User preference recommendation method, device, medium and computer program product

Abstract

The invention relates to the technical field of intelligent recommendation, and particularly provides a user preference recommendation method and device, a medium and a computer program product. The user preference recommendation method comprises the steps of obtaining user feature information of a target user and virtual feature information of a virtual character associated with the target user; respectively carrying out clustering calculation on the user feature information and the virtual feature information, and determining a real user group and a virtual character group to which the target user belongs; determining common users between the real user group and the virtual character group, and forming a target group based on the common users; the number of the common users is greater than or equal to two; and generating a user preference recommendation result of the target user based on the preference information corresponding to each common user in the target group. According to the method, the feature information of the virtual character is introduced, and double-path grouping recommendation is performed based on the virtual feature information and the user feature information, so that the accuracy of the recommended content is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of intelligent recommendation technology, and in particular to a method, device, medium and computer program product for recommending user preferences. Background Art

[0002] Nowadays, intelligent recommendation algorithms play an important role in Internet applications. They use artificial intelligence technology to conduct in-depth analysis of user behaviors, interests or needs, and accurately recommend matching content or services. Through intelligent recommendation, users can find the content they are interested in faster, saving a lot of search and screening time. It can not only meet the differentiated needs of different users, but also improve the pertinence and effectiveness of recommendation services.

[0003] In the related art, intelligent recommendation methods often analyze the correlation features between users and content, determine the user's preference information based on the correlation features, and then determine the content that the user is interested in. However, content-based recommendation methods often can only rely on the user's historical behavior and preference data for recommendation, resulting in a small range of recommended content, high repetition and serious monotony, and unable to provide diversified user preference recommendation results. Summary of the invention

[0004] In view of this, exemplary embodiments of the present disclosure provide a user preference recommendation method, device, medium, and computer program product to solve the problems existing in the related art.

[0005] One aspect of the exemplary embodiments of the present disclosure provides a method for recommending user preferences, the method comprising:

[0006] Acquire user characteristic information of a target user and virtual characteristic information of a virtual character associated with the target user;

[0007] Performing cluster calculations on the user characteristic information and the virtual characteristic information respectively to determine the real user group and the virtual character group to which the target user belongs;

[0008] Determine common users between the real user group and the virtual character group, and form a target group based on the common users; the number of the common users is greater than or equal to two;

[0009] Based on the preference information corresponding to each common user in the target group, a user preference recommendation result of the target user is generated.

[0010] Another aspect of the exemplary embodiments of the present disclosure provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method described in the exemplary embodiments of the present disclosure.

[0011] According to another aspect of the exemplary embodiments of the present disclosure, a computer-readable storage medium is provided, on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the method described in the exemplary embodiments of the present disclosure is implemented.

[0012] According to another aspect of the exemplary embodiments of the present disclosure, a computer program product is provided, including a computer program / instruction. When the computer program / instruction is executed by a processor, the method described in the exemplary embodiments of the present disclosure is implemented.

[0013] As will be described in detail below, according to a user preference recommendation method of an embodiment of the present disclosure, the user feature information of the target user and the virtual feature information of the virtual character associated with the target user are obtained, and the feature information of the virtual character associated with the target user is introduced. The user feature information and the virtual feature information are clustered and calculated respectively to determine the real user group corresponding to the target user in the real domain and the virtual character group to which the target user belongs in the virtual domain. The common users between the real user group and the virtual character group are determined, and the target group is formed based on the common users, so as to determine the common users with similar feature information to the target user in the real domain and the virtual domain, and form the target group. Among them, the number of common users is greater than or equal to two. Finally, based on the preference information corresponding to each common user in the target group, the user preference recommendation result of the target user is generated. Therefore, the user preference recommendation method provided by the present disclosure can solve the technical problem that the recommendation results are inaccurate and simplistic due to the lack of user behavior data in virtual scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The above and other purposes, features and advantages of the present disclosure will become more apparent by describing the embodiments of the present disclosure in more detail in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0015] Figure 1 A flowchart of a user preference recommendation method provided as an example in the present disclosure;

[0016] Figure 2 A schematic diagram of a feature extraction stage provided as an example of the present disclosure;

[0017] Figure 3 A schematic diagram of a two-way group association phase provided as an example of the present disclosure;

[0018] Figure 4 This is a schematic diagram of the fusion recommendation stage provided as an example in the present disclosure;

[0019] Figure 5 A flowchart of a user preference recommendation method provided as an example in the present disclosure;

[0020] Figure 6 A schematic block diagram of functional modules of an exemplary device for recommending user preferences provided by the present disclosure;

