Object processing method and device

Abstract

This application provides an object processing method and apparatus, wherein the object processing method includes: obtaining personalized data of a target object; determining multiple candidate state data of the target object based on the personalized data; parsing each candidate state data based on the personalized data to obtain the state strength of the candidate state data; determining target state data from the multiple candidate state data based on the state strengths, and executing the target state data on the target object. This solution can improve the personalization and autonomy of the object state.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and more particularly to an object processing method, an object processing apparatus, a computing device, and a computer-readable storage medium. Background Art

[0002] With the development of Internet technology, virtual stars, virtual pets, virtual friends, non-player characters (NPCs) in games and other objects have been widely used.

[0003] In related technologies, the state of an object is typically determined based on user input. This state can include actions and speech, among other things. However, this approach can easily lead to the object's state becoming rigid, similar to typical human-computer interactions. Therefore, a solution is needed that allows for a state that is more human-like, with human autonomy and personalization. Summary of the Invention

[0004] In view of this, the embodiments of the present application provide an object processing method to solve the technical defects existing in the prior art. The embodiments of the present application also provide an object processing apparatus, a computing device, and a computer-readable storage medium.

[0005] According to a first aspect of an embodiment of the present application, there is provided an object processing method, comprising:

[0006] Obtain personalized data of target objects;

[0007] determining a plurality of candidate state data of the target object according to the personalized data;

[0008] Based on the personalized data, each candidate state data is analyzed respectively to obtain the state strength of the candidate state data;

[0009] According to each state strength, target state data is determined from the plurality of candidate state data, and the target state data is executed for the target object.

[0010] According to a second aspect of an embodiment of the present application, there is provided an object processing apparatus, including:

[0011] a candidate state data determination module configured to obtain personalized data of a target object; and determine a plurality of candidate state data of the target object based on the personalized data;

[0012] a state strength determination module configured to analyze each candidate state data based on the personalized data to obtain the state strength of the candidate state data;

[0013] The target state data determination module is configured to determine target state data from the plurality of candidate state data according to each state strength, and execute the target state data for the target object.

[0014] Optionally, the number of the personalized data is multiple;

[0015] The candidate status data determination module is further configured to:

[0016] Determine the target source type of each personalized data of the target object respectively, and for each target source type, find the target state determination method corresponding to the target source type from the pre-established correspondence between the source type and the state determination method;

[0017] For each target source type, using the target state determination method corresponding to the target source type, the personalized data of the target object with the target source type is processed to obtain state data of the target source type;

[0018] The obtained state data of each target source type is determined as a plurality of candidate state data of the target object.

[0019] Optionally, the target source type includes: a demand type that characterizes the demand of the target object;

[0020] The candidate status data determination module is further configured to:

[0021] Searching for the demand-type status data corresponding to the personalized data of the target object from the pre-established correspondence between the personalized data and the demand-type status data;

[0022] Based on the searched requirement-type status data, status data of the requirement type is determined.

[0023] Optionally, the number of personalized data with the required type is multiple;

[0024] The candidate status data determination module is further configured to:

[0025] Obtaining the demand intensity of each demand corresponding to each personalized data of the demand type, and determining target personalized data whose demand intensity reaches a first intensity threshold from the personalized data of the demand type;

[0026] The demand-type state data corresponding to the target personalized data is searched from the pre-established correspondence between personalized data and demand-type state data.

[0027] Optionally, the candidate status data determination module is further configured to:

[0028] Determining, based on the target personalized data, the state strength of the demand-type state data corresponding to the target personalized data;

[0029] The state intensities corresponding to the searched demand-type state data are used as weights, and a preset weight random model is used to determine the state data of the demand type from the searched demand-type state data.

[0030] Optionally, the target source type includes: a perception type;

[0031] The candidate status data determination module is further configured to:

[0032] determining a target object type of the target object in the personalized data of the perception type;

[0033] From the pre-established correspondence between object types, personalized data and feedback status data, feedback status data corresponding to both the target object type and the personalized data of the perception type is searched.

[0034] Optionally, the target source type includes: a target type associated with a specified target of the target object;

[0035] The candidate status data determination module is further configured to:

[0036] Obtaining the execution order of multiple sub-goals in the specified goal;

[0037] According to the execution order, determining, from the multiple sub-goals, the sub-goals whose target objects have not been completed and whose execution order meets the preset priority condition;

[0038] From the correspondence between the personalized data, the sub-goals and the target-type state data, the target-type state data corresponding to both the personalized data of the target type and the determined sub-goals is searched.

[0039] Optionally, the state strength determination module is further configured to:

[0040] Determine the target execution type for each candidate state data;

[0041] According to the target execution type of each candidate state data, from the pre-established correspondence between execution type and success rate model, a success rate model corresponding to the target execution type is searched and determined as the success rate model of the candidate state data;

[0042] For each candidate state data, determining the success rate of the candidate state data using the success rate model of the candidate state data according to the personalized data corresponding to the candidate state data;

[0043] The state strength of each candidate state data is determined based on the success rate of the candidate state data.

[0044] Optionally, the state strength determination module is further configured to:

[0045] Determining at least one of a basic strength, a success weight, and a failure weight of each candidate state data according to the personalized data corresponding to the candidate state data;

[0046] Determine the failure rate of each candidate state data according to the success rate of the candidate state data;

[0047] For each candidate state data, the success rate and failure rate of the candidate state data are processed using at least one of the basic strength, the success weight, and the failure weight of the candidate state data to obtain the state strength of the candidate state data.

[0048] Optionally, the state strength determination module is further configured to:

[0049] Obtaining current status data of the target object;

[0050] From the current state data and the plurality of candidate state data, first state data whose state strength reaches a first preset threshold is determined as target state data.

[0051] Optionally, the state strength determination module is further configured to:

[0052] Determine whether the state type of the first state data is the same as that of the current state data; if different, use the first state data as target state data when the first state data reaches a second preset threshold.

[0053] According to a third aspect of an embodiment of the present application, a computing device is provided, including:

[0054] memory and processor;

[0055] The memory is used to store computer-executable instructions, and the processor implements the steps of the object processing method when executing the computer-executable instructions.

[0056] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of the object processing method are implemented.

