Online intelligent agent automatic deployment management system

Abstract

The invention belongs to the technical field of automatic operation and maintenance, and particularly relates to an online agent automatic deployment management system. The foreground control system is used for providing a visual control panel for an administrator to set and adjust deployment, configuration information and behavior modes of the intelligent agent; the background monitoring system is used for carefully supervising the operation process of the intelligent agent and providing real-time and accurate information feedback for an administrator; the deployment operation system is used for converting an agent behavior strategy predetermined for the online platform into a specific behavior; the strategy algorithm system is used for providing an efficient and convenient automatic scheme for the operation of the intelligent agent. The invention is used for automatically deploying, managing and monitoring agents on a network online platform, and the system consists of a plurality of modules which work cooperatively to support various functions.

Description

Technical Field

[0001] The present invention belongs to the technical field of automated operation and maintenance, and specifically relates to an online intelligent body automated deployment management system. Background Art

[0002] In today's digital age, various online platforms have become key places for business process automation, human-computer interaction, and information dissemination. In addition to the direct participation and interaction of human users, the number of automated tasks performed by agents on these platforms is increasing, including automated operations such as clicking, forwarding, and publishing content. Although these automated functions help improve efficiency, their deployment and management face challenges in complexity and flexibility.

[0003] Currently, most existing management tools focus on providing services to human users, such as content publishing plans, interaction monitoring, and effect analysis, but lack effective solutions for the deployment, maintenance, and dynamic management of intelligent agents. Although some tools support automated operations, they are often limited in terms of fine management, real-time monitoring, and intelligent automated operations. In particular, in terms of operation, existing methods mostly make decisions through preset templates or programs to complete the automated operation process of intelligent agents, lacking more flexible and intelligent modeling solutions. Summary of the invention

[0004] The present invention provides an online intelligent agent automatic deployment management system, which is used for automatically deploying, managing and monitoring intelligent agents on a network online platform. The system is composed of multiple modules working in coordination to support various functions.

[0005] The present invention is achieved through the following technical solutions:

[0006] An online intelligent agent automatic deployment management system, the system includes a front-end control system, a back-end monitoring system, a deployment and operation system, and a strategy algorithm system.

[0007] The front-end control system is used to provide an intuitive control panel for administrators to set and adjust the deployment, configuration information and behavior mode of the intelligent agent;

[0008] The background monitoring system is used to monitor the operation process of the intelligent agent in detail and provide real-time and accurate information feedback to the administrator;

[0009] The deployment and operation system is used to convert the agent behavior strategy predetermined for the online platform into specific behavior;

[0010] The strategy algorithm system is used to provide an efficient and convenient automation solution for the operation of intelligent entities.

[0011] Furthermore, the front-end control system includes an account management module, a portrait management module, an algorithm configuration module and an account grouping module;

[0012] The account management module is used to import the account information of the online platform where the agent is located to add the agent to the deployment and operation module;

[0013] The portrait management module is used to complete the configuration of agent information;

[0014] The algorithm configuration module and the account grouping module are aggregated and work together to manage the grouping of intelligent entities.

[0015] Furthermore, the account management module can freely complete the online / offline function of the agent by clicking the corresponding start / stop button;

[0016] The agent portrait in the portrait management module includes four parts: age, label, location, and introduction, which are described in natural language and depict the anthropomorphic background information of the agent;

[0017] In the account grouping module, the administrator customizes the agent account groups, and then uses the algorithm configuration module to configure different strategy algorithms for different groups of agent clusters; after the configuration is completed, the agents in different clusters can run simultaneously in the system with different algorithm logics, thereby exhibiting differentiated behavior patterns, so that the administrator can analyze the effectiveness of the strategy algorithm.

[0018] Furthermore, the background monitoring system includes action behavior monitoring, account status monitoring and group monitoring;

[0019] The action behavior monitoring module is used for the administrator to analyze the execution results of the decision algorithm;

[0020] The account status monitoring module is used to monitor the operating status of the agent account;

[0021] The group monitoring module is used to provide cluster monitoring of intelligent agent accounts for administrators.