[0021] Figure 7 A structural block diagram of an electronic device provided as an example in the present disclosure;

[0022] Figure 8 A schematic diagram of a computer program product provided by an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0024] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0025] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc. mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0026] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0027] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0028] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0029] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0030] As an optional but non-limiting implementation, in response to receiving an active request from the user, the method of sending a prompt message to the user may be, for example, a pop-up window, in which the prompt message may be presented in text. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device. It is understandable that the above notification and the process of obtaining user authorization are only illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0031] Before introducing the embodiments of the present disclosure, the following definitions are given for the relevant terms involved in the embodiments of the present disclosure:

[0032] Nowadays, intelligent recommendation algorithms play an important role in Internet applications. They use artificial intelligence technology to conduct in-depth analysis of user behaviors, interests or needs, and accurately recommend matching content or services. Through intelligent recommendation, users can find the content they are interested in faster, saving a lot of search and screening time. It can not only meet the differentiated needs of different users, but also improve the pertinence and effectiveness of recommendation services.

[0033] In the related art, intelligent recommendation methods often analyze the correlation features between users and content, determine the user's preference information based on the correlation features, and then determine the content that the user is interested in. However, content-based recommendation methods often can only rely on the user's historical behavior and preference data for recommendation, resulting in a small range of recommended content, high repetition and serious monotony, and unable to provide diversified user preference recommendation results.

[0034] Therefore, in order to solve the above problems, the exemplary embodiments of the present disclosure provide a method for recommending user preferences, which extracts feature information of the real domain and the meta-virtual domain respectively. Then, users of the real domain and the meta-virtual domain are grouped according to the user feature information to obtain real domain groups and meta-virtual domain groups. Finally, the real domain groups and meta-virtual domain groups are merged to recommend the preference information of other users in the same group to the user.

[0035] For example, Figure 1 A flowchart of a user preference recommendation method provided by an exemplary embodiment of the present disclosure is shown in FIG. Figure 1 As shown, the following steps may be specifically included:

[0036] Step S110: extracting feature information of the real user and the virtual character respectively.

[0037] In the embodiment, the features of the real user and the virtual character associated with the real user are extracted based on their respective domain dimensions, and the features are collected based on their respective spatial dimensions to obtain user feature information corresponding to the real user and virtual feature information corresponding to the virtual character.

[0038] Specifically, feature extraction is performed on real users to obtain user feature information, which may include user portrait features, terminal attribute features, and behavior attribute features.

[0039] Among them, user portrait features may include basic information such as the user's age, gender, occupation, interests and hobbies.

[0040] Terminal attribute characteristics may include the device type, operating system, and browser version used by the user. Terminal attribute characteristics can help the system understand the user's technical environment and usage habits.

[0041] Behavioral attribute characteristics may include purchasing or operation behavior characteristics. Behavioral attribute characteristics can analyze user needs and preferences by recording user shopping history, browsing history, click behavior and other data.

[0042] Extract features of virtual characters to obtain virtual feature information, which focuses on describing performance and behavior in the virtual environment.

[0043] Virtual feature information may include the style, dress, behavior, and duration of stay of the virtual character.

[0044] Among them, the style and dress of the virtual character can include the appearance, clothing and accessories of the virtual character, etc. These features can be extracted and classified through image processing or visual analysis technology.

[0045] The behavior of a virtual character may include the way the virtual character moves, interactive behaviors, or changes in facial expressions, etc. The behavior can be accurately recorded through animation capture and behavior recognition technology.

[0046] The length of time spent in the virtual space is an important indicator for measuring the activity of virtual characters, which can be obtained by tracking the length of time that virtual characters stay in different virtual scenes.

[0047] For example, Figure 2 This is a schematic diagram of the feature extraction stage provided as an example of the present disclosure, as shown in FIG. Figure 2 As shown, there are real users a1-a12 and virtual characters v1-v9.

[0048] Extract features of real users: Extract features of real users a1-a12 respectively to obtain user feature information, which may include: user portrait features, terminal attribute features, and behavior attribute features.

[0049] Extract features of virtual characters: Extract features of virtual characters v1-v9 respectively to obtain virtual feature information, which may include the style, dress, behavior and duration of stay of the virtual characters.

[0050] Based on this, by extracting and analyzing the features of real users and virtual characters associated with real users, a correlation relationship between the two can be established to provide data support for subsequent user grouping.

[0051] Step S120: performing two-way group association according to the user characteristic information and the virtual characteristic information.

[0052] Exemplarily, it is first necessary to perform clustering calculations on real users and virtual characters according to user feature information and virtual feature information, and divide them into different user groups.