[0057] In the solution provided by this application, personalized data of a target object is obtained; multiple candidate state data for the target object are determined based on the personalized data; each candidate state data is parsed based on the personalized data to obtain the state strength of the candidate state data; target state data is determined from the multiple candidate state data based on the state strengths, and the target state data is executed for the target object. The personalized data of the target object can represent the target object's own characteristics. Furthermore, the state strengths of the candidate state data obtained based on the personalized data of the target object can reflect the differences in the probability of the target object executing different state data based on the target object's own characteristics. Therefore, determining the target state data based on the state strengths ensures that the state achieved by executing the target state data is consistent with the target object's own characteristics. Therefore, a personalized state that better suits the target object's own characteristics can be achieved for the target object. Furthermore, the implementation of this solution does not rely on user input, but is based on the personalized data of the target object, which is equivalent to the target object autonomously achieving the state. Therefore, the autonomy of the object's state realization can be improved. As can be seen, this solution can enhance the personalization and autonomy of the object's state. Furthermore, this application standardizes and abstracts the data related to object state implementation using the target object's personalized data, state data, and state strength. This ensures that this application can be applied to implement state implementation for different objects and a large number of objects, eliminating the need for specialized state implementation development for each object. Therefore, this solution can reduce development time for different objects and a large number of objects, improving development efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a flowchart of an object processing method provided by an embodiment of the present application;

[0059] Figure 2 This is a flow chart of a method for determining state data of a requirement type in an object processing method provided by an embodiment of the present application;

[0060] Figure 3 This is an example diagram of a method for determining state strength in an object processing method provided in one embodiment of the present application;

[0061] Figure 4 is a processing flow chart of another object processing method provided by an embodiment of the present application;

[0062] Figure 5 This is a schematic structural diagram of an object processing device provided in one embodiment of the present application;

[0063] Figure 6 This is a structural block diagram of a computing device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0064] The following description sets forth many specific details to facilitate a thorough understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the scope of the present application. Therefore, the present application is not limited to the specific implementations disclosed below.

[0065] The terms used in one or more embodiments of the present application are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of the present application. The singular forms "a", "the" and "the" used in one or more embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of the present application refers to and includes any or all possible combinations of one or more associated listed items.

[0066] It should be understood that although the terms "first," "second," and the like may be used to describe various information in one or more embodiments of the present application, such information should not be limited to these terms. These terms are merely used to distinguish information of the same type from one another. For example, "first" may also be referred to as "second," and similarly, "second" may also be referred to as "first," without departing from the scope of one or more embodiments of the present application.

[0067] This application provides an object processing method, an object processing apparatus, a computing device, and a computer-readable storage medium, which are described in detail in the following embodiments.

[0068] Figure 1 A flowchart of an object processing method provided according to an embodiment of the present application is shown, which specifically includes the following steps:

[0069] S102: Obtain personalized data of the target object.

[0070] The target subject's personalized data is used to characterize the target subject's own characteristics, equivalent to a portrait of the target subject, and can be of multiple types. Exemplarily, the target subject's personalized data may include at least one of: needs data, emotional data, temperament data, memory data, and interpersonal data. Specifically, needs data is used to indicate the state the target subject desires to achieve. For example, needs data such as eating and sleeping can be specifically represented by need labels, need stages, and need intensity. Emotional data is used to indicate the target subject's emotions and can influence their state. Exemplarily, emotional data can be represented using the following emotional formula: E(N+1) = E(N) * attenuation factor + E(N) * personality influence matrix + E(N) * mutual influence matrix + weighted external stimulus change. E(N) represents the emotion vector at the previous moment, and E(N+1) represents the emotion vector at the next moment.

[0071] Furthermore, temperament data is used to indicate the temperament of the target object. Temperament may include the character and / or preferences of the target object. For example, a character may be righteous, willful, or short-tempered; preferences may include smoking, drinking, reading, traveling, and the like. For example, specific "values" or "labels" may be used as temperament data. Values are suitable for representing personalities that vary in degree, such as righteousness, willfulness, and the desire to survive; labels are suitable for representing "yes" or "no" personalities, such as smoking, drinking, and the like.

[0072] Furthermore, memory data refers to information stored in the target object's memory bank. The target object can store information perceived in any way in the memory bank. Interpersonal relationship data is used to indicate the relationship between the target object and other objects. To further personalize the target object's status, interpersonal relationship data can be configured according to the following two principles: First, target objects can be interconnected with each other, forming a network-like interpersonal network. Second, interpersonal relationships can be represented in a multi-layered, bidirectional manner, rather than a single "favorability" scale. For example, interpersonal relationship data could indicate that target object A admires target object B, while target object B does not know target object A. Specifically, numerical interpersonal relationship data can include: a love-hate value, which indicates the degree of willingness to help the target object; a negative value indicates the degree of harm; an admiration-dislike value, which indicates the degree of willingness to be with the target object; a negative value indicates a desire not to get close; and a pressure-fear value, which indicates a belief that the target object will obey; a negative value indicates a fear that the target object will not resist. In specific applications, the bidirectional and multi-dimensional characterization of interpersonal relationships may result in an increase in data volume. Therefore, a limit can be set on the number of interpersonal networks for each target object. If the limit is exceeded, the target object in the network will be deleted. Specific deletion rules may include: for a target object, deleting a specified number of objects in the target object's interpersonal network whose preset memory strength is less than a memory strength threshold; if the memory strengths are the same, deleting objects that were added earlier in the interpersonal network.

[0073] S104: Determine multiple candidate state data of the target object according to the personalized data.

[0074] In specific applications, multiple methods can be used to determine multiple candidate state data for a target object based on personalized data. For example, multiple candidate state data corresponding to the personalized data of the target object can be found from a pre-established correspondence between personalized data and state data. Alternatively, the type of personalized data can be determined, and different methods can be used to determine multiple candidate state data for different types of personalized data. To facilitate understanding and provide a reasonable layout, the second example will be described in detail below as an optional embodiment.

[0075] S106: Analyze each candidate state data based on the personalized data to obtain the state strength of the candidate state data.

[0076] In specific applications, based on personalized data, each candidate state data is parsed separately to obtain the state strength of the candidate state data, which can be multiple. For example, the personalized data corresponding to each candidate state data can be input into a pre-trained state strength model to obtain the state strength of the candidate state data. Among them, the state strength model is a neural network model trained using sample personalized data and the state strength labels of the sample state data corresponding to the sample personalized data. Or, for example, the target execution type of each candidate state data can be determined, and then the success rate calculation model corresponding to the target execution type can be used to obtain the success rate of the candidate state data, and the state strength of the candidate state data can be determined based on the success rate. In order to facilitate understanding and reasonable layout, the second example will be specifically described in the form of an optional embodiment.

[0077] S108 , determining target state data from a plurality of candidate state data according to the strength of each state, and executing the target state data on the target object.

[0078] In specific applications, there are various ways to determine target state data from multiple candidate state data based on the strength of each state, and execute the target state data on the target object. For example, the target state data can be determined based on the state strength of the multiple candidate state data reaching a first preset threshold. Alternatively, the target state data can be determined based on candidate state data from the multiple candidate state data that is of a different type than the current state data of the target object. For ease of understanding and reasonable layout, the second example will be described in detail below in the form of an optional embodiment.

