Virtual behavior generation method and device in semi-supervised environment, equipment and medium

Abstract

The invention relates to the technical field of virtual reality, in particular to a virtual behavior generation method and device in a semi-supervised environment, equipment and a medium, and the method comprises the steps: obtaining a to-be-responded target event of a virtual behavior system; identifying a target generation type of a virtual behavior triggered by the target event; searching historical target data from user log data according to the target generation type, the user log data being obtained after a user manages an automatically stored operation log of the virtual behavior system; and generating a target virtual behavior corresponding to the target event according to the historical target data, and controlling a virtual behavior system to execute the target virtual behavior so as to respond to the target event. Therefore, the problems of deviation between behaviors and user intentions, incapability of forming personalized behavior styles, heavy operation burden, low response speed, difficulty in meeting efficient interaction requirements and the like in complex situations of related technologies are solved.

Description

Technical Field

[0001] This application relates to the field of virtual reality technology, and particularly to a virtual behavior generation method, device, equipment and medium in a semi-supervised environment. Background Art

[0002] With the rapid development of VR (Virtual Reality), AI (Artificial Intelligence) and big data technologies, virtual behavior systems have gradually become important technical tools in modern society. Such systems aim to establish in-depth interactions with users through virtual environments and endow virtual agents with the ability of autonomous decision-making and communication, and are widely used in scenarios such as virtual social interaction, distance education, digital office, and human-machine collaboration. In these applications, the autonomy and humanization level of virtual behavior systems directly affect the user experience and the actual effectiveness of the systems.

[0003] In related technologies, virtual behavior systems mainly rely on fully automated behavior generation mechanisms, that is, the systems make decisions and communicate independently based on algorithmic logic. Although this method can meet basic functional requirements, it is prone to deviations between behaviors and user intentions in complex situations. For example, when the system faces scenarios it has never encountered or is induced by other users, it may make wrong decisions or unreasonable behaviors, thus causing user dissatisfaction or disorder in the virtual environment.

[0004] In addition, fully automated behavior systems lack the learning and reference of users' historical behaviors and cannot form personalized behavior styles, resulting in a single experience and insufficient interactivity for users in long-term use. On the other hand, virtual behavior systems that rely entirely on manual control by users are difficult to meet the requirements of efficient interaction due to heavy operation burdens and slow response speeds. Summary of the Invention

[0005] This application provides a virtual behavior generation method, device, equipment and medium in a semi-supervised environment to solve problems in related technologies such as being prone to deviations between behaviors and user intentions in complex situations, being unable to form personalized behavior styles, having heavy operation burdens, slow response speeds, and being difficult to meet the requirements of efficient interaction.

[0006] The first aspect of this application provides a virtual behavior generation method in a semi-supervised environment, including the following steps: obtaining a target event waiting for a response of a virtual behavior system; identifying a target generation type of a virtual behavior triggered by the target event; searching for historical target data from user log data according to the target generation type, where the user log data is obtained after managing the automatically saved operation logs of the virtual behavior system by the user; generating a target virtual behavior corresponding to the target event according to the historical target data, and controlling the virtual behavior system to execute the target virtual behavior to respond to the target event.

[0007] Optionally, the virtual behavior system is an autonomous behavior system associated with an individual stored in the virtual world, having autonomous decision-making and communication functions.

[0008] Optionally, the target generation types include behavior decision types and communication decision types, the historical target data includes historical behavior data and historical communication data, and finding historical target data from the user log data according to the target generation type includes: if the target generation type is a behavior decision type, finding historical behavior data from the user log data; if the target generation type is a communication decision type, finding historical communication data from the user log data.