[0053] Exemplarily, the clustering calculation method can use a deep neural network and combine it with community analysis of graph data for classification.

[0054] Specifically, they may include:

[0055] On the deep neural network side: first collect and organize user feature information and virtual feature information, and standardize these data to make them suitable for input into the input layer of the neural network.

[0056] Design and build a deep neural network model. The input layer of the model receives user feature information and virtual feature information. The middle layer contains several hidden layers, each of which consists of multiple neurons. The activation function can use nonlinear functions such as ReLU (Rectified Linear Unit). The number of nodes in the output layer depends on the number of categories that need to be classified.

[0057] Use the preprocessed training data to train the model. You can use methods such as the cross entropy loss function and the Adam optimizer to improve the classification accuracy of the model. During the training process, the model is evaluated and adjusted through the validation set to prevent overfitting.

[0058] After training is complete, the model is used to classify new users. Based on the classification results, users are assigned to different groups.

[0059] On the social analysis side of graph data: first, build a user graph, and construct users and their relationships into a graph structure, with users as nodes and user feature information as edges.

[0060] Then, a community detection algorithm (such as Louvain algorithm or Label Propagation algorithm) is used to identify communities in the graph. A community is a subgraph with densely connected nodes in the graph, indicating close relationships between users.

[0061] Finally, based on the detection results, users are divided into different communities, which represent the clustering results of users in the virtual domain and the real domain.

[0062] After obtaining the user groups output by the deep neural network model and the community results output by the community analysis of the graph data respectively, the user groups and the community results can be combined to form a more accurate clustering result.

[0063] For example, Figure 3 The schematic diagram of the two-way group association phase provided by the present disclosure is as follows: Figure 3 As shown, after obtaining the user feature information corresponding to the real user and the virtual feature information corresponding to the virtual character, execute S310: perform clustering calculations on the real user and the virtual character respectively, and divide the real users into real user groups AG-1 to AG-N respectively, and similarly divide the virtual characters into virtual character groups VG-1 to VG-N.

[0064] S320: Determine all real user groups AG-1 to AG-N as real user groups to be recommended.

[0065] S330: Determine all virtual character groups VG-1 to VG-N as virtual character groups to be merged.

[0066] S340: In order to realize the association conversion from virtual characters to real users, it is necessary to map according to the affiliation between real users and virtual characters. Taking the following Table 1 as an example, the specific operation of S340 is as follows: Assuming that real user a1 has virtual characters v1 and v20, when virtual character v1 needs to be converted, it can be mapped to the real user dimension of a1. According to this principle, v1, v2, and v3 in the virtual user group can be converted and mapped to obtain the real associated user group AV-1 containing these virtual characters, and AV-1 contains display users a1, a2, and a3.

[0067] Table 1 Affiliation between real users and associated virtual characters

[0068] Real users Related virtual characters a1 v1, v20 a2 v2, v22 a3 v3, v33 a4 v4, v44

[0069] S350: Obtain the reality-related user group AV-1.

[0070] Finally, the real user group to be recommended in S320 and the real associated user group in S350 are combined into a same space user group, that is, the same space user group includes the AG-1 group and the AV-1 group.

[0071] Based on this, this mapping and conversion process realizes the connection between the characteristics of virtual characters and the needs of real users. Through the combination of deep neural network models and clustering algorithms, users and virtual characters can be efficiently grouped and associated, providing a solid foundation for personalized recommendations. Ultimately, the system can recommend virtual characters that best suit the user's interests based on their preferences and behaviors, improving user experience and satisfaction.

[0072] Step S130: Perform user preference recommendations based on feature information and user grouping.

[0073] For example, user preference information can be determined through virtual-real fusion and collaborative recommendation. First, a similarity or intersection algorithm can be used to define virtual-real fusion domain rules, identify common elements of the virtual-real fusion domain, namely users, and obtain virtual-real fusion results.

[0074] Specifically, by comparing the groups of real users and virtual characters, using similarity or intersection algorithms, we can find out the common users in both groups. These common users constitute the core of the virtual-real fusion domain and form the virtual-real fusion result.

[0075] Then, the virtual-real fusion result is extracted according to the user dimension. The detailed preference characteristics of each user are extracted by analyzing the user feature information and virtual feature information. Then, the recommendation items for the user are determined by using the collaborative recommendation method.

[0076] For example, Figure 4This is a schematic diagram of the fusion recommendation stage provided by an exemplary embodiment of the present disclosure, such as Figure 4 As shown, it may specifically include:

[0077] First, merge the actual user group and the actual associated user group. For example, in the separated groups:

[0078] The real user group AG-1 includes real users a1, a3 and a4;

[0079] The reality-associated user group AV-1 includes real users a1, a2, and a3 corresponding to the virtual characters.