[0022] Furthermore, the account status monitoring module provides data results of the agent being followed, forwarded, and interacted by human users;

[0023] The group monitoring module supports the system to monitor the behavior and status of each group, and presents it to the administrator in the form of the mean of the indicators within the group for algorithm adjustment and optimization.

[0024] Furthermore, the deployment and operation system includes an account deployment module, an algorithm execution module and an action execution module.

[0025] The account deployment module is used to allocate an independent thread space to each agent;

[0026] The algorithm execution module is used to provide the configured agent information to the algorithm module to generate a decision strategy;

[0027] The action execution module is used to guide the automated operation program to complete the online behavior execution operation of the intelligent agent according to the strategy.

[0028] Furthermore, after the administrator configures the agent information in the user terminal, the information is transmitted to the information storage module of the database; the same is true for the agent operation status information;

[0029] The decision strategy is generated through an algorithm to guide the automated operation program to complete the online behavior execution operation of the intelligent agent according to the strategy; while the intelligent agent is executing various behaviors, it transmits the operation execution results to the monitoring module background to complete the update of monitoring information;

[0030] The deployment and operation system deploys accounts in a multi-threaded manner, allocating an independent thread space to each agent; the portrait and algorithm configuration information of each agent are assigned to their own threads; Selenium and ChromeDriver are used inside the thread to operate the browser;

[0031] When processing the peak operation of a large number of intelligent agents, the deployment and operation system uses a thread pool mechanism to ensure that each behavior can be successfully executed in the system under limited computing resources.

[0032] Furthermore, the strategy algorithm system includes a behavior sequence modeling module and a content generation modeling module.

[0033] The sequence modeling module is used for policy-driven automated behavior execution;

[0034] The content generation modeling module is used for text generation.

[0035] Furthermore, the sequence modeling module includes agent behavior frequency and agent behavior sequence,

[0036] The agent behavior frequency modeling adopts the power law distribution function to model the agent's behavior interval Δt:

[0037] P(Δt)~Δt -α (1)

[0038] The approximate optimal solution of α in the above function (1) is obtained to be between [1min, 10h]; that is, each agent will perform approximately 10 to 30 actions per day;

[0039] The agent behavior sequence modeling adopts Markov chain to model the agent's behavior state:

[0040] P(X n+1 =x|X0,X1,X2,…,X n ) (2)

[0041] Among them, x is the behavior variable, representing the possible actions of the agent; X0, X1, X2, …, X n is the sequence of actions executed by the agent, which constitutes the current state of the agent; X n+1 is the action that the agent will perform;

[0042] By simulating the behavior of a large number of human users, we can obtain the execution probability of the next action in each state in the above models (1) and (2). The system will guide the agent to perform the corresponding action with the corresponding probability;

[0043] By combining the user's behavior frequency and behavior sequence modeling scheme, a complete automated behavior execution strategy is constructed; mechanical operations will be directly executed by deploying and running modules; and advanced operations involving text content need to be combined with automated text generation algorithms to complete.

[0044] Furthermore, the content generation modeling module first uses large-scale online platform text data to further train the large language model so that it can understand and simulate specific user group behaviors and language patterns;

[0045] Subsequently, using the preset topics and combining the agent account portraits and behavior sequences under specific grouping clusters, the large language model is used to generate posts and comments under the specified topics.

[0046] The beneficial effects of the present invention are:

[0047] By integrating the above system modules, the present invention can greatly reduce the reliance on human resources for operating and maintaining online agents. Overall, the system uses advanced strategy algorithms, combined with a stable account cluster operation scheme, to drive agents to perform corresponding behavioral operations, aiming to improve the user experience on the online platform and promote the healthy development and growth of the platform.

