, and use the evaluation accuracy as the iterative evaluation result. Step S222: When the target application scenario is the second application scenario based on the success rate, determine the evaluation success rate between the output result data and the preset success result, and use the evaluation success rate as the iterative evaluation result. Step S223: When the target application scenario is the third application scenario based on the accuracy rate, determine the evaluation accuracy rate between the output result data and the preset correct result, and use the evaluation accuracy rate as the iterative evaluation result.

[0037] In this embodiment, different criteria can be selected for the output result data in different scenarios. This embodiment uses accuracy, success rate, and correctness as examples for explanation. The first application scenario is where accurate product attributes need to be obtained. Accuracy is considered when the information extracted from historical product documents (standard results defined by the user from historical data) matches the current product library field information (output result data). For example, taking the extraction of financial product description information as an example, if the product attributes extracted from historical product descriptions match the actual product attributes in the current product library, it indicates that the agent information extraction is accurate. When a certain evaluation data is input, the preset standard result of that evaluation data is obtained from the full historical data, and then compared with the output result data. Finally, based on multiple evaluation data, the overall accuracy can be determined, thus completing the evaluation. The second application scenario is where only data acquisition is required; successful data output is considered successful. For example, taking the "financial risk rating agent" evaluation as an example, if the risk rating information of historical products is successfully output according to business rules, it indicates success. Success can also be associated with accuracy; after obtaining the required data, the iterative evaluation result is determined to be passed. The third application scenario involves reviewing information disclosure documents to determine the accuracy of identified issues. If accurate, the review is considered correct. This means not only acquiring the necessary data but also assessing its accuracy. For example, when a specific evaluation data point is input, a preset correct result is retrieved from the entire historical dataset and compared with the output data. The overall accuracy rate is then determined based on multiple evaluation data points, completing the evaluation. The difference between accuracy and correctness is that accuracy only retrieves accurate data, such as daily revenue for a given year; correctness also considers relevant information to determine the accuracy of the daily revenue. Furthermore, batch evaluations are conducted using the parsed production documents. Algorithms calculate the accuracy, success rate, and correctness of the intelligent agent, helping business personnel understand its performance in the production environment and ensuring it achieves the expected results before official release. It is worth noting that once the evaluation is confirmed to be passed, the verified agent will be released to the production environment. At the same time, the agent platform verification environment (the agent environment for debugging and evaluation in this application) will be isolated from the production environment to avoid the evaluation process from affecting production functions.

[0038] It's worth noting that multiple evaluation criteria can be used to make judgments within a specific scenario, allowing for targeted updates to the agent in different situations. In this case, the full dataset of evaluations ensures that the updated agent is suitable for different production environments, while also guaranteeing its effectiveness in production.

[0039] Furthermore, the preset iteration conditions include accuracy threshold, success rate threshold, and correctness threshold. After determining the iteration evaluation results based on the full historical data in the debugging environment, the process includes: Step S23: If the evaluation accuracy in the iterative evaluation results is greater than the accuracy threshold, then it is determined that the iterative evaluation results of the first application scenario meet the preset iteration conditions. Step S24: If the success rate of the iterative evaluation results is greater than the success rate threshold, then it is determined that the iterative evaluation results of the second application scenario meet the preset iterative conditions. Step S25: If the evaluation accuracy rate in the iterative evaluation results is greater than the accuracy rate threshold, then it is determined that the iterative evaluation results of the third application scenario meet the preset iterative conditions.

[0040] In this embodiment, after determining the iterative evaluation results, a comprehensive judgment is made based on different iterative evaluation results and scenarios to determine whether iterative updates can be controlled. For example, if the evaluation accuracy rate in the iterative evaluation results is greater than the accuracy threshold, then the iterative evaluation results of the first application scenario are determined to meet the preset iteration conditions; if the evaluation success rate is greater than the success rate threshold, then the iterative evaluation results of the second application scenario are determined to meet the preset iteration conditions; if the evaluation accuracy rate is greater than the accuracy rate threshold, then the iterative evaluation results of the third application scenario are determined to meet the preset iteration conditions. Of course, multiple data can also be evaluated simultaneously. For example, for the fourth application scenario that requires both accuracy and success rate, it is necessary to separately determine whether the evaluation accuracy rate and evaluation success rate meet the requirements. If both meet the requirements, then the iterative evaluation results of the fourth application scenario meet the preset iteration conditions. Furthermore, when the user generates change rule data, the corresponding evaluation basis can be generated to process different demand scenarios in a targeted manner.