[0080] The intersection algorithm can be used to perform virtual-real fusion calculation in the virtual-real fusion domain. If it is found that the proportion of the same users between AG-1 and VG-1 is greater than the preset threshold, collaborative recommendation can be performed. If it is found that the proportion of the same users between AG-1 and VG-1 is less than the preset threshold, virtual-real fusion and subsequent collaborative recommendation will not be performed.

[0081] Based on this, the intersection algorithm plays a bridging role in virtual-real fusion computing, and realizes effective data fusion by determining the common elements of real user groups and virtual character groups. Subsequently, the system can conduct in-depth analysis and personalized recommendations based on these fused data, thereby improving user experience and satisfaction.

[0082] After obtaining the virtual-real fusion results, the common users in the virtual-real fusion domain are determined from the virtual-real fusion results: a1 and a3, and then a1 and a3 can be inferred based on different features in each dimension of virtual and real. Since the common users have stronger similarity in the two-dimensional grouping, the common users can make collaborative recommendations to each other based on their respective preferences.

[0083] The following uses shared users a1 and a3 as an example to describe how to use the collaborative recommendation algorithm to make cross-user recommendations based on the shared users' preference characteristics:

[0084] Assume that the preferences of user a1 include: preference 1, preference 2, and preference 3, and the preferences of user a3 include: preference 3, preference 5, and preference 6.

[0085] Then the user preference information recommendation result can be determined: recommend a3's preference 5 and preference 6 to user a1, and recommend a1's preference 1 and preference 2 to user a3, thereby completing the collaborative recommendation across users in the same group.

[0086] Similarly, the above-mentioned combination of virtual and real fusion and collaborative recommendation is also applicable to other shared users in the group.

[0087] Based on this, the fusion recommendation stage can effectively integrate user data in both virtual and real dimensions, and use collaborative recommendation to achieve accurate personalized recommendations. This recommendation method not only improves the accuracy and effectiveness of recommendations, but also enhances the consistency and satisfaction of users' experience in different environments.

[0088] Based on the above embodiment, in another embodiment provided by the present disclosure, a method for recommending user preferences for new users in the real domain may include:

[0089] First, the virtual character corresponding to the new user in the real domain is determined, and the feature information of the virtual character is extracted.

[0090] Since the user feature information of the new real user is not extracted, the new real user may not be included in any existing real user group. However, according to the virtual feature information extracted from the virtual character, the virtual character associated with the new real user may be included in the virtual character group.

[0091] Finally, new users in the real domain can be recommended based on the preference information of other users in the virtual character group.

[0092] For example, suppose a1 is a new user in the real domain. Although a1 has no corresponding real user group in the real domain, a1 is included in the virtual character group VG-1 in the virtual domain. In addition, VG-1 also includes real users a2 and a3. The preference data of other members in the virtual character group VG-1 (such as a2 and a3) can be used to recommend a1.

[0093] Based on this, even if there is a lack of sufficient initial data in the real domain and it is impossible to classify new users into a certain real user group, the feature information in the virtual domain can be used for collaborative recommendation.

[0094] One or more technical solutions provided in the exemplary embodiments of the present disclosure first extract feature information of real users and virtual characters respectively through the feature extraction stage to obtain user feature information and virtual feature information, thereby comprehensively capturing multi-dimensional information of users in real and virtual environments. Compared with relying solely on historical behavior data of users in reality for feature extraction, this method greatly enriches the user's feature information and makes subsequent recommendation results more accurate.

[0095] Then, in the two-way grouping association stage, the user feature information and virtual feature information are used to build and train the deep neural network model, and the user grouping is completed for real users and virtual characters respectively. Then, the association mapping between virtual characters and real users is performed to obtain the real associated user group divided by virtual feature information. Through the association mapping between virtual characters and real users, the corresponding relationship between user feature information and virtual feature information can be established, which expands the range of recommended options.

[0096] Finally, in the fusion recommendation stage, the real user group and the real associated user group are fused to obtain the virtual-real fusion result. Combined with the preference extraction and collaborative recommendation capabilities, the preference extraction and recommendation of the users in the group of the virtual-real fusion result are performed. Through this virtual-real fusion recommendation strategy, the real preferences of users can be captured more accurately, and more personalized recommendation services can be provided, further improving the diversity and novelty of recommendations.

[0097] Therefore, the user preference recommendation method provided in the exemplary embodiment of the present disclosure can effectively solve the problems existing in the traditional recommendation method, such as insufficient accuracy, narrow recommendation scope, high content duplication, and serious monotony.