[0079] In the solution provided by this application, the personalized data of a target object can represent the target object's own characteristics. Furthermore, the state strength of the candidate state data obtained based on the target object's personalized data can reflect the differences in the probability of the target object executing different state data, depending on the target object's own characteristics. Therefore, determining the target state data based on the state strength ensures that the state achieved by executing the target state data is consistent with the target object's own characteristics. Therefore, a personalized state that better suits the target object's own characteristics can be achieved for the target object. Furthermore, the implementation of this solution does not rely on user input, but is instead based on the target object's personalized data, equivalent to the target object autonomously achieving its state. Therefore, the autonomy of object state realization can be improved. Furthermore, this application uses the target object's personalized data, state data, and state strength to standardize, abstract, and unify the data related to achieving the object's state. This ensures that this application can be applied to achieving state realization for different objects and a large number of objects, eliminating the need for specialized state realization development for each object. Therefore, this solution can reduce development time for different objects and a large number of objects, thereby improving development efficiency.

[0080] In an optional embodiment, the number of personalized data is multiple;

[0081] Accordingly, the above-mentioned determination of multiple candidate status data of the target object based on the personalized data may specifically include the following steps:

[0082] Determine the target source type of each personalized data of the target object respectively, and for each target source type, find the target state determination method corresponding to the target source type from the pre-established correspondence between the source type and the state determination method;

[0083] For each target source type, using the target state determination method corresponding to the target source type, the personalized data of the target object with the target source type is processed to obtain the state data of the target source type;

[0084] The obtained state data of each target source type is determined as a plurality of candidate state data of the target object.

[0085] In specific applications, the target source types of personalized data can be divided according to the source differences of the personalized data. Exemplary target source types of personalized data may include: time type, demand type, perception type, goal type, and task type. This embodiment divides personalized data according to source differences, thereby processing personalized data of different source types in different ways, thus balancing the diversity and accuracy of status data. The following describes in detail, using optional embodiments, how to obtain status data corresponding to personalized data of different source types.

[0086] In an optional embodiment, the target source type includes: time type;

[0087] Accordingly, for each target source type, the target state determination method corresponding to the target source type is used to process the personalized data of the target object with the target source type to obtain the state data of the target source type, which may specifically include the following steps:

[0088] From the pre-established correspondence between personalized data and time-based state data, the time-based state data corresponding to the personalized data of the target object is searched.

[0089] In specific applications, time-type state data refers to state data triggered by time. For example, in time period T1, the target object executes target state S1, and in time period T2, the target object executes target state S2. The correspondence between pre-established personalized data and time-type state data can be regarded as a timetable, including time periods and time-type state data corresponding to the time periods. Each target object can only have one timetable. In application scenarios where the target object is an NPC, different timetables can be bound to the target object based on the target object's occupation type. The schedule is a very cost-effective mechanism for expressing "real and vivid" images, with low production and performance costs. For example, a schedule can be used to present a virtual scene about a village: the target state of the target adult is to work from sunrise to sunset and rest, and the target state of the target child is to play after dinner, sleep after playing tired, and other regular states.

[0090] In addition, when the source type of personalized data is multiple, the state data target object recorded in the timetable can be executed, which can be specifically achieved through this application. Figure 1 In the embodiment, the method of determining the target state determines whether to execute. In this way, while achieving a regular state that conforms to "human nature" through time, it can avoid mechanical fixed state implementation and further improve the personalization of the state.

[0091] In an optional embodiment, the target source type includes: a demand type that characterizes the demand of the target object;

[0092] Accordingly, for each target source type, the target state determination method corresponding to the target source type is used to process the personalized data of the target object with the target source type to obtain the state data of the target source type, which may specifically include the following steps:

[0093] Searching for the demand-type state data corresponding to the personalized data of the target object from the pre-established correspondence between the personalized data and the demand-type state data;

[0094] Based on the searched requirement-type status data, status data of the requirement type is determined.

[0095] In specific applications, the pre-established correspondence between personalized data and need-based state data can serve as a need tag, and state data matching the need tag can be used to satisfy the needs of the target subject. For example, a pre-established correspondence between personalized data and need-based state data might include: a hunger value of 70 corresponds to consuming food with a fullness value of 30. Then, by executing the target state data for the target subject, the target subject can satisfy the need "eating" by consuming noodles with a fullness value of 30. Furthermore, the need stage refers to the ratio between the decay result of the initial value set for the need and the initial value of the need. For example, needs such as eating and sleeping have initial values that decay according to a preset decay rule, generating a decay result. This is equivalent to setting a progress bar for the need, and the decay result, i.e., the ratio of the current value of the progress bar to the initial value of the need, determines the current stage, such as fullness or hunger. When the decay result falls below the need threshold, the target subject can execute state data that increases the decay result, satisfying the target subject's need, such as eating. This achieves a state that can change autonomously, similar to human needs, and the state changes in a gradient, further enhancing the humanization and personalization of the state. Furthermore, different preset decay rules can be set for the above different stages. For example, different decay speeds can be set to achieve a more humane state based on the different states of wakefulness and sleep of the target subject. For example, hunger increases quickly when awake and decreases slowly when asleep, etc.

[0096] In addition, the demand intensity is described in detail below in the form of an optional embodiment.

[0097] In an optional embodiment, the number of personalized data with the required type is multiple;

[0098] Accordingly, searching for the demand-type status data corresponding to the personalized data of the target object from the pre-established correspondence between the personalized data and the demand-type status data may specifically include the following steps:

[0099] Obtaining the demand intensity of each demand corresponding to each personalized data of the demand type, and determining target personalized data whose demand intensity reaches a first intensity threshold from the personalized data of the demand type;

[0100] The demand-type state data corresponding to the target personalized data is searched from the pre-established correspondence between personalized data and demand-type state data.

[0101] In specific applications, need strength refers to the intensity of a target's desire to fulfill a need. Therefore, by obtaining the need strength corresponding to a need in personalized data, it is possible to determine the target need, i.e., the state data of the need type, even if the target has multiple needs. This improves the accuracy of determining the target's state.

[0102] Furthermore, there may be multiple specific ways to obtain the demand intensity of the demand corresponding to each personalized data of the demand type. For example, each personalized data of the demand type may be input into a pre-trained demand intensity model to obtain the demand intensity of the demand corresponding to the personalized data. The demand intensity model is pre-trained using the personalized data of the sample demand type and the demand intensity label of the personalized data of the sample demand type. Alternatively, for example, if each personalized data of the demand type includes a demand stage, the demand intensity of the demand corresponding to each personalized data of the demand type may be found from the pre-established correspondence between the demand stage and the demand intensity. For example, the demand stage may include: 0 for full, 0 for normal, 0.1 for slightly hungry, 0.4 for hungry, and 1 for extremely hungry. When the target object's demand stage is extremely hungry, the demand intensity increases by 4 times. According to the embodiment of the present application, it is likely that the target object will give up what he is doing and execute the status data about eating, thereby further improving personalization and autonomy.

[0103] Furthermore, there may be multiple ways to determine the demand-type status data based on the demand-type status data searched, which are described below in the form of optional embodiments.

[0104] In an optional implementation, the above-mentioned determination of the demand-type status data based on the searched demand-type status data may specifically include the following steps:

[0105] The searched demand-type status data is determined to be demand-type status data.