[0048] The system of the present invention can provide real-time feedback on the results of agent behavior through real-time monitoring of agent account activities and agent status, thereby optimizing the administrator user experience.

[0049] The present invention provides an economical, efficient and convenient online network platform intelligent body automation management tool, which greatly improves the overall efficiency and quality of operations.

[0050] The present invention aims to solve the above problems in a highly automated and intelligent way. In addition, the system also provides a set of comprehensive management tools, which enables managers to easily configure and monitor various activities of intelligent agents, thereby greatly improving the automation level of intelligent agent management and the transparency of operations.

[0051] By applying this system, the present invention can not only improve the response speed and processing capacity of online services, but also open up new possibilities for online platform intelligent management and automated service fields. In addition, the implementation of this system also plays an important role in improving user service experience and ensuring the healthy development of the online platform ecosystem. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a structural schematic diagram of the present invention.

[0053] Figure 2 It is a flow chart of the operation of the deployment and operation module of the present invention in the system.

[0054] Figure 3 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0055] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0056] It should be understood that when used in this specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0057] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, the singular forms "a", "an" and "the" are intended to include plural forms unless the context clearly indicates otherwise.

[0058] The following is attached to this application specification Figure 1-3, the technical solutions in the embodiments of the present application are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0059] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0060] Implementation Method 1

[0061] This embodiment provides an online intelligent agent automatic deployment management system, which includes a front-end control system, a back-end monitoring system, a deployment and operation system, and a strategy algorithm system.

[0062] The front-end control system is used as the interface for administrators to interact with the system, providing an intuitive control panel for administrators to set and adjust the deployment, configuration information and behavior patterns of the intelligent body; this part is presented in the form of a web interface and exchanges data with the deployment and operation module through the Http protocol.

[0063] The background monitoring system is used for detailed supervision of the operation process of the intelligent body and provides real-time and accurate information feedback to the administrator; the front end of this part is also presented in the form of a web interface, and the back end is aggregated with the deployment and operation module.

[0064] The deployment and operation system is used to convert the agent behavior strategy predetermined for the online platform into specific behaviors; the operation process of the deployment and operation module in the system is as follows: Figure 2 shown.

[0065] The strategy algorithm system is used to provide an efficient and convenient automation solution for the operation of intelligent entities.

[0066] Furthermore, the front-end control system includes an account management module, a portrait management module, an algorithm configuration module and an account grouping module;

[0067] The account management module is used to import the account information of the online platform where the agent is located to add the agent to the deployment and operation module;

[0068] The portrait management module is used to complete the configuration of agent information;

[0069] The algorithm configuration module and the account grouping module are aggregated and work together to manage the grouping of intelligent entities.

[0070] Furthermore, in the account management module, the administrator can add the agent to the deployment and operation module by importing the account information of the online platform where the agent is located. Specifically, by clicking the corresponding start / stop button, the agent's online / offline function can be freely completed;

[0071] The agent portrait in the portrait management module includes four parts: age, label, location, and introduction, which are described in natural language and depict the anthropomorphic background information of the agent. This part, combined with the strategy algorithm part, can generate more realistic and reasonable posting content for the agent during operation.

[0072] In the account grouping module, the administrator customizes the agent account groups, and then uses the algorithm configuration module to configure different strategy algorithms for different groups of agent clusters; during system operation, the underlying algorithm part can be modeled in a variety of ways, and the administrator can freely configure the behavior algorithm and generation algorithm type of each group of agent clusters; after the configuration is completed, agents in different clusters can run simultaneously in the system with different algorithm logics, thereby exhibiting differentiated behavior patterns, so that the administrator can analyze the effectiveness of the strategy algorithm.

[0073] Specifically, the front-end control system cooperates with the account management module, portrait management module, algorithm configuration module and account grouping module to provide an intuitive control panel for administrators to set and adjust the deployment, configuration information and behavior patterns of the intelligent agent.