[0098] Based on the above embodiments, the present disclosure also provides a method for recommending user preferences. Figure 5 This is a flow chart of a user preference recommendation method provided by an exemplary embodiment of the present disclosure, such as Figure 5 As shown, the method may include the following steps:

[0099] Step S510: Acquire user characteristic information of a target user and virtual characteristic information of a virtual character associated with the target user.

[0100] In the embodiment, user feature information of a target user in a real domain is obtained, and virtual feature information of a virtual character associated with the target user in a virtual space is obtained.

[0101] These feature information include users' behavior data and interests in real life and in virtual space. By collecting these multi-dimensional feature information, data support can be provided for subsequent user recommendations.

[0102] Step S520: performing cluster calculations on the user characteristic information and the virtual characteristic information respectively, and determining the real user group and the virtual character group to which the target user belongs.

[0103] In the embodiment, clustering calculations are performed on user feature information and virtual feature information respectively. Clustering of user feature information will determine the real user group to which the target user belongs, while clustering of virtual feature information will determine the virtual character group to which the target user belongs. Such a clustering process helps to discover the user's potential communities and interest circles in real and virtual environments.

[0104] Clustering algorithms may include K-means, hierarchical clustering, or density clustering.

[0105] Step S530: determining common users between the real user group and the virtual character group, and forming a target group based on the common users, where the number of common users is greater than or equal to two.

[0106] In the embodiment, by determining the common users between the two groups, users who are similar to the target user in real life and have intersections in the virtual environment can be identified. These common users form a target group, which represents a group of users with a high degree of similarity to the target user.

[0107] Step S540: generating a user preference recommendation result of the target user based on the preference information corresponding to each common user in the target group.

[0108] In the embodiment, based on the preference information corresponding to each common user in the target group, the user preference recommendation results of the target user are generated. By comprehensively analyzing the behavior data, interests and interaction records of the users in the target group, the system can extract common preferences and personalized needs. Using this information, the recommendation results generated by the recommendation system are more accurate and diversified, and can better meet the interests and needs of the target users.

[0109] Based on this, by obtaining the user feature information of the target user and the virtual feature information of the virtual character associated with the target user, the feature information of the virtual character associated with the target user is introduced. The user feature information and the virtual feature information are clustered and calculated respectively to determine the real user group corresponding to the target user in the real domain and the virtual character group corresponding to the virtual domain. Multiple common users between the real user group and the virtual character group are determined, and a target group is formed based on the multiple common users, so as to determine the common users with similar feature information to the target user in the real domain and the virtual domain, and form the target group. Finally, based on the preference information corresponding to each common user in the target group, a user preference recommendation result for the target user is generated. Therefore, the user preference recommendation method provided by the present disclosure can solve the technical problem that the recommendation results are inaccurate and simplistic due to the lack of user behavior data in virtual scenes.

[0110] Based on the above embodiment, in another embodiment provided by the present disclosure, the above user preference recommendation method may further include:

[0111] Obtaining affiliation information; the affiliation information represents the association relationship between the real user and the virtual character; based on the affiliation information, determining the first real user associated with each virtual character in the virtual character group.

[0112] In the embodiment, the affiliation information between the real user and the virtual character is obtained, and the affiliation information can be represented as the above Table 1.

[0113] According to the affiliation information, real users respectively associated with each virtual character in the virtual character group are determined, and the corresponding real users are determined as first real users.

[0114] Based on this, the first real user associated with each virtual character in the virtual character group can be determined based on the affiliation information, which can more accurately identify the user's preference information in the virtual space. This association relationship helps to reveal the user's preference information in different environments, thereby improving the accuracy of the recommendation system and making the recommendation results more in line with the user's real preferences.

[0115] Based on the above embodiment, in another embodiment provided by the present disclosure, the above determining a plurality of common users between the real user group and the virtual character group, and forming a target group based on the plurality of common users may include:

[0116] Obtaining a corresponding second real user in the real user group;

[0117] A plurality of common users between the first real user and the second real user are determined, and the plurality of common users are grouped into a target group.

[0118] In the embodiment, all real users in the real user group are obtained, and all real users are determined as second real users.

[0119] The intersection algorithm can be used to perform virtual-real fusion calculation on the first real user and the second real user to determine the common users between the two. If the proportion of common users is greater than a preset threshold, collaborative recommendation can be performed; if the proportion of common users is less than the preset threshold, virtual-real fusion and subsequent collaborative recommendation will not be performed.

[0120] When the proportion of common users is greater than a preset threshold, all common users are grouped into a target group.