[0106] In another optional implementation, the above-mentioned determination of the demand type status data based on the searched demand type status data may specifically include the following steps:

[0107] Determining, based on the target personalized data, the state strength of the demand-type state data corresponding to the target personalized data;

[0108] The state intensities corresponding to the searched demand-type state data are used as weights, and a preset weight random model is used to determine the state data of the demand type from the searched demand-type state data.

[0109] Among them, the weighted random algorithm is a common algorithm in deep learning and is generally used in the following scenario: there is a set S, which contains four items, such as A, B, C, and D. At this time, you want to randomly extract one item from it, but with different extraction probabilities, for example, you want the probability of extracting A to be 50%, the probability of extracting B and C to be 20%, and the probability of extracting D to be 10%. Generally speaking, you can assign a weight to each item, and the probability of extraction is proportional to this weight. Figure 2 In an object processing method provided in one embodiment of the present application, a flow chart of a method for determining state data of a required type is shown. For a game scene, the target object may be a character in the game. The method includes the following steps:

[0110] Polling status; obtaining all current demand tags TN of the character with demand intensity greater than 0; searching for demand-type state data corresponding to the demand tag TN; if the search is successful, calculating the state intensity corresponding to each demand-type state data, deleting the demand-type state data with state intensity not greater than 0, and obtaining the first set; if the first set is not empty, determining from the first set a second set of demand-type state data with state intensity greater than the hesitation corresponding to the state intensity; determining whether the current state data of the character belongs to the second set, and if not, using the state intensity corresponding to the state data in the second set as the weight, using the preset weight random model to determine the state data of the demand type in the second set; if the current state data of the character belongs to the second set, ending; if the search fails, clearing the state data of the character; if the first set is empty, clearing the state data of the character.

[0111] In specific applications, the polling state refers to determining, in a polling manner, whether the demand intensity corresponding to the personalized data of each demand type reaches a first intensity threshold. The polling method may specifically include triggering a demand intensity determination every 10 minutes during game play. The demand tag TN corresponds to the target personalized data whose demand intensity reaches the first intensity threshold in the above-mentioned embodiment, and may specifically be a set, such as the set TN(). Based on this, searching for demand-type state data corresponding to the demand tag TN may specifically include determining, from objects whose distance from the character reaches a distance threshold, objects corresponding to the demand tag TN, that is, objects carrying any of the tags in the demand tags TN, that are executable. For example, the demand tags TN include: eating, noodles, and hamburgers. Among the objects whose distance from the character reaches the distance threshold, objects carrying any of the tags in the demand tags TN include noodle shops and hamburger shops. If the noodle shop is full, the execution is unavailable, and therefore, the executable object is the hamburger shop. Based on this, and to reduce the unreasonable occurrence of state data with target object execution intensity less than 0, demand-type state data with a state intensity less than 0 may be deleted to obtain the first set. Furthermore, in order to reduce the jitter caused by frequent state switching, a second set of demand-type state data whose state strength is greater than the hesitation corresponding to the state strength is determined from the first set, and when it is determined that the current state data of the character does not belong to the second set, the state strength corresponding to the state data in the second set is used as the weight, and the state data of the demand type is determined in the second set using a preset weight random model. Among them, the hesitation is used to characterize the tendency of the target object between the two state data, and specifically can include the sum between the state strength corresponding to the current state data of the target object and the basic hesitation. The basic hesitation can be set according to specific needs. In addition, a target object can have multiple needs, that is, multiple personalized data of the demand type, at the same time, and a state data can also match at the same time, that is, correspond to multiple personalized data.

[0112] In an optional embodiment, the target source type includes: a perception type;

[0113] Accordingly, for each target source type, the target state determination method corresponding to the target source type is used to process the personalized data of the target object with the target source type to obtain the state data of the target source type, which may specifically include the following steps:

[0114] determining a target object type of the target object in the personalized data of the perception type;

[0115] From the pre-established correspondence between object type, personalized data and feedback status data, feedback status data corresponding to both the target object type and the personalized data of the perception type is searched.

[0116] In specific applications, when a target object receives information, it can provide a variety of feedback or no feedback. This feedback can be implemented through the execution of state data, significantly enhancing the personalization of the target object. In the above scenario, the information received by the target object is personalized data of the perception type. The receiving mode in this scenario can specifically be at least one of the perception modes, such as hearing, seeing, smelling, and touching. Furthermore, the target object type in the personalized data of the perception type is determined, which can specifically include whether "I," i.e., the target object, is the subject, object, or bystander of the information. Feedback is then provided according to the script in the information configuration table. The information configuration table represents a pre-established correspondence between object type, personalized data, and feedback-type state data, while the script represents the feedback-type state data. For example, if target object A robs target object B in an alley, when target object A threatens target object B with a knife, this state data generates a message: ID = Robbery, Subject Target Object A, Object Target Object B, Time X, Location Y. This message associates target objects A and B and represents the perception type (visual). At this point, target C passes by and sees target A. Now, targets A, B, and C all receive the same message. For target A, A determines that it is the subject and searches the configuration table for the subject response script, but finds it empty, so it continues the previous knife threat interaction. For target B, B determines that it is the object and searches the script for three possible responses: resist, flee, or pay. It chooses one based on its personality. For target C, C determines that it is a bystander and searches the script for three possible responses: ignore, call the police, or bravely help. It chooses one based on its personality. For complex, continuous interactions involving cause and effect, independent modules must be implemented for context management to ensure that the NPC determines appropriate state data. Let's use the robbery example mentioned above again. If victim B chooses to resist robber A and then loses, in real life, he or she would likely pay money and beg for mercy. However, if there is no module in the game that "tells" B that A is attacking him or her for money, then B will not be able to generate the state data of paying money. This requires event context information, which can be thought of as a "script." These scripts are independent of each other, and each target object can correspond to a "script," a form of personalized data. For example, the target object "victim" corresponds to the robbery context information, while the target object "robber" corresponds to the robbery context management module.

[0117] In an optional embodiment, the target source type includes: a target type associated with a specified target of the target object;

[0118] Accordingly, for each target source type, the target state determination method corresponding to the target source type is used to process the personalized data of the target object with the target source type to obtain the state data of the target source type, which may specifically include the following steps:

[0119] Get the execution order of multiple sub-goals in the specified goal;

[0120] According to the execution order, the sub-goals whose target objects have not been completed and whose execution order meets the preset priority conditions are determined from the multiple sub-goals;

[0121] From the correspondence between the personalized data, the sub-goals and the target-type state data, the target-type state data corresponding to both the personalized data of the target type and the determined sub-goals is searched.