[0009] Optionally, if the historical target data is historical behavior data, generating a target virtual behavior corresponding to the target event according to the historical target data includes: performing cluster analysis on the historical behavior data; constructing a behavior graph according to the cluster analysis result, where the behavior graph includes multiple historical decision-making schemes; searching for a historical behavior decision-making scheme in the behavior graph that matches the target event, if no historical behavior decision-making scheme that matches the target event is found, then searching for adjacent nodes related to the target event in the behavior graph; generating a new behavior decision-making scheme according to the historical behavior decision-making scheme of the adjacent nodes, if no adjacent nodes related to the target event are found in the behavior graph, then generating a risk event.

[0010] Optionally, if the historical target data is historical communication data, generating a target virtual behavior corresponding to the target event according to the historical target data includes: training a generative adversarial network and a large language processing model using the historical communication data; using the trained generative adversarial network and large language processing model to generate a new communication decision-making scheme consistent with the user's habits.

[0011] The second aspect of this application provides a virtual behavior generation device in a semi-supervised environment, including: an acquisition module for acquiring a target event waiting for a response of the virtual behavior system; an identification module for identifying the target generation type of the virtual behavior triggered by the target event; a search module for finding historical target data from the user log data according to the target generation type, where the user log data is obtained after managing the automatically saved operation logs of the user on the virtual behavior system; a generation module for generating a target virtual behavior corresponding to the target event according to the historical target data, and controlling the virtual behavior system to execute the target virtual behavior to respond to the target event.

[0012] The third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the program to implement the virtual behavior generation method in the semi-supervised environment as described above.

[0013] The fourth aspect of the present application provides a computer-readable storage medium, on which a computer program is stored. The program is executed by a processor to implement the virtual behavior generation method in the semi-supervised environment as described above.

[0014] The fifth aspect of the present application provides a computer program product. When the computer program is executed, it is used to implement the virtual behavior generation method in the semi-supervised environment as described above.

[0015] Thus, the present application includes the following beneficial effects:

[0016] In the embodiments of the present application, by obtaining the target event waiting for a response in the virtual behavior system, identifying the target generation type of the virtual behavior triggered by the event, searching for historical target data from the user log data, generating the target virtual behavior corresponding to the target event according to the historical target data, and controlling the virtual behavior system to execute the target virtual behavior to respond to the target event, it is possible to learn the user's behavior habits for a long time, achieve personalized behavior simulation, and at the same time ensure the user's full control over the log, make autonomous decisions without manual intervention, and be able to build a more reliable and efficient human-computer interaction experience in a complex virtual environment. Thus, the problems in the related art, such as the deviation of behavior from the user's intention in complex situations, the inability to form a personalized behavior style, heavy operation burden, slow response speed, and difficulty in meeting the requirements of efficient interaction, are solved.

[0017] The additional aspects and advantages of the present application will be partially given in the following description, partially become obvious from the following description, or be understood through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and / or additional aspects and advantages of the present application will become obvious and easy to understand from the following description of the embodiments in conjunction with the drawings, where:

[0019] Figure 1 is a schematic flowchart of the virtual behavior generation method in the semi-supervised environment provided by the embodiments of the present application;

[0020] Figure 2 is a schematic diagram of the virtual behavior system in the semi-supervised environment provided by an embodiment of the present application;

[0021] Figure 3 is a schematic diagram of autonomous decision-making search provided by an embodiment of the present application;

[0022] Figure 4 is an example diagram of the virtual behavior generation device in the semi-supervised environment provided by the embodiments of the present application;

[0023] Figure 5 is a schematic diagram of the structure of an electronic device provided by the embodiments of the present application. Detailed Implementation Manner

[0024] The embodiments of the present application will be described in detail below. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present application, but should not be construed as a limitation to the present application.