[0121] Based on this, the intersection algorithm plays a bridging role in virtual-real fusion computing, and realizes effective data fusion by determining the common users of real user groups and virtual character groups. Subsequently, the system can conduct in-depth analysis and personalized recommendations based on these fused data, thereby improving user experience and satisfaction.

[0122] Based on the above embodiment, in another embodiment provided by the present disclosure, the above step S520 may include:

[0123] Obtaining a trained group classification model; the training process of the group classification model includes: using user feature training information and virtual feature training information as training samples, inputting the training samples into a pre-constructed group classification model, and obtaining user grouping prediction data; updating parameters of the group classification model based on the user grouping prediction data, and obtaining a trained group classification model;

[0124] Input user feature information into the group classification model to obtain the actual user group corresponding to the target user;

[0125] The virtual feature information is input into the group classification model to obtain the virtual character group corresponding to the target user.

[0126] In an embodiment, in order to determine the real user group and the virtual character group of the target user, the real user group and the virtual character group corresponding to the target user may be determined based on the user feature information, the virtual feature information and the trained group classification model.

[0127] Firstly, user feature training information and virtual feature training information are collected and sorted, and these data are standardized to make them suitable for input into the pre-constructed group classification model.

[0128] The input layer of the group classification model receives user feature training information and virtual feature training information. The middle layer contains several hidden layers, each of which consists of multiple neurons. The activation function can use nonlinear functions such as ReLU (Rectified Linear Unit). The number of nodes in the output layer depends on the number of categories that need to be classified.

[0129] Use the preprocessed training data to train the model. You can use methods such as the cross entropy loss function and the Adam optimizer to improve the classification accuracy of the model. During the training process, the model is evaluated and adjusted through the validation set to prevent overfitting.

[0130] After training is completed, the model is used to classify the target users. According to the classification results, the target users are assigned to the corresponding real user groups and virtual character groups.

[0131] Based on this, by acquiring and training the group classification model and inputting user feature information and virtual feature information into the model for prediction, accurate grouping of target users can be achieved. In addition, the group classification model can not only effectively group users, but also dynamically adapt to new user feature information and continuously optimize the recommendation effect.

[0132] Based on the above embodiment, in another embodiment provided by the present disclosure, the above step S520 may include:

[0133] A real user graph is constructed based on user feature information; the nodes in the real user graph represent real users, the edges in the real user graph represent the relationships between real users, and the weights of the edges are set based on the user feature information of the real users;

[0134] Constructing a virtual character graph based on the virtual feature information; the nodes in the virtual character graph represent the virtual characters, the edges in the virtual character graph represent the relationships between the virtual characters, and the weights of the edges are set based on the virtual feature information of the virtual characters;

[0135] Use community detection algorithms to process real user graphs, find other real users with similar user feature information, and form real user groups;

[0136] The community detection algorithm is used to process the virtual character graph to find other virtual characters with similar virtual feature information to form a virtual character group.

[0137] In an embodiment, in order to determine the real user group and virtual character group of the target user, a real user graph and a virtual character graph can be constructed based on the user feature information and the virtual feature information respectively, and processed using a community detection algorithm to form corresponding user groups and virtual character groups.

[0138] First, a real user graph is constructed based on user feature information. In the real user graph, each node represents a real user, and the edges between nodes represent the relationship between real users. In order to better reflect the similarity between users, the weights of these edges are set according to the feature information of real users. For example, if two users have high similarity in interests, behaviors, and other features, the weight of the edge connecting them will be higher.

[0139] Similarly, an avatar graph is constructed based on the avatar feature information. In the avatar graph, each node represents an avatar, and the edges between nodes represent the relationship between avatars. The weight of the edge is set based on the feature information of the avatar. For example, if two avatars are similar in attributes, behavior patterns, or other characteristics, the weight of the edge connecting them will also be high.

[0140] Next, a community detection algorithm may be used to process the real user graph to find other real users with similar user feature information to form a real user group.

[0141] Exemplarily, the community detection algorithm may adopt the Louvain algorithm or the Label Propagation algorithm, which can effectively discover the close relationships and community structures between users by identifying densely connected subgraphs in the graph, and then determine the actual user group of the target user.

[0142] Similarly, a community detection algorithm is used to process the virtual character graph to find other virtual characters with similar virtual feature information to form a virtual character group. The virtual character group reflects the close relationship and similarity between characters in the virtual environment, and can provide rich virtual character data support for the recommendation system.

[0143] Based on this, by constructing a real user graph that reflects the relationship between real users and a virtual character graph that reflects the relationship between virtual characters, and using a community detection algorithm to identify the corresponding real user groups and virtual character groups, not only the accuracy and diversity of the recommendation results are improved, but it can also dynamically adapt to changes in user behavior and continuously optimize the recommendation effect.