[0122] In a specific application, if the designated goal of a target object is to become a general, then this designated goal can be broken down into multiple sub-goals to be executed in sequence, such as joining the army, achieving meritorious service, becoming a centurion, suffering a defeat in battle, becoming a centurion, and becoming a thousand-man commander. On this basis, in order to be more humane and personalized, when selecting the target-type state data to be executed by the target object, preset priority conditions can be used, combined with the execution order, to reduce the rigidity of the target object in achieving the above-mentioned designated goal. For example, in the process of becoming a general, one may encounter setbacks such as defeat in battle, punishment, demotion, and capture. Therefore, if the sub-goals that the target object has not completed include: suffering a defeat in battle, becoming a centurion, and becoming a thousand-man commander, and the preset priority condition is that a defeat must have occurred before becoming a centurion, then the target-type state data is data used to achieve the state of suffering a defeat.

[0123] In an optional implementation, the above-mentioned individual analysis of each candidate state data based on the personalized data to obtain the state strength of the candidate state data may specifically include the following steps:

[0124] Determine the target execution type for each candidate state data;

[0125] According to the target execution type of each candidate state data, from the pre-established correspondence between execution type and success rate model, a success rate model corresponding to the target execution type is searched and determined as the success rate model of the candidate state data;

[0126] For each candidate state data, determining the success rate of the candidate state data using the success rate model of the candidate state data according to the personalized data corresponding to the candidate state data;

[0127] The state strength of each candidate state data is determined based on the success rate of the candidate state data.

[0128] The target execution type of each candidate state data can be categorized based on the differences in the target object's candidate state data processing objects. Specifically, it can include: object type, other type, and probability type. The object type refers to the processing performed on the target object under the candidate state data; the other type refers to the processing performed by the target object on another target object under the candidate state data; and the probability type refers to the state achieved by the target object with a certain probability under the candidate state data.

[0129] Specifically, the type of thing can be calculated as follows: the analysis of things only needs to consider likes or dislikes, without adding a layer of complex operations of interpersonal relationships. Therefore, a universal success rate function can be established for the analysis of things, and encapsulated into a unified function name. Doing so can decouple external calls and internal algorithms, facilitate modification, and operate without interfering with each other. For example, the success rate function = success rate (A, T), which is used to obtain the value of the processing object T of the state data in the mind of the target object A. The following lists the algorithms corresponding to some types of processing objects T:

[0130] Success rate (target object A, life) = target object A's desire for survival × target object A's strength of need for life; success rate (target object A, food) = target object A's appetite × target object A's strength of need for food; success rate (target object A, sleep) = target object A's strength of need for sleep; success rate (target object A, item) = (target object A's desire for wealth × item price) ÷ target object A's total asset value. Based on this success rate function, we can establish success rates for good / evil labels and link them to the analysis of resistance and courageous behavior data: Good / evil success rate (A, T) = success rate (A, target object's object type label) × target object's strength of good / evil; resistance success rate (target object A, target object's good / evil label) = target object A's desire to resist × good / evil success rate (A, T); courageous behavior success rate (target object A, target object's good / evil label) = target object A's sense of justice × good / evil success rate (A, T).

[0131] The personality type of another person can be calculated using three main methods: the inference function, the favor-revenge function, and the interpersonal relationship function. The former functions serve as the basis for the latter. Specifically, the inference function represents the inference that target subject A makes about target subject B's personality. The accuracy of this inference is determined by the emotional intelligence (EQ) data. An EQ of 100 indicates a completely accurate guess, while an EQ of 0 indicates a completely biased judgment. Therefore, for example, the inference function = A.Inference(BX) = [(100 - EQ of target subject A) × A.X + EQ of target subject A × BX] / 100, representing target subject A's inference of target subject B's personality X. The inference function = A.Inference(F(B,T)) = ((100 - A.EQ) × F(A,T) + A.EQ × F(B,T)) / 100, representing target subject A's inference of the F-type parsing of target subject B on processing subject T. F-type parsing refers to target subject B processing T's state data according to the F-type. Inference function = A.Inference(F(B,A)) = ((100-A.EQ)×F(A,B)+A.EQ×F(B,A)) / 100, which represents the value of target object A's inference of target object B's F-type analysis of target object A.

[0132] The "enmity function" converts interaction results, or state data, into interpersonal data and incorporates it into the target object's memory: historical records. This memory can serve as personalized data of a perceptual type. When target object A causes a consequence X on target object B, the interpersonal function can be used to calculate the impact of this consequence on the interpersonal data between target object A and target object B. Exemplary interpersonal functions may include functions for calculating appreciation and dislike: Appreciation (target object A, female target) = (female target's endurance - 50) / 400 + (female target's intelligence - 50) / 400 + (female target's emotional quotient - 50) / 100 + (female target's appearance - 50) / 200; Appreciation (target object A, male target) = (male target's strength - 50) / 200 + (male target's endurance - 50) / 400 + (target object's intelligence - 50) / 200 + (male target's emotional quotient - 50) / 200 + (male target's appearance - 50) / 400. The love-hate function used to calculate care and hate values considers the gains and losses of the target to be proportionally treated as the target's own gains and losses. Thus, a value of 1 indicates that the target is treated as equal to the target itself; a value greater than 1 indicates a high probability of sacrificing themselves to save the target's life; and a value of -1 indicates extreme hatred, with the possibility of mutual destruction. The love-hate function (target object A, target) = target object A's kindness / evilness + target object A's discrimination × success rate (target object A, target) + probability of target object A receiving a favor (target) × target object A's desire to repay a favor - probability of target object A being harmed (target) × target object A's desire for revenge. The pressure function used to calculate coercion and fear values = coercion (target object A, target) = target object A's harm (target) × target object A's estimated vulnerability (target) - target object A's harm (target) × target object A's vulnerability.

[0133] For probability types, probability calculations are uniformly based on the [0, 1] range. Specifically, these can include preset probabilities and probabilities calculated based on personalized data corresponding to the candidate state data. For example, if the candidate state data is escape, and the corresponding personalized data is agility and the target's vehicle type, the probability can be calculated as agility multiplied by the preset probability corresponding to the vehicle type.

[0134] In an optional implementation, determining the state strength of each candidate state data based on the success rate of the candidate state data may specifically include the following steps:

[0135] Determining at least one of a basic strength, a success weight, and a failure weight of each candidate state data according to the personalized data corresponding to the candidate state data;

[0136] Determine the failure rate of each candidate state data according to the success rate of the candidate state data;

[0137] For each candidate state data, the success rate and failure rate of the candidate state data are processed using at least one of the basic strength, success weight, and failure weight of the candidate state data to obtain the state strength of the candidate state data.