[0025] The virtual behavior generation method, device, equipment, and medium in a semi-supervised environment according to the embodiments of the present application will be described below with reference to the accompanying drawings. In view of the problems in the related art mentioned in the above background technology, such as the deviation between behavior and user intention, the inability to form a personalized behavior style, heavy operation burden, slow response speed, and difficulty in meeting the requirements of efficient interaction in complex scenarios, the present application provides a virtual behavior generation method in a semi-supervised environment. In this method, by obtaining the target event waiting for a response in the virtual behavior system, identifying the target generation type of the virtual behavior triggered by the event, searching for historical target data from the user log data, generating the target virtual behavior corresponding to the target event according to the historical target data, and controlling the virtual behavior system to execute the target virtual behavior to respond to the target event, it can learn the user's behavior habits for a long time, realize personalized behavior simulation, and at the same time ensure the user's full control over the log, make autonomous decisions without manual intervention, and be able to build a more reliable and efficient human-computer interaction experience in a complex virtual environment. Thus, the problems in the related art, such as the deviation between behavior and user intention, the inability to form a personalized behavior style, heavy operation burden, slow response speed, and difficulty in meeting the requirements of efficient interaction in complex scenarios, are solved.

[0026] Specifically, Figure 1 is a schematic flowchart of a virtual behavior generation method in a semi-supervised environment provided by an embodiment of the present application.

[0027] As Figure 1 shown, the virtual behavior generation method in this semi-supervised environment includes the following steps:

[0028] In step S101, obtain the target event waiting for a response in the virtual behavior system.

[0029] In the embodiment of the present application, the virtual behavior system is an autonomous behavior system associated with an individual stored in the virtual world, and has the functions of autonomous decision-making and communication.

[0030] Among them, being associated with an individual refers to the connection between the virtual behavior system and a specific user; autonomy refers to the ability of the system to perform tasks and make decisions without direct manual intervention.

[0031] It can be understood that the virtual behavior system of the embodiments of the present application is a technical framework existing in the virtual world, which can be associated with individual users, has autonomy, and has the ability of self-decision-making and communication.

[0032] In step S102, identify the target generation type of the virtual behavior triggered by the target event.

[0033] Among them, the target generation type includes a behavior decision type and a communication decision type.

[0034] It can be understood that when the embodiments of the present application obtain the target event waiting for a response of the virtual behavior system, they will first evaluate the target generation type of the virtual behavior triggered by the event. If the target event involves a decision-making requirement, it is a behavior decision type; for the target event with a communication requirement, it is a communication decision type.

[0035] In step S103, find historical target data from the user log data according to the target generation type, where the user log data is obtained after managing the operation logs automatically saved by the user for the virtual behavior system.

[0036] Among them, the user log data refers to all the operation logs automatically saved by the virtual behavior system. These logs record the interaction details between the user and the system, which will be described in detail below and will not be elaborated here.

[0037] It can be understood that after the embodiments of the present application determine the target generation type corresponding to the target event, they find relevant historical target data from the user log data, and the user log data therein is obtained by automatically saving and managing the user's past operation logs.

[0038] In the embodiments of the present application, the user's management of the operation logs includes deletion and retention. Among them, the user deletes the logs that do not conform to the user's habits.

[0039] It can be understood that in the embodiments of the present application, the user's management of the operation logs includes deletion and retention. By selectively deleting the operation records that no longer reflect the current behavior pattern or preference settings, the validity and relevance of the log data are maintained.

[0040] In the embodiments of the present application, the target generation type includes a behavior decision type and a communication decision type, and the historical target data includes historical behavior data and historical communication data. Finding historical target data from the user log data according to the target generation type includes: if the target generation type is a behavior decision type, finding historical behavior data from the user log data; if the target generation type is a communication decision type, finding historical communication data from the user log data.

[0041] It can be understood that the historical target data in the embodiments of the present application includes historical behavior data and historical communication data. When the recognized target generation type is a behavior decision type, the historical behavior data in the user log data is searched; when the target generation type is a communication decision type, the historical communication data is searched, ensuring that whether it is for behavior decision-making or communication interaction, a response that is both personalized and meets the user's expectations can be provided.