[0144] Based on the above embodiment, in another embodiment provided by the present disclosure, the above step S540 may include:

[0145] Obtain preference information corresponding to each common user in the target group to form initial preference information;

[0146] Obtaining first preference information corresponding to the target user;

[0147] Based on the initial preference information and the first preference information, a user preference recommendation result of the target user is generated.

[0148] In the embodiment, the preference information corresponding to each common user in the target group is obtained, and the information is integrated into the initial preference information. By analyzing the preference information of the users in the target group, the common preferences in the group are understood.

[0149] Then, the preference information corresponding to the target user is obtained, and the preference information corresponding to the target user is determined as the first preference information.

[0150] Furthermore, the content in the initial preference information that is repeated with the first preference information is removed to obtain the preference information after the removal, and the preference information after the removal is determined as the user preference recommendation result of the target user.

[0151] Based on this, this recommendation method based on the combination of group and personal preference information can make full use of the common preferences within the group and the personalized preferences of the target users to generate more accurate and rich recommendation results.

[0152] Based on the above embodiment, in another embodiment provided by the present disclosure, the above step S510 may include:

[0153] Obtaining user portrait information of the target user, and performing feature extraction on the user portrait information to obtain user feature information; the user feature information includes user portrait features, terminal attribute features, or behavior attribute features;

[0154] Acquire the attribute information of the virtual character, and perform feature extraction on the attribute information to obtain virtual feature information; the virtual feature information includes the style, dress, behavior, or length of stay in the virtual space of the virtual character.

[0155] In the embodiment, the user portrait information of the target user in the real domain is obtained, and the attribute information of the virtual character associated with the target user in the virtual space is obtained. Then, the feature extraction of the domain dimension of each domain is performed on the target user and the virtual character associated with the target user, and the feature is collected based on the respective space dimension. The user feature information corresponding to the target user and the virtual feature information corresponding to the virtual character are obtained.

[0156] Specifically, feature extraction is performed on the target user to obtain user feature information, which may include: user portrait features, terminal attribute features, and behavior attribute features.

[0157] Among them, user portrait features may include basic information such as the user's age, gender, occupation, interests and hobbies.

[0158] Terminal attribute characteristics may include the device type, operating system, and browser version used by the user. Terminal attribute characteristics can help the system understand the user's technical environment and usage habits.

[0159] Behavioral attribute characteristics may include purchasing or operation behavior characteristics. Behavioral attribute characteristics can analyze user needs and preferences by recording user shopping history, browsing history, click behavior and other data.

[0160] Extract features of virtual characters to obtain virtual feature information, which focuses on describing performance and behavior in the virtual environment.

[0161] Virtual feature information may include the style, dress, behavior, and duration of stay of the virtual character.

[0162] Among them, the style and dress of the virtual character can include the appearance, clothing and accessories of the virtual character, etc. These features can be extracted and classified through image processing or visual analysis technology.

[0163] The behavior of a virtual character may include the way the virtual character moves, interactive behaviors, or changes in facial expressions, etc. The behavior can be accurately recorded through animation capture and behavior recognition technology.

[0164] The length of time spent in the virtual space is an important indicator for measuring the activity of virtual characters, which can be obtained by tracking the length of time that virtual characters stay in different virtual scenes.

[0165] Based on this, by extracting and analyzing the features of real users and virtual characters associated with real users, a correlation relationship between the two can be established to provide data support for subsequent user grouping.

[0166] One or more technical solutions provided in the exemplary embodiments of the present disclosure first extract feature information of real users and virtual characters respectively through the feature extraction stage to obtain user feature information and virtual feature information, thereby comprehensively capturing multi-dimensional information of users in real and virtual environments. Compared with relying solely on historical behavior data of users in reality for feature extraction, this method greatly enriches the user's feature information and makes subsequent recommendation results more accurate.

[0167] Then, in the two-way grouping association stage, the user feature information and virtual feature information are used to build and train the deep neural network model, and the user grouping is completed for real users and virtual characters respectively. Then, the association mapping between virtual characters and real users is performed to obtain the real associated user group divided by virtual feature information. Through the association mapping between virtual characters and real users, the corresponding relationship between user feature information and virtual feature information can be established, which expands the range of recommended options.

[0168] Finally, in the fusion recommendation stage, the real user group and the real associated user group are fused to obtain the virtual-real fusion result. Combined with the preference extraction and collaborative recommendation capabilities, the preference extraction and recommendation of the users in the group of the virtual-real fusion result are performed. Through this virtual-real fusion recommendation strategy, the real preferences of users can be captured more accurately, and more personalized recommendation services can be provided, further improving the diversity and novelty of recommendations.