[0138] For example, Figure 3 In an object processing method provided by one embodiment of the present application, an example diagram of the determination method of state strength is shown as follows: state strength = inevitable result + success rate × expected result of success + (1-success rate) × expected result of failure. Among them, the inevitable result is equivalent to the basic strength, the expected result of success is equivalent to the success weight, and the expected result of failure is equivalent to the failure weight. For example, the candidate state data includes option 1: resistance, and option 4: call for help. Among them, option 1: the state strength of resistance = combat success rate (A, B) + resistance (A, beating) - fear (A, B) + (1-combat success rate (A, B)) × (-valuation (A, serious injury) - valuation (A, C)). Among them, the combat success rate (A, B) is the success rate, and since there is no expected result of success, there is no success weight, so there is no need to weight the success rate. Resistance (A, beating) - fear (A, B) represents the inevitable result, which is equivalent to the basic strength. The expected result of failure, that is, the failure weight is -valuation (A, serious injury) - valuation (A, C). Here, A represents the target object, and B and C both represent objects that interact with target object A through candidate state data, such as resistance. For example, B could be a robber and C could be the looted property. Similarly, the state strengths for Option 2: Escape, Option 3: Surrender, and Option 4: Call for Help can be obtained, respectively. The difference lies in the specific base strength, success weight, and failure weight determination methods.

[0139] In an optional embodiment, the above-mentioned determination of target state data from a plurality of candidate state data according to the strength of each state may specifically include the following steps:

[0140] Get the current state data of the target object;

[0141] From the current state data and a plurality of candidate state data, first state data whose state strength reaches a first preset threshold is determined as target state data.

[0142] In specific applications, the first preset threshold can be set according to specific needs. In order to reduce state realization conflicts and further improve state accuracy, when there are multiple first state data, the first state data with the largest state intensity can be selected as the target state data.

[0143] In an optional embodiment, determining, from the current state data and a plurality of candidate state data, first state data whose state strength reaches a first preset threshold as the target state data may specifically include the following steps:

[0144] Determine whether the state types of the first state data and the current state data are the same; if different, use the first state data as the target state data when the first state data reaches a second preset threshold.

[0145] Among them, the second preset threshold can specifically be the sum of the state strength of the current state data and the hesitation. The hesitation is used to characterize the tendency of the target object between the two state data. If a decision is currently being executed, it is also necessary to determine whether to interrupt the current decision and switch to executing a new decision. In order to prevent jitter when switching back and forth between two decisions of similar strength, a protection mechanism is set: if the source of the decision with the highest strength is different from the source of the current decision, then the second preset threshold must be reached before the current decision is switched. Reaching the second preset threshold is also a kind of personalized data of the target object, which can be called hesitation.

[0146] The following combined Figure 4 , further explain the object processing method. Figure 4 A processing flow chart of another object processing method provided in an embodiment of the present application is shown, which specifically includes the following steps:

[0147] Initialize archive reading; wait for source update; when a source update event occurs, get the state data {Xi} of all sources; end if not obtained; if at least one is obtained, determine the first state data; determine whether the current state data is empty; if empty, determine the target state data based on the first state data; execute the target state data for the target object; if not empty, look for the first state data of the same state type as the current state data; if found, determine whether the state strength of the second state data reaches the second preset threshold; if reached, determine the target state data based on the second state data; determine whether the target state data is the same as the current state data, if the same, maintain the execution of the current state data; if different, interrupt the execution of the current state data, and switch to executing the target state data for the target object; if not found, determine the target state data based on the first state data, and execute to determine whether the target state data is the same as the current state data.

[0148] In specific applications, based on the first state data, determining the target state data may include: using the first state data as the target state data, or, when there are multiple first state data, using the first state data with the largest state strength as the target state data. Among them, if the current state data is X0, then using the first state data, or the first state data with the largest state strength as the target state data may include: X0=X, X∈{Xi}, that is, using the first state data as the value of the current state data. In addition, the state type of the state data can be specifically divided according to the source type of personalized data, such as demand-type state data, target-type state data, etc., see the above for details. Figure 1 In an alternative embodiment, the state data may carry a state type tag, such as a source serial number. Accordingly, searching for first state data having the same state type as the current state data may include determining, from {Xi}, first state data Xn having the same source serial number as the current state data.

[0149] And, exemplarily, the state strength of the first state data is XB, and the second preset threshold value = the state strength XA of the current state data + hesitation. Therefore, XB>the second preset threshold value, indicating that the target object's willingness to execute the first state data exceeds the execution strength of the current state data, and there is no hesitation, then when the first state data and the current state data are different, the execution of the current state data can be interrupted, and the execution can be switched to the target state data determined based on the first state data. Among them, the second state data is the first state data of the same state type as the current state data. The method of determining the target state data based on the second state data is similar to the method of determining the target state data based on the first state data. The difference is that the state data is the second state data. The same content will not be repeated here. For details, see the description of the method of determining the target state data based on the first state data.

[0150] Furthermore, this application standardizes and abstracts the data related to object state implementation using the target object's personalized data, state data, and state strength. This ensures that this application can be applied to implement state implementation for different objects and a large number of objects, eliminating the need for specialized state implementation development for each object. Therefore, this solution can reduce development time for different objects and a large number of objects, improving development efficiency.

[0151] This embodiment can further improve the accuracy and personalization of the target object's status by determining whether the status type and status data are the same.

[0152] Corresponding to the above method embodiment, the present application also provides an object processing device embodiment, Figure 5 FIG. 1 shows a schematic diagram of the structure of an object processing device provided by an embodiment of the present application. Figure 5 As shown, the device includes:

[0153] The candidate state data determination module 502 is configured to obtain personalized data of a target object; and determine a plurality of candidate state data of the target object based on the personalized data;

[0154] The state strength determination module 504 is configured to analyze each candidate state data based on the personalized data to obtain the state strength of the candidate state data;

[0155] The target state data determination module 506 is configured to determine target state data from the plurality of candidate state data according to each state strength, and execute the target state data for the target object.

[0156] In the solution provided by this application, the personalized data of a target object can represent the target object's own characteristics. Furthermore, the state strength of the candidate state data obtained based on the target object's personalized data can reflect the differences in the probability of the target object executing different state data, depending on the target object's own characteristics. Therefore, determining the target state data based on the state strength ensures that the state achieved by executing the target state data is consistent with the target object's own characteristics. Therefore, a personalized state that better suits the target object's own characteristics can be achieved for the target object. Furthermore, the implementation of this solution does not rely on user input, but is instead based on the target object's personalized data, equivalent to the target object autonomously achieving its state. Therefore, the autonomy of object state realization can be improved. Furthermore, this application uses the target object's personalized data, state data, and state strength to standardize, abstract, and unify the data related to achieving the object's state. This ensures that this application can be applied to achieving state realization for different objects and a large number of objects, eliminating the need for specialized state realization development for each object. Therefore, this solution can reduce development time for different objects and a large number of objects, thereby improving development efficiency.

[0157] In an optional implementation, the number of the personalized data is multiple;

[0158] The candidate state data determination module 502 is further configured to:

[0159] Determine the target source type of each personalized data of the target object respectively, and for each target source type, find the target state determination method corresponding to the target source type from the pre-established correspondence between the source type and the state determination method;

[0160] For each target source type, using the target state determination method corresponding to the target source type, the personalized data of the target object with the target source type is processed to obtain state data of the target source type;

[0161] The obtained state data of each target source type is determined as a plurality of candidate state data of the target object.