[0042] In the embodiments of the present application, if the historical target data is historical behavior data, generating a target virtual behavior corresponding to the target event based on the historical target data includes: performing cluster analysis on the historical behavior data; constructing a behavior graph according to the results of the cluster analysis, where the behavior graph includes multiple historical decision-making schemes; searching for the historical behavior decision-making scheme in the behavior graph that matches the target event. If no historical behavior decision-making scheme that matches the target event is found, searching for adjacent nodes related to the target event in the behavior graph; generating a new behavior decision-making scheme according to the historical behavior decision-making scheme of the adjacent nodes. If no adjacent nodes related to the target event are found in the behavior graph, generating a risk event.

[0043] Among them, cluster analysis is a statistical analysis method used to group similar behavior data together to better understand and process this data; the behavior graph is a model constructed based on the results of the cluster analysis, representing the relationships between different historical decision-making schemes. Each node represents an event and its corresponding decision-making scheme, and the weight of the edge can represent the similarity between different events; the adjacent node is the node in the behavior graph that is closest to but does not exactly match the target event, and may provide useful reference information for generating a new decision-making scheme.

[0044] It can be understood that when the embodiments of the present application recognize that the historical target data is historical behavior data, in order to generate a corresponding virtual behavior for the target event, it is necessary to perform cluster analysis on these historical behavior data and construct a behavior graph. This graph contains multiple historical decision-making schemes as nodes and connections reflecting the similarity between these schemes. Then, search for the historical behavior decision-making scheme in the behavior graph that best matches the target event; if a direct match cannot be found, look for adjacent nodes related to the target event, and generate a new behavior decision-making scheme according to the historical behavior decision-making scheme of the adjacent nodes; if neither a direct match of the historical decision-making scheme nor a related adjacent node is found in the behavior graph, mark this event as a risk event and report this situation to the user, so as to ensure that the system's response not only conforms to the user's habits but also ensures safety in uncertain situations, helps improve decision-making accuracy, and at the same time maintains the consistency and security of the user experience.

[0045] In an embodiment of the present application, if the historical target data is historical communication data, generating a target virtual behavior corresponding to a target event based on the historical target data includes: training a generative adversarial network and a large language processing model using the historical communication data; using the trained generative adversarial network and large language processing model to generate a new communication decision-making scheme consistent with the user's habits.

[0046] Among them, the generative adversarial network is a deep learning model composed of a generator and a discriminator, which can generate realistic new data samples through the game process between the two; the large language processing model refers to an artificial intelligence system that can understand and generate natural language, usually based on deep learning technology, and is used for tasks such as text generation and sentiment analysis.

[0047] It can be understood that when the historical target data is identified as historical communication data in an embodiment of the present application, using the user's historical communication data to train a generative adversarial network and a large language processing model allows the model to learn the user's communication habits, including language style, word preference, and communication mode. After sufficient training, the model can be used to generate a new communication decision-making scheme highly consistent with the user's habits, enhancing the interaction ability of the virtual behavior system and ensuring the accuracy and consistency in simulating user communication.

[0048] In step S104, a target virtual behavior corresponding to a target event is generated based on the historical target data, and the virtual behavior system is controlled to execute the target virtual behavior to respond to the target event.

[0049] It can be understood that in an embodiment of the present application, the most suitable target virtual behavior for a corresponding target event is determined based on the historical target data, and the virtual behavior system is controlled to execute this behavior, thereby effectively responding to the target event. Whether it is for complex decision-making or natural and smooth communication interaction, the virtual behavior system can provide both personalized and accurate feedback, ensuring the consistency of the user experience and the efficiency of the system.

[0050] According to the virtual behavior generation method in a semi-supervised environment proposed by an embodiment of the present application, by obtaining the target event waiting for a response of the virtual behavior system, identifying the target generation type of the virtual behavior triggered by the event, searching for historical target data from user log data, generating a target virtual behavior corresponding to the target event based on the historical target data, and controlling the virtual behavior system to execute the target virtual behavior to respond to the target event, it can learn the user's behavior habits for a long time, achieve personalized behavior simulation, ensure the user's full control over the logs, make autonomous decisions without manual intervention, and build a more reliable and efficient human-computer interaction experience in a complex virtual environment.