[0041] Furthermore, based on the first and / or second embodiments of this application described above, a third embodiment of the intelligent agent iteration method of this application is proposed. In this embodiment, before the step of obtaining the change rule data received by the change factor unit, the method includes: Step a: Generate iterative object data in response to iteration requirements, where iteration requirements include version iteration and business changes; Step b: Obtain the changed data and initial data from the iterable object data, and use the changed data and initial data as change rule data.

[0042] In this embodiment, when iteration is required, iteration object data is generated based on iteration requirements. These requirements may include version iterations and business changes. When these states or situations are detected, iteration object data is generated. This data can be user-inputted data or automatically generated data based on iteration requirements. For example, if a change is detected in business A, a new mapping relationship or instruction retrieval relationship can be directly generated based on the changed business A, thus obtaining iteration object data. Finally, the changed data and initial data in the iteration object data are determined and used as change rule data. Changed data refers to data where the rules change, and initial data refers to rule data directly or indirectly related to the changed data. Data with no relation to the changed data is excluded; changed data and data directly or indirectly related to the changed data are not included in the change rule data to facilitate subsequent rule modifications. It is worth noting that the reason for choosing the initial data here is that although this type of data has not been changed, the changed data affects this type of data. Therefore, the iterative evaluation results of the initial data output will also be evaluated in subsequent judgments. That is, the whole process will not only focus on whether the result of the rule change meets the requirements, but also judge whether the results directly or indirectly related after the rule change meet the requirements, thereby ensuring the subsequent application effect of the intelligent agent.

[0043] In another embodiment, prior to the step of acquiring the change rule data received by the change factor unit, the collaborative intelligent agent iteration method further includes: Step c: Obtain the input initial demand data and create the initial production environment of the intelligent agent for industry-technology co-creation based on the initial demand data. The initial production environment includes a formal production environment for production and an iterative production environment for iteration. The iterative production environment and the formal production environment are isolated from each other. The iterative production environment is used to acquire the change rule data received by the change factor unit, and to iterate the formal production environment based on the change rule data when the iterative evaluation results meet the preset iterative conditions.

[0044] In this embodiment, before executing the entire iterative update step, the collaboratively created intelligent agent is developed. This involves acquiring initial requirement data, i.e., the basis data for initial development, and then creating an initial production environment for the collaboratively created intelligent agent based on this initial requirement data. This initial production environment includes a formal production environment for production and an iterative production environment for iteration, which are isolated from each other. Finally, the iterative production environment acquires the change rule data received by the change factor unit and iterates the formal production environment based on the change rule data when the iterative evaluation results meet preset iteration conditions. It is worth noting that during the iterative update process, business personnel can determine whether the change rule data meets their needs. If not, new initial requirement data needs to be generated to build a new intelligent agent environment. Another scenario is that after generating change rule data multiple times, either the environment to be debugged cannot be generated, or the iterative evaluation results do not meet the preset iteration conditions. In such cases, business personnel can determine that new initial requirement data needs to be generated to build a new intelligent agent environment, thus greatly improving their judgment on the basis for building a new intelligent agent environment.

[0045] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the intelligent agent iteration method of this application. Any simple transformations based on this technical concept are within the protection scope of this application.

[0046] This application also provides an iterative device for collaborative business and technology creation of intelligent agents. This device is applied to an intelligent agent involved in collaborative business and technology creation. The intelligent agent includes a change factor unit for debugging by business personnel and an iterative controller. The iterative controller is connected to the change factor unit, as shown in the reference... Figure 4 The iterative controller includes: The environment construction module A10 is used to acquire the change rule data received by the change factor unit and construct the environment to be debugged based on the change rule data. The evaluation and processing module A20 is used to acquire the full historical data of the intelligent agent created by industry and technology collaboration, and to determine the iterative evaluation results in the debugging environment based on the full historical data. The iterative update module A30 is used to iterate the intelligent agent of industry-technology co-creation based on the change rule data when the iterative evaluation results meet the preset iterative conditions.

[0047] The collaborative intelligent agent iteration device provided in this application adopts the collaborative intelligent agent iteration method in the above embodiments, aiming to provide a new intelligent agent iteration method that achieves intelligent agent iteration without the participation of developers. Compared with the prior art, the beneficial effects of the collaborative intelligent agent iteration device provided in this application are the same as those of the collaborative intelligent agent iteration method provided in the above embodiments, and other technical features in the controller are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0048] This application provides an intelligent agent iteration device for collaborative innovation, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute the intelligent agent iteration method for collaborative innovation described in Embodiment 1 above.