[0074] Furthermore, the background monitoring system includes action behavior monitoring, account status monitoring and group monitoring;

[0075] The action behavior monitoring module is responsible for tracking each interactive action of the agent (including posting, clicking, etc.) and presenting it in digital form for the administrator to analyze the execution results of the decision algorithm;

[0076] The account status monitoring module is used to monitor the operating status of the agent account;

[0077] The group monitoring module is used to provide cluster monitoring of intelligent agent accounts for administrators.

[0078] Furthermore, the account status monitoring module provides data results of the agent being followed, forwarded and interacted by human users. In addition, if an account is abnormal (such as abnormal disconnection, inoperability, etc.), an alarm can be sent in real time to ensure the stable operation of the platform;

[0079] The group monitoring module supports the system to monitor the behavior and status of each group and presents it to the administrator in the form of the average value of the indicators within the group. Through this module, the administrator can analyze the strategy effects of different groups to adjust and optimize the algorithm.

[0080] Specifically, the action monitoring module, account status monitoring module, and group monitoring module work together to provide administrators with a clear operational perspective and a quick response mechanism. Administrators can efficiently and conveniently analyze the operating status of the intelligent agent and adjust the operating strategy in a timely manner.

[0081] Furthermore, the deployment and operation system includes an account deployment module, an algorithm execution module and an action execution module.

[0082] The account deployment module is used to allocate an independent thread space to each agent;

[0083] The algorithm execution module is used to provide the configured agent information to the algorithm module to generate a decision strategy;

[0084] The action execution module is used to guide the automated operation program to complete the online behavior execution operation of the intelligent agent according to the strategy.

[0085] Furthermore, after the administrator configures the agent information in the user terminal, the information is transmitted to the information storage module of the database; the same is true for the agent operation status information;

[0086] The decision strategy is generated through an algorithm to guide the automated operation program to complete the online behavior execution operation of the agent according to the strategy; while the agent is executing various behaviors, it transmits the operation execution results (including behavior records and agent account status) to the monitoring module background to complete the update of monitoring information;

[0087] The deployment and operation system deploys accounts in a multi-threaded manner, allocating an independent thread space for each agent; the portrait and algorithm configuration information of each agent are assigned to their own threads to avoid mutual interference; Selenium and ChromeDriver are used inside the thread to operate the browser, and a series of behaviors such as login, logout, and click are automatically executed in the browser by simulating the behavior of real users, ensuring that the agent can run smoothly in the online platform;

[0088] When handling the peak of a large number of intelligent agents, the deployment and operation system uses a thread pool mechanism to ensure that each behavior can be successfully executed in the system under limited computing resources. In addition, the behavior execution of the intelligent agent adopts an asynchronous mechanism to ensure that the system is not affected by a single account failure, thereby improving the stability of the system.

[0089] Furthermore, the strategy algorithm system includes a behavior sequence modeling module and a content generation modeling module.

[0090] The sequence modeling module is used for policy-driven automated behavior execution, aiming to allow the agent to simulate the behavior frequency and behavior sequence of real human users. It is modeled through preset algorithm rules;

[0091] The content generation modeling module is used for text generation. It uses advanced natural language processing technology to automatically generate text content with high relevance and authenticity, and applies it to posting and commenting behaviors.

[0092] Furthermore, the sequence modeling module includes agent behavior frequency and agent behavior sequence. For agent behavior frequency modeling, the present invention uses a power law distribution function to model the agent behavior interval Δt:

[0093] P(Δt)~Δt -α (1)

[0094] By regressing the behavior of a large number of human users, we can obtain the approximate optimal solution of α in the above function (1). Considering the actual situation, the final threshold of Δt is limited to [1min, 10h]. In the end, each agent will perform about 10 to 30 behaviors per day.