[0051] The virtual behavior generation method in a semi-supervised environment will be further described below through a specific embodiment.

[0052] As Figure 2 shown, a virtual behavior system in a semi-supervised environment provided by this embodiment is an emotion recognition method that can achieve a high emotion recognition accuracy rate, avoid the leakage of personal privacy information, and help improve the comprehensive understanding of the emotional state by relevant researchers. This embodiment includes the following steps:

[0053] 1. The autonomous virtual behavior system associated with an individual stored in the virtual world will automatically save all operation logs.

[0054] 2. The user regularly backs up the logs and checks the log integrity, and deletes the logs that do not conform to the current user's behavior habits.

[0055] 3. The system performs clustering analysis on the user's historical behavior data to construct a behavior map.

[0056] 4. When the system needs to make an autonomous decision, it first searches in the behavior map for the historical decision-making plan with the highest matching degree to the corresponding event. If the search is successful, it imitates or directly applies this plan; if the search fails, it looks for adjacent nodes related to the current event in the behavior map and comprehensively generates a new decision-making plan from multiple nodes. The autonomous decision-making search method is shown in Figure 3 .

[0057] 5. When no matching information can be found in the behavior map, the system stops making decisions and marks the current event as a risk behavior and reports it to the user.

[0058] 6. When the system needs to communicate autonomously, it uses the user's historical communication data to train a generative adversarial network and a large language processing model to generate communication behaviors such as text, voice, and video that are consistent with the user's habits.

[0059] As Figure 3 shown, in the behavior map of the virtual behavior system, the relationships between different behavior decisions will be represented by a graph. Each node represents a different event and the corresponding decision-making plan, and the weight of the edge between nodes represents the similarity between different events.

[0060] When an event identical to the current event can be found in the behavior map, the decision-making plan of the historical event is directly applied.

[0061] When an event identical to the current event cannot be found in the behavior map, but there are similar events, that is, the similarity between the event searched in the behavior map and the current event is greater than the threshold δ similar at this time, multiple events searched in the behavior map are marked, and a region Z with an area of threshold S that can contain the most marked events is found. The behavior system generates a new decision-making plan from these events included in region Z.

[0062] When there is no similar event, that is, the similarity between the event searched in the behavior graph and the current event is less than the threshold δ similar at this time, mark the current event as a risk event and report it to the user, and stop making decisions on this event.

[0063] Next, a virtual behavior generation device in a semi-supervised environment according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0064] Figure 4 It is a block diagram of a virtual behavior generation device in a semi-supervised environment according to an embodiment of the present application.

[0065] As Figure 4 shown, the virtual behavior generation device 10 in this semi-supervised environment includes: an acquisition module 201, an identification module 202, a search module 203, and a generation module 204.

[0066] Among them, the acquisition module 201 is used to acquire a target event waiting for a response in the virtual behavior system; the identification module 202 is used to identify a target generation type of the virtual behavior triggered by the target event; the search module 203 is used to search for historical target data from user log data according to the target generation type, where the user log data is obtained after managing the operation logs automatically saved by the user for the virtual behavior system; the generation module 204 is used to generate a target virtual behavior corresponding to the target event according to the historical target data, and control the virtual behavior system to execute the target virtual behavior to respond to the target event.

[0067] In the embodiment of the present application, the virtual behavior system is an autonomous behavior system associated with an individual stored in a virtual world, and has autonomous decision-making and communication functions.

[0068] In the embodiment of the present application, the user's management of the operation logs includes deletion and retention, where the user deletes the logs that do not conform to the user's habits.