[0049] The following is for reference. Figure 5 The diagram illustrates a structural schematic of an intelligent agent iteration device suitable for implementing the industry-technology co-creation embodiments of this application. The intelligent agent iteration device for industry-technology co-creation in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The intelligent agent iteration device for collaborative innovation shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0050] like Figure 5As shown, the collaborative intelligent agent iteration device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the collaborative intelligent agent iteration device. The processing unit 1001, the ROM 1002, and the RAM 1004 are interconnected via a bus 1005. An input / output interface 1006 is also connected to the bus. Typically, the following devices may be connected to the input / output interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows collaboratively developed intelligent agent iteration devices to communicate wirelessly or wiredly with other collaboratively developed intelligent agent iteration devices to exchange data. Although the figures show collaboratively developed intelligent agent iteration devices with various devices, it should be understood that implementing or possessing all of the shown devices is not required. More or fewer devices may be implemented alternatively.

[0051] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or loaded from storage device 1003, or loaded from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0052] The collaborative intelligent agent iteration device provided in this application employs the collaborative intelligent agent iteration method described in the above embodiments, aiming to provide a new intelligent agent iteration approach that achieves intelligent agent iteration without the involvement of developers. Compared with the prior art, the beneficial effects of the collaborative intelligent agent iteration device provided in this application are the same as those of the collaborative intelligent agent iteration method provided in the above embodiments, and other technical features in this collaborative intelligent agent iteration device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0053] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0054] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0055] This application provides a readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the collaborative intelligent agent iteration method in the above embodiments.

[0056] The readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor devices, apparatuses, or components, or any combination thereof. More specific examples of readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or flash memory, optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution apparatus, device, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), or any suitable combination thereof.

[0057] The aforementioned readable storage medium may be included in the collaborative intelligent agent iteration device; or it may exist independently and not be assembled into the collaborative intelligent agent iteration device.

[0058] The aforementioned readable storage medium carries one or more programs, which, when executed by the collaborative intelligent agent iteration device, cause the collaborative intelligent agent iteration device to execute: Acquire the change rule data received by the change factor unit, and construct the debugging environment based on the change rule data; Acquire the full historical data of the intelligent agent created through industry-technology collaboration, and determine the iterative evaluation results based on the full historical data in the environment to be debugged; When the iterative evaluation results meet the preset iterative conditions, the intelligent agent of industry-technology co-creation is iterated based on the change rule data.

[0059] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0060] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation that may be implemented by apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using dedicated hardware-based apparatus to perform the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0061] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0062] The readable storage medium provided in this application stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned collaborative intelligent agent iteration method. It aims to provide a novel intelligent agent iteration method that achieves intelligent agent iteration without the involvement of developers. Compared to the prior art, the beneficial effects of the readable storage medium provided in this application are the same as those of the collaborative intelligent agent iteration method provided in the above embodiments, and will not be elaborated upon here.

[0063] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described collaborative intelligent agent iterative method.

[0064] The computer program product provided in this application aims to solve the technical problem of poor iteration effect of intelligent agents in industry-technology co-creation. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the intelligent agent iteration method in industry-technology co-creation provided in the above embodiments, and will not be repeated here.

[0065] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the content of this application specification and drawings, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. An iterative method for intelligent agents through industry-technology co-creation, characterized in that, The aforementioned business-technology co-creation intelligent agent iteration method is applied to the business-technology co-creation intelligent agent, which includes change factor units debugged by business personnel. The business-technology co-creation intelligent agent iteration method includes: The change rule data received by the change factor unit is obtained, and the debugging environment is constructed based on the change rule data; Obtain the full historical data of the intelligent agent created through industry-technology collaboration, and determine the iterative evaluation results based on the full historical data in the environment to be debugged; When the iterative evaluation results meet the preset iterative conditions, the intelligent agent of the industry-technology co-creation is iterated according to the change rule data.

2. The intelligent agent iterative method for industry-technology co-creation as described in claim 1, characterized in that, The intelligent agent for industry-technology co-creation also includes a fixed logic unit for developers to debug, and the step of constructing the debugging environment based on the change rule data includes: Acquire the solidified logic data of the solidified logic unit, and determine the initial debugging environment based on the change rule data and the solidified logic data, and determine the debugging verification result of the preset debugging data in the initial debugging environment; When the debugging verification result is that the debugging verification is successful, the initial debugging environment is used as the environment to be debugged.