[0162] In an optional embodiment, the target source type includes: a demand type that characterizes the demand of the target object;

[0163] The candidate state data determination module 502 is further configured to:

[0164] Searching for the demand-type status data corresponding to the personalized data of the target object from the pre-established correspondence between the personalized data and the demand-type status data;

[0165] Based on the searched requirement-type status data, status data of the requirement type is determined.

[0166] In an optional embodiment, the number of personalized data with the required type is multiple;

[0167] The candidate state data determination module 502 is further configured to:

[0168] Obtaining the demand intensity of each demand corresponding to each personalized data of the demand type, and determining target personalized data whose demand intensity reaches a first intensity threshold from the personalized data of the demand type;

[0169] The demand-type state data corresponding to the target personalized data is searched from the pre-established correspondence between personalized data and demand-type state data.

[0170] In an optional implementation, the candidate state data determination module 502 is further configured to:

[0171] Determining, based on the target personalized data, the state strength of the demand-type state data corresponding to the target personalized data;

[0172] The state intensities corresponding to the searched demand-type state data are used as weights, and a preset weight random model is used to determine the state data of the demand type from the searched demand-type state data.

[0173] In an optional embodiment, the target source type includes: a perception type;

[0174] The candidate state data determination module 502 is further configured to:

[0175] determining a target object type of the target object in the personalized data of the perception type;

[0176] From the pre-established correspondence between object types, personalized data and feedback status data, feedback status data corresponding to both the target object type and the personalized data of the perception type is searched.

[0177] In an optional embodiment, the target source type includes: a target type associated with a specified target of the target object;

[0178] The candidate state data determination module 502 is further configured to:

[0179] Obtaining the execution order of multiple sub-goals in the specified goal;

[0180] According to the execution order, determining, from the multiple sub-goals, the sub-goals whose target objects have not been completed and whose execution order meets the preset priority condition;

[0181] From the correspondence between the personalized data, the sub-goals and the target-type state data, the target-type state data corresponding to both the personalized data of the target type and the determined sub-goals is searched.

[0182] In an optional implementation, the state strength determination module 504 is further configured to:

[0183] Determine the target execution type for each candidate state data;

[0184] According to the target execution type of each candidate state data, from the pre-established correspondence between execution type and success rate model, a success rate model corresponding to the target execution type is searched and determined as the success rate model of the candidate state data;

[0185] For each candidate state data, determining the success rate of the candidate state data using the success rate model of the candidate state data according to the personalized data corresponding to the candidate state data;

[0186] The state strength of each candidate state data is determined based on the success rate of the candidate state data.

[0187] In an optional implementation, the state strength determination module 504 is further configured to:

[0188] Determining at least one of a basic strength, a success weight, and a failure weight of each candidate state data according to the personalized data corresponding to the candidate state data;

[0189] Determine the failure rate of each candidate state data according to the success rate of the candidate state data;

[0190] For each candidate state data, the success rate and failure rate of the candidate state data are processed using at least one of the basic strength, the success weight, and the failure weight of the candidate state data to obtain the state strength of the candidate state data.

[0191] In an optional implementation, the state strength determination module 504 is further configured to:

[0192] Obtaining current status data of the target object;

[0193] From the current state data and the plurality of candidate state data, first state data whose state strength reaches a first preset threshold is determined as target state data.

[0194] In an optional implementation, the state strength determination module 504 is further configured to:

[0195] Determine whether the state type of the first state data is the same as that of the current state data; if different, use the first state data as target state data when the first state data reaches a second preset threshold.

[0196] The above is a schematic scheme of an object processing device of this embodiment. It should be noted that the technical solution of the object processing device and the technical solution of the object processing method mentioned above belong to the same concept. For details not described in detail in the technical solution of the object processing device, please refer to the description of the technical solution of the object processing method mentioned above. In addition, the various components in the device embodiment should be understood as functional modules that must be established to implement each step of the program flow or each step of the method, and each functional module is not an actual functional division or separation definition. The device claim defined by such a group of functional modules should be understood as a functional module architecture that mainly implements the solution through the computer program recorded in the specification, and should not be understood as a physical device that mainly implements the solution through hardware.

[0197] Figure 6 6 shows a block diagram of a computing device 600 according to an embodiment of the present application. Components of the computing device 600 include, but are not limited to, a memory 610 and a processor 620. The processor 620 is connected to the memory 610 via a bus 630, and a database 650 is used to store data.

[0198] The computing device 600 also includes an access device 640 that enables the computing device 600 to communicate via one or more networks 660. Examples of such networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 640 may include one or more of any type of network interface (e.g., a network interface card (NIC)) whether wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like.

[0199] In one embodiment of the present application, the above components of the computing device 600 and Figure 6 Other components not shown in the figure may also be connected to each other, for example, via a bus. Figure 6 The computing device structure block diagram shown is for illustrative purposes only and is not intended to limit the scope of the present application. Those skilled in the art may add or replace other components as needed.

[0200] The computing device 600 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook computer, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smartwatch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or PC. The computing device 600 may also be a mobile or stationary server.

[0201] The processor 620 is configured to execute computer executable instructions of the object processing method.

[0202] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the object processing method described above are based on the same concept. For details not described in detail in the technical scheme of the computing device, please refer to the description of the technical scheme of the object processing method described above.

[0203] An embodiment of the present application further provides a computer-readable storage medium storing computer instructions, which are used in an object processing method when executed by a processor.

[0204] The above is a schematic diagram of a computer-readable storage medium according to this embodiment. It should be noted that the technical solution of the storage medium and the technical solution of the object processing method described above are based on the same concept. For details not described in detail in the technical solution of the storage medium, please refer to the description of the technical solution of the object processing method described above.

[0205] The foregoing description describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0206] The computer instructions include computer program code, which may be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium may include any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunications signal, and a software distribution medium.

[0207] It should be noted that for the aforementioned method embodiments, for ease of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0208] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0209] The preferred embodiments of the present application disclosed above are intended only to help illustrate the present application. The optional embodiments do not describe all details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made based on the content of this application. This application selects and describes these embodiments in detail in order to better explain the principles and practical applications of this application, so that those skilled in the art can better understand and utilize this application. This application is limited only by the claims and their full scope and equivalents.