[0069] In the embodiment of the present application, the search module 203 is further used for: the target generation type includes a behavior decision type and a communication decision type, the historical target data includes historical behavior data and historical communication data, and search for historical target data from the user log data according to the target generation type. If the target generation type is the behavior decision type, search for historical behavior data from the user log data; if the target generation type is the communication decision type, search for historical communication data from the user log data.

[0070] In an embodiment of the present application, the search module 203 is further configured to: if the historical target data is historical behavior data, generate a target virtual behavior corresponding to the target event according to the historical target data, perform clustering analysis on the historical behavior data; construct a behavior map according to the clustering analysis result, where the behavior map includes multiple historical decision-making schemes; search for a historical behavior decision-making scheme in the behavior map that matches the target event, and if no historical behavior decision-making scheme that matches the target event is found, search for adjacent nodes related to the target event in the behavior map; generate a new behavior decision-making scheme according to the historical behavior decision-making scheme of the adjacent nodes, and if no adjacent nodes related to the target event are found in the behavior map, generate a risk event.

[0071] In an embodiment of the present application, the search module 203 is further configured to: if the historical target data is historical communication data, generate a target virtual behavior corresponding to the target event according to the historical target data, including: training a generative adversarial network and a large language processing model using the historical communication data; using the trained generative adversarial network and large language processing model to generate a new communication decision-making scheme consistent with the user's habits.

[0072] It should be noted that the foregoing explanation of the embodiment of the virtual behavior generation method in the semi-supervised environment also applies to the virtual behavior generation device in the semi-supervised environment of this embodiment, and will not be repeated here.

[0073] The virtual behavior generation device in the semi-supervised environment proposed according to the embodiment of the present application can obtain the target event waiting for a response in the virtual behavior system, identify the target generation type of the virtual behavior triggered by the event, search for historical target data from the user log data, generate a target virtual behavior corresponding to the target event according to the historical target data, and control the virtual behavior system to execute the target virtual behavior to respond to the target event. It can learn the user's behavior habits for a long time, realize personalized behavior simulation, and at the same time ensure the user's full control of the log, make autonomous decisions without manual intervention, and be able to build a more reliable and efficient human-computer interaction experience in a complex virtual environment.

[0074] Figure 5 The structural schematic diagram of the electronic device provided for the embodiment of the present application. The electronic device may include:

[0075] A memory 301, a processor 302, and a computer program stored on the memory 301 and executable on the processor 302.

[0076] When the processor 302 executes the program, it implements the virtual behavior generation method in the semi-supervised environment provided in the foregoing embodiment.

[0077] Further, the electronic device further includes:

[0078] A communication interface 303 for communication between the memory 301 and the processor 302.

[0079] A memory 301 for storing computer programs that can run on the processor 302.

[0080] The memory 301 may include a high-speed RAM (Random Access Memory) memory and may also include non-volatile memory, such as at least one disk memory.

[0081] If the memory 301, the processor 302, and the communication interface 303 are implemented independently, the communication interface 303, the memory 301, and the processor 302 can be interconnected through a bus to complete communication with each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 only a thick line is used to represent it in the figure, but it does not mean that there is only one bus or one type of bus.

[0082] Optionally, in a specific implementation, if the memory 301, the processor 302, and the communication interface 303 are integrated on a single chip, the memory 301, the processor 302, and the communication interface 303 can complete communication with each other through an internal interface.

[0083] The processor 302 may be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0084] The embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the virtual behavior generation method in the semi-supervised environment as described above is implemented.

[0085] The embodiments of the present application also provide a computer program product, including a computer program or instruction, and when the computer program or instruction is executed, the virtual behavior generation method in the semi-supervised environment as described above is implemented.

[0086] In the description of this specification, the descriptions with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of this application. In this specification, the schematic expressions of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0087] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of these features. In the description of this application, the meaning of "N" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0088] Any process or method description shown in a flowchart or described in other ways herein can be understood to represent a module, segment, or part of code including one or N executable instructions for implementing a customized logic function or process. And the scope of the preferred embodiments of this application includes additional implementations, where the functions can be executed in a manner that is not in the order shown or discussed, including in a substantially simultaneous manner according to the involved functions or in the reverse order, which should be understood by those skilled in the art to which the embodiments of this application belong.