3. The intelligent agent iterative method for industry-technology co-creation as described in claim 1, characterized in that, The step of determining the iterative evaluation result based on the full historical data in the environment to be debugged includes: For each evaluation data point in the full historical data, the output result data is determined in the environment to be debugged, and the target application scenario corresponding to the environment to be debugged is determined. The iterative evaluation results are determined based on the output data and the target application scenario.

4. The intelligent agent iterative method for industry-technology co-creation as described in claim 3, characterized in that, The step of determining the iterative evaluation result based on the output result data and the target application scenario includes: When the target application scenario is a first application scenario based on accuracy, the evaluation accuracy between the output result data and the preset standard result is determined, and the evaluation accuracy is used as the iterative evaluation result; When the target application scenario is a second application scenario based on the success rate, the evaluation success rate between the output result data and the preset success result is determined, and the evaluation success rate is used as the iterative evaluation result; When the target application scenario is a third application scenario based on the accuracy rate, the evaluation accuracy rate between the output result data and the preset correct result is determined, and the evaluation accuracy rate is used as the iterative evaluation result.

5. The intelligent agent iterative method for industry-technology co-creation as described in claim 4, characterized in that, The preset iteration conditions include an accuracy threshold, a success rate threshold, and a correctness threshold. Following the step of determining the iteration evaluation results based on the full historical data in the debugging environment, the following steps are included: If the evaluation accuracy in the iterative evaluation results is greater than the accuracy threshold, then the iterative evaluation results of the first application scenario are determined to meet the preset iteration conditions. If the success rate in the iterative evaluation results is greater than the success rate threshold, then the iterative evaluation results of the second application scenario are determined to meet the preset iteration conditions. If the evaluation accuracy rate in the iterative evaluation results is greater than the accuracy rate threshold, then the iterative evaluation results of the third application scenario are determined to meet the preset iteration conditions.

6. The intelligent agent iterative method for industry-technology co-creation as described in any one of claims 1 to 5, characterized in that, Before the step of obtaining the change rule data received by the change factor unit, the following steps are included: Iterative object data is generated in response to iterative requirements, wherein the iterative requirements include version iteration and business changes; Obtain the changed data and initial data from the iterative object data, and use the changed data and initial data as change rule data.

7. The intelligent agent iterative method for industry-technology co-creation as described in any one of claims 1 to 5, characterized in that, Before the step of acquiring the change rule data received by the change factor unit, the intelligent agent iteration method of industry-technology co-creation further includes: The initial demand data is obtained and an initial production environment for the collaborative intelligent agent is created based on the initial demand data. The initial production environment includes a formal production environment for production and an iterative production environment for iteration. The iterative production environment and the formal production environment are isolated from each other. The iterative production environment is used to acquire the change rule data received by the change factor unit, and to iterate the formal production environment based on the change rule data when the iterative evaluation result meets the preset iterative conditions.

8. A collaborative innovation intelligent agent iteration device, characterized in that, The aforementioned intelligent agent iteration device for business-technology co-creation is applied to an intelligent agent for business-technology co-creation. The intelligent agent includes a change factor unit for business personnel to debug and an iteration controller. The iteration controller is connected to the change factor unit and includes: The environment construction module is used to acquire the change rule data received by the change factor unit and construct the environment to be debugged based on the change rule data. The evaluation and processing module is used to acquire the full historical data of the intelligent agent created by the industry-technology co-creation, and to determine the iterative evaluation results based on the full historical data in the environment to be debugged. The iterative update module is used to iterate the intelligent agent of industry-technology co-creation according to the change rule data when the iterative evaluation result meets the preset iterative conditions.

9. A collaborative intelligent agent iteration device, characterized in that, The collaborative innovation intelligent agent iteration device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the collaborative innovation intelligent agent iteration method as described in any one of claims 1 to 7.

10. A readable storage medium, characterized in that, The readable storage medium is a computer-readable storage medium, and the computer-readable storage medium stores an intelligent agent iteration program for collaborative innovation, wherein when the intelligent agent iteration program for collaborative innovation is executed by a processor, it implements the steps of the intelligent agent iteration method for collaborative innovation as described in any one of claims 1 to 7.

Images