Claims

1. An object processing method, characterized in that: include: Obtaining personalized data of a target object, and classifying the target source type of the personalized data according to source differences of the personalized data, wherein the personalized data includes at least one of demand data, emotion data, temperament data, memory data, and interpersonal relationship data, the number of the personalized data is multiple, and the target object includes a virtual star, a virtual pet, a virtual friend, and a non-player target object in a game; Determining, based on the personalized data, a plurality of candidate state data of the target object, wherein determining, based on the personalized data, the plurality of candidate state data of the target object comprises: respectively determining a target source type of each personalized data of the target object, and for each target source type, searching, from a pre-established correspondence between source types and state determination methods, for a target state determination method corresponding to the target source type, wherein the target source type comprises at least one of a time type, a demand type, a perception type, a target type, and a task type; for each target source type, processing, using the target state determination method corresponding to the target source type, the personalized data of the target object having the target source type to obtain state data of the target source type; and determining the obtained state data of each target source type as the plurality of candidate state data of the target object, wherein, for each target source type, processing, using the target state determination method corresponding to the target source type, the personalized data of the target object having the target source type to obtain state data of the target source type comprises: searching, from a pre-established correspondence between personalized data and state data of the target source type, for state data of the target source type corresponding to the personalized data; Based on the personalized data, each candidate state data is parsed to obtain the state strength of the candidate state data, including: inputting the personalized data corresponding to each candidate state data into a pre-trained state strength model to obtain the state strength of the candidate state data, wherein the state strength model is a neural network model trained using sample personalized data and state strength labels of sample state data corresponding to the sample personalized data; According to each state strength, target state data is determined from the plurality of candidate state data, and the target state data is executed for the target object.

2. The method according to claim 1, characterized in that The demand type includes: a demand type that characterizes the demand of the target object; The process of processing the personalized data of the target object having the target source type using the target state determination method corresponding to the target source type for each target source type to obtain state data of the target source type includes: Searching for the demand-type status data corresponding to the personalized data of the target object from the pre-established correspondence between the personalized data and the demand-type status data; Based on the searched requirement-type status data, status data of the requirement type is determined.

3. The method according to claim 2, characterized in that The number of personalized data with the required type is multiple; The searching for the demand-type status data corresponding to the personalized data of the target object from the pre-established correspondence between the personalized data and the demand-type status data includes: Obtaining the demand intensity of each demand corresponding to each personalized data of the demand type, and determining target personalized data whose demand intensity reaches a first intensity threshold from the personalized data of the demand type; The demand-type state data corresponding to the target personalized data is searched from the pre-established correspondence between personalized data and demand-type state data.

4. The method according to claim 3, characterized in that The step of determining the state data of the demand type based on the searched demand type state data includes: Determining, based on the target personalized data, the state strength of the demand-type state data corresponding to the target personalized data; The state intensities corresponding to the searched demand-type state data are used as weights, and a preset weight random model is used to determine the state data of the demand type from the searched demand-type state data.

5. The method according to claim 1, wherein for each target source type, using a target state determination method corresponding to the target source type, processing the personalized data of the target object having the target source type to obtain state data of the target source type comprises: determining a target object type of the target object in the personalized data of the perception type; From the pre-established correspondence between object types, personalized data and feedback status data, feedback status data corresponding to both the target object type and the personalized data of the perception type is searched.

6. The method according to claim 1, characterized in that The target type includes: a target type associated with a specified target of a target object; The process of processing the personalized data of the target object having the target source type using the target state determination method corresponding to the target source type for each target source type to obtain state data of the target source type includes: Obtaining the execution order of multiple sub-goals in the specified goal; According to the execution order, determining, from the multiple sub-goals, the sub-goals whose target objects have not been completed and whose execution order meets the preset priority condition; From the correspondence between the personalized data, the sub-goals and the target-type state data, the target-type state data corresponding to both the personalized data of the target type and the determined sub-goals is searched.

7. The method according to claim 1, characterized in that The step of respectively parsing each candidate state data based on the personalized data to obtain the state strength of the candidate state data includes: Determine the target execution type for each candidate state data; According to the target execution type of each candidate state data, from the pre-established correspondence between execution type and success rate model, a success rate model corresponding to the target execution type is searched and determined as the success rate model of the candidate state data; For each candidate state data, determining the success rate of the candidate state data using the success rate model of the candidate state data according to the personalized data corresponding to the candidate state data; The state strength of each candidate state data is determined based on the success rate of the candidate state data.

8. The method according to claim 7, characterized in that The determining the state strength of each candidate state data based on the success rate of the candidate state data includes: Determining at least one of a basic strength, a success weight, and a failure weight of each candidate state data according to the personalized data corresponding to the candidate state data; Determine the failure rate of each candidate state data according to the success rate of the candidate state data; For each candidate state data, the success rate and failure rate of the candidate state data are processed using at least one of the basic strength, the success weight, and the failure weight of the candidate state data to obtain the state strength of the candidate state data.

9. The method according to claim 1, characterized in that The step of determining target state data from the plurality of candidate state data according to the strength of each state includes: Obtaining current status data of the target object; From the current state data and the plurality of candidate state data, first state data whose state strength reaches a first preset threshold is determined as target state data.

10. The method according to claim 9, characterized in that The step of determining, from the current state data and the plurality of candidate state data, first state data whose state intensity reaches a first preset threshold as target state data includes: Determine whether the state type of the first state data is the same as that of the current state data; if different, use the first state data as target state data when the first state data reaches a second preset threshold.

11. An object processing device, characterized in that: include: The candidate state data determination module is configured to obtain personalized data of a target object, and classify target source types of the personalized data according to source differences of the personalized data, wherein the personalized data includes at least one of demand data, emotional data, temperament data, memory data, and interpersonal relationship data, the number of the personalized data is multiple, and the target object includes a virtual star, a virtual pet, a virtual friend, and a non-player target object in a game; determine multiple candidate state data of the target object based on the personalized data, and the determining multiple candidate state data of the target object based on the personalized data includes: determining the target source type of each personalized data of the target object respectively, and for each target source type, searching for a target state determination method corresponding to the target source type from a pre-established correspondence between the source type and the state determination method, wherein The target source type includes at least one of a time type, a demand type, a perception type, a target type, and a task type; for each target source type, using a target state determination method corresponding to the target source type, processing the personalized data of the target object having the target source type to obtain state data of the target source type; determining the obtained state data of each target source type as a plurality of candidate state data of the target object, wherein, for each target source type, using a target state determination method corresponding to the target source type, processing the personalized data of the target object having the target source type to obtain state data of the target source type includes: searching for state data of the target source type corresponding to the personalized data from a pre-established correspondence between personalized data and state data of the target source type; The state strength determination module is configured to analyze each candidate state data based on the personalized data to obtain the state strength of the candidate state data, including: inputting the personalized data corresponding to each candidate state data into a pre-trained state strength model to obtain the state strength of the candidate state data, wherein the state strength model is a neural network model trained using sample personalized data and state strength labels of sample state data corresponding to the sample personalized data; The target state data determination module is configured to determine target state data from the plurality of candidate state data according to each state strength, and execute the target state data for the target object.

12. A computing device, characterized in that include: memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the steps of the object processing method according to any one of claims 1 to 10.

13. A computer-readable storage medium storing computer-executable instructions, characterized in that: When the instruction is executed by a processor, the steps of the object processing method according to any one of claims 1 to 10 are implemented.

Images