[0089] It should be understood that each part of this application can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, the steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware as in another embodiment, any one of the following well-known technologies in the art or a combination of them can be used: discrete logic circuits having logic gate circuits for implementing logic functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays, field-programmable gate arrays, etc.

[0090] Those of ordinary skill in the art of this technology can understand that all or part of the steps carried by the methods for implementing the above embodiments can be completed by instructing relevant hardware through a program. The above program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0091] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A virtual behavior generation method in a semi-supervised environment, characterized in that, including the following steps: Obtain the target event waiting for a response in the virtual behavior system; Identify the target generation type of the virtual behavior triggered by the target event; Search for historical target data from the user log data according to the target generation type, where the user log data is obtained after the user manages the automatically saved operation logs of the virtual behavior system; Generate the target virtual behavior corresponding to the target event according to the historical target data, and control the virtual behavior system to execute the target virtual behavior to respond to the target event.

2. The virtual behavior generation method in a semi-supervised environment according to claim 1, wherein The virtual behavior system is an autonomous behavior system associated with an individual stored in a virtual world, with autonomous decision-making and communication functions.

3. The virtual behavior generation method in a semi-supervised environment according to claim 1, wherein The user's management of the operation logs includes deletion and retention, where the user deletes the logs that do not conform to the user's habits.

4. The virtual behavior generation method in a semi-supervised environment according to claim 1, wherein The target generation type includes a behavior decision type and a communication decision type. The historical target data includes historical behavior data and historical communication data. Searching for historical target data from the user log data according to the target generation type includes: If the target generation type is the behavior decision type, search for historical behavior data from the user log data; If the target generation type is the communication decision type, search for historical communication data from the user log data.

5. The virtual behavior generation method in a semi-supervised environment according to claim 4, characterized in that, If the historical target data is historical behavior data, generating the target virtual behavior corresponding to the target event according to the historical target data includes: Perform clustering analysis on the historical behavior data; Construct a behavior map according to the clustering analysis results, where the behavior map includes multiple historical decision-making schemes; Search for the historical behavior decision-making scheme in the behavior map that matches the target event. If no historical behavior decision-making scheme that matches the target event is found, search for the neighboring nodes related to the target event in the behavior map; Generate a new behavior decision-making scheme according to the historical behavior decision-making scheme of the neighboring nodes. If no neighboring nodes related to the target event are found in the behavior map, generate a risk event.

6. The virtual behavior generation method in a semi-supervised environment according to claim 4, wherein If the historical target data is historical communication data, generating the target virtual behavior corresponding to the target event according to the historical target data includes: Use the historical communication data to train and generate a generative adversarial network and a large language processing model; Use the trained generative adversarial network and the large language processing model to generate a new communication decision-making scheme consistent with the user's habits.

7. A virtual behavior generation device in a semi-supervised environment, characterized in that, including: An acquisition module for obtaining the target event waiting for a response in the virtual behavior system; An identification module for identifying the target generation type of the virtual behavior triggered by the target event; A search module for searching for historical target data from the user log data according to the target generation type, where the user log data is obtained after the user manages the automatically saved operation logs of the virtual behavior system; A generation module for generating the target virtual behavior corresponding to the target event according to the historical target data, and controlling the virtual behavior system to execute the target virtual behavior to respond to the target event.

8. An electronic device, characterized in that, including: A memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the virtual behavior generation method in a semi-supervised environment according to any one of claims 1-6.

9. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instruction is executed, it implements the virtual behavior generation method in a semi-supervised environment according to any one of claims 1-6.

10. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instruction is executed, it implements the virtual behavior generation method in a semi-supervised environment according to any one of claims 1-6.

Images