[0095] The agent behavior sequence modeling, the present invention uses Markov chain to model the agent behavior state:

[0096] P(X n+1 =x|X0,X1,X2,…,X n ) (2)

[0097] Among them, x is a behavior variable, representing the actions that the agent may perform (including posting, clicking, etc.); X0, X1, X2, …, X n is the sequence of actions that the agent has executed, which constitutes the current state of the agent; X n+1 is the action that the agent will perform;

[0098] By simulating the behavior of a large number of human users, we can obtain the execution probability of the next action in each state in the above models (1) and (2). The system will guide the agent to perform the corresponding action with the corresponding probability. In practical applications, only the most recent 4 action spaces are considered, namely:

[0099] P(X n+1 = z | X n-3 ,X n-2 ,X n-1 ,X n ) (3)

[0100] Based on different sources of human user data, modeling schemes with different distributions can be constructed, which can be configured as a variety of behavioral algorithms as parameters for agent grouping clusters for administrators to choose;

[0101] By combining the user's behavior frequency and behavior sequence modeling scheme, a complete automated behavior execution strategy can be built; mechanical operations such as clicks and forwarding will be directly executed through the deployment and operation module; and advanced operations involving text content such as posting and commenting need to be combined with automated text generation algorithms to complete.

[0102] Furthermore, the content generation modeling module first uses large-scale online platform text data to further train the large language model so that it can understand and simulate specific user group behaviors and language patterns;

[0103] The large language model may be a chatglm3-6B model or other models that can achieve the same function;

[0104] Subsequently, using the preset topics and combining the agent account portraits and behavior sequences under specific grouping clusters, the large language model is used to generate posts and comments under the specified topics.

[0105] Before calling this solution, the policy-driven automatic behavior execution module is used to obtain the posting and commenting behavior sequences of the agent cluster within a period of time, and then the generation model is guided to generate the posting and commenting text content of the corresponding agent;

[0106] In the post generation part, we first randomly select a topic (such as sports, military, etc.) from the topic set preset for the algorithm, and then combine it with the agent's account profile (including age, tags, location, and introduction) to guide the large model to generate posts on the specified topic that are consistent with the account profile. The formal expression is as follows:

[0107] c post =G(t,p)

[0108] where c post represents the generated post content, t represents the selected topic vocabulary, p represents the text description of the agent account portrait, and G represents the language model further trained on large-scale online platform text data; by changing the topic t to a specified product or opinion, text content that promotes brand marketing and opinion dissemination can also be generated;

[0109] In the comment generation part, we first randomly select a text from the generated post content, and then combine it with the account portrait of the agent to guide the large model to generate comments under the specified content that are consistent with the account portrait. The formal expression is as follows:

[0110] ccomment =G(c post ,p)

[0111] Based on different preset topic sets and large language models, different generation schemes can be constructed for administrators to configure the generation algorithm of the agent cluster.

[0112] Specifically, the text generated by the strategy algorithm module will be published to the online platform by the deployment and operation module at an appropriate time; this strategy-driven automated text generation greatly improves the quality and efficiency of the content, and also plays a key role in maintaining the continuity and consistency of the intelligent agent's activities; through intelligent content generation, the present invention can increase the interactive depth of the intelligent agent account, while improving the overall effectiveness and user appeal of the intelligent agent modeling strategy.

[0113] Implementation Method 2

[0114] This embodiment uses the online agent automatic deployment management system as described in the first embodiment to implement a working method of the online agent automatic deployment management system.

[0115] First, based on the existing behavior sequence information of the agent, the behavior decision module is used to model the agent's next behavior (such as posting, commenting); after determining the behavior action, the system uses the large language model to output the content to be published based on the agent's portrait and the corresponding patterns of different generated behaviors; finally, through the automated operation module, the agent can perform the corresponding actions on the corresponding online platform. The process is described in detail as follows:

[0116] Behavior decision modeling: The behavior of online agents is driven by the decision module, which uses machine learning algorithms to model the historical behavior sequence of the agent. In this process, the system predicts and sets the agent's future behavior strategy by analyzing and learning the agent's past interaction patterns and outputs it to subsequent modules.

[0117] Content generation strategy: After determining the behavior, the system will use natural language generation technology to generate text content that conforms to the agent portrait and the predetermined behavior pattern. The model input contains key information in the agent portrait, such as age, interests, location, and user behavior tendencies, and then uses the trained specific online platform styled large language model to automatically generate highly attractive and engaging content to provide content support for posting and commenting behaviors.

[0118] Automatic operation module: After the content is generated, the system implements the behavior of the agent through the automatic operation module. This module uses multi-threading technology to ensure the efficient execution of the target behavior of each agent. In addition, this module also has an exception handling mechanism, which can adjust the strategy or retry the operation in time when encountering problems such as login failure and response timeout, so as to ensure the continuity and stability of the online activities of the agent.

[0119] Implementation Method 3

[0120] An embodiment of the present invention provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the memory is used to store software programs and modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory and the processor are connected via a bus. Specifically, the processor implements any step in the first embodiment above by running the computer program stored in the memory.

[0121] It should be understood that in the embodiments of the present invention, the processor referred to may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0122] The memory may include a read-only memory, a flash memory, and a random access memory, and provides instructions and data to the processor. A part or all of the memory may also include a nonvolatile random access memory.

[0123] It should be understood that if the above-mentioned integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned implementation method, and can also be completed by instructing the relevant hardware through a computer program. The above-mentioned computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method implementations when executed by the processor. Among them, the above-mentioned computer program includes computer program code, and the above-mentioned computer program code can be in source code form, object code form, executable file or some intermediate form. The above-mentioned computer-readable medium may include: any entity or device capable of carrying the above-mentioned computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the above-mentioned computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.

[0124] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest range consistent with the principles and novel features disclosed herein.

[0125] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the above-mentioned device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the implementation method can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method implementation method, which will not be repeated here.

[0126] It should be noted that the methods and detailed examples provided in the above embodiments can be combined with the devices and equipment provided in the embodiments, and references can be made to each other, and no further details will be given.

[0127] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0128] In the embodiments provided by the present invention, it should be understood that the disclosed device / terminal equipment and method can be implemented in other ways. For example, the device / equipment implementation described above is only illustrative, for example, the division of the above modules or units is only a logical function division, and in actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0129] The above-mentioned embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above-mentioned embodiments, or replace some of the technical features therein by equivalents. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. An online intelligent agent automatic deployment management system, characterized in that: The system includes a front-end control system, a back-end monitoring system, a deployment and operation system, and a strategy algorithm system. The front-end control system is used to provide an intuitive control panel for administrators to set and adjust the deployment, configuration information and behavior mode of the intelligent agent; The background monitoring system is used to monitor the operation process of the intelligent agent in detail and provide real-time and accurate information feedback to the administrator; The deployment and operation system is used to convert the agent behavior strategy predetermined for the online platform into specific behavior; The strategy algorithm system is used to provide an efficient and convenient automation solution for the operation of intelligent entities.

2. According to claim 1, an online intelligent agent automatic deployment management system is characterized in that: The front-end control system includes an account management module, a portrait management module, an algorithm configuration module and an account grouping module; The account management module is used to import the account information of the online platform where the agent is located to add the agent to the deployment and operation module; The portrait management module is used to complete the configuration of the agent information; The algorithm configuration module and the account grouping module are aggregated and work together to manage the grouping of intelligent entities.

3. According to claim 2, an online intelligent agent automatic deployment management system is characterized in that: The account management module can freely complete the online / offline function of the intelligent agent by clicking the corresponding start / stop button; The agent portrait in the portrait management module includes four parts: age, label, location, and introduction, which are described in natural language and depict the anthropomorphic background information of the agent; In the account grouping module, the administrator customizes the agent account groups, and then uses the algorithm configuration module to configure different strategy algorithms for different groups of agent clusters; after the configuration is completed, the agents in different clusters can run simultaneously in the system with different algorithm logics, thereby exhibiting differentiated behavior patterns, so that the administrator can analyze the effectiveness of the strategy algorithm.

4. According to claim 1, an online intelligent agent automatic deployment management system is characterized in that: The background monitoring system includes action behavior monitoring, account status monitoring and group monitoring; The action behavior monitoring module is used for the administrator to analyze the execution results of the decision-making algorithm; The account status monitoring module is used to monitor the operating status of the agent account; The group monitoring module is used to provide cluster monitoring of intelligent agent accounts for administrators.

5. According to claim 4, an online intelligent agent automatic deployment management system is characterized in that: The account status monitoring module provides data results of the agent being followed, forwarded and interacted by human users; The group monitoring module supports the system to monitor the behavior and status of each group, and presents it to the administrator in the form of the mean of the indicators within the group for algorithm adjustment and optimization.

6. According to claim 1, an online intelligent agent automatic deployment management system is characterized in that: The deployment and operation system includes an account deployment module, an algorithm execution module and an action execution module. The account deployment module is used to allocate an independent thread space to each agent; The algorithm execution module is used to provide the configured agent information to the algorithm module to generate a decision strategy; The action execution module is used to guide the automated operation program to complete the online behavior execution operation of the intelligent agent according to the strategy.

7. The online agent automatic deployment management system according to claim 6, characterized in that: After the administrator configures the agent information on the user terminal, the information is transmitted to the information storage module of the database; the same applies to the agent operation status information; The decision strategy is generated through the algorithm, guiding the automated operation program to complete the online behavior execution operation of the intelligent agent according to the strategy; While executing various behaviors, the agent transmits the operation execution results to the monitoring module background to complete the update of monitoring information; The deployment and operation system deploys accounts in a multi-threaded manner, allocating an independent thread space to each agent; the portrait and algorithm configuration information of each agent are assigned to their own threads; Selenium and ChromeDriver are used inside the thread to operate the browser; When processing the peak operation of a large number of intelligent agents, the deployment and operation system uses a thread pool mechanism to ensure that each behavior can be successfully executed in the system under limited computing resources.

8. The online agent automatic deployment management system according to claim 1, characterized in that: The strategy algorithm system includes a behavior sequence modeling module and a content generation modeling module. The sequence modeling module is used for policy-driven automated behavior execution; The content generation modeling module is used for text generation.

9. The online agent automatic deployment management system according to claim 8, characterized in that: The sequence modeling module includes agent behavior frequency and agent behavior sequence, The agent behavior frequency modeling adopts the power law distribution function to model the agent's behavior interval Δt: P(Δt)~Δt -α (1) The approximate optimal solution of α in the above function (1) is obtained to be between [1min, 10h]; that is, each agent will perform approximately 10 to 30 actions per day; The agent behavior sequence modeling adopts Markov chain to model the agent's behavior state: P(X n+1 =x∣X0,X1,X2,…,X n ) (2) Among them, x is the behavior variable, representing the possible actions of the agent; X0, X1, X2, …, X n is the sequence of actions executed by the agent, which constitutes the current state of the agent; X n+1 is the action that the agent will perform; By simulating the behavior of a large number of human users, the execution probability of the next action in each state in the above models (1) and (2) is obtained, and the system will guide the intelligent agent to perform the corresponding action with the corresponding probability; By combining the user's behavior frequency and behavior sequence modeling scheme, a complete automated behavior execution strategy is constructed; mechanical operations will be directly executed by deploying and running modules; and advanced operations involving text content need to be combined with automated text generation algorithms to complete.

10. The online agent automatic deployment management system according to claim 8, characterized in that: The content generation modeling module uses large-scale online platform text data to further train the large language model so that it can understand and simulate specific user group behaviors and language patterns; Then, by using the preset topics and combining the agent account portraits and behavior sequences under specific grouping clusters, the large language model is used to generate posts and comments under the specified topics.

Images