[0169] Therefore, the user preference recommendation method provided in the exemplary embodiment of the present disclosure can effectively solve the problems existing in the traditional recommendation method, such as insufficient accuracy, narrow recommendation scope, high content duplication, and serious monotony.

[0170] The above mainly introduces the solutions provided by the exemplary embodiments of the present disclosure. It is understandable that in order to implement the above functions, the electronic device includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present disclosure.

[0171] The exemplary embodiments of the present disclosure can divide the functional units of the electronic device according to the above method examples. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the exemplary embodiments of the present disclosure is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0172] In the case of dividing each functional module according to each function, an exemplary embodiment of the present disclosure provides a user preference recommendation device, which may be a server or a chip applied to a server. Figure 6 This is a schematic block diagram of the functional modules of an exemplary user preference recommendation device provided by the present disclosure. Figure 6 As shown, the user preference recommendation device 600 includes:

[0173] The data acquisition module 610 is used to acquire user characteristic information of a target user and virtual characteristic information of a virtual character associated with the target user;

[0174] The data processing module 620 is used to perform clustering calculations on the user characteristic information and the virtual characteristic information respectively, and determine the real user group and the virtual character group to which the target user belongs;

[0175] The data processing module 620 is further used to determine common users between the real user group and the virtual character group, and form a target group based on the common users; the number of the common users is greater than or equal to two;

[0176] The data processing module 620 is further configured to generate a user preference recommendation result of the target user based on the preference information corresponding to each common user in the target group.

[0177] In another embodiment provided by the present disclosure, the data processing module 620 is further used to obtain affiliation information; the affiliation information represents the association relationship between the real user and the virtual character; based on the affiliation information, determine the first real user associated with each virtual character in the virtual character group.

[0178] In another embodiment provided by the present disclosure, the data processing module 620 is further used to obtain the corresponding second real user in the real user group; determine multiple common users between the first real user and the second real user, and form the multiple common users into the target group.

[0179] In another embodiment provided by the present disclosure, the data processing module 620 is also used to obtain a trained group classification model; the training process of the group classification model includes: taking user feature training information and virtual feature training information as training samples, inputting the training samples into a pre-constructed group classification model, and obtaining user grouping prediction data; updating the parameters of the group classification model based on the user grouping prediction data to obtain a trained group classification model; inputting the user feature information into the group classification model to obtain a real user group corresponding to the target user; and inputting the virtual feature information into the group classification model to obtain a virtual character group corresponding to the target user.

[0180] In another embodiment provided by the present disclosure, the data processing module 620 is also used to construct a real user graph based on the user feature information; the nodes in the real user graph represent real users, the edges in the real user graph represent the relationships between real users, and the weights of the edges are set based on the user feature information of the real users; construct a virtual character graph based on the virtual feature information; the nodes in the virtual character graph represent virtual characters, the edges in the virtual character graph represent the relationships between virtual characters, and the weights of the edges are set based on the virtual feature information of the virtual characters; use a community detection algorithm to process the real user graph, find other real users with similar user feature information, and form the real user group; use a community detection algorithm to process the virtual character graph, find other virtual characters with similar virtual feature information, and form the virtual character group.

[0181] In another embodiment provided by the present disclosure, the data processing module 620 is also used to obtain preference information corresponding to each common user in the target group to form initial preference information; obtain first preference information corresponding to the target user; and generate user preference recommendation results for the target user based on the initial preference information and the first preference information.

[0182] In another embodiment provided by the present disclosure, the data processing module 620 is further used to obtain user portrait information of the target user, and perform feature extraction on the user portrait information to obtain user feature information; the user feature information includes user portrait features, terminal attribute features or behavior attribute features; obtain attribute information of the virtual character, and perform feature extraction on the attribute information to obtain virtual feature information; the virtual feature information includes the style, dress, behavior and action of the virtual character or the length of stay in the virtual space.

[0183] The exemplary embodiment of the present disclosure also provides an electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication. The memory stores a computer program that can be executed by the at least one processor, and the computer program is used to cause the electronic device to perform the method according to the embodiment of the present disclosure when executed by the at least one processor.

[0184] The exemplary embodiments of the present disclosure also provide a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is used to cause the computer to perform the method according to the embodiments of the present disclosure.

[0185] Figure 7 The block diagram of the electronic device provided as an example of the present disclosure, the block diagram of the electronic device 700 that can be used as the server or client of the present disclosure will now be described, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples, and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0186] like Figure 7 As shown, the electronic device 700 includes a computing unit 701, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the electronic device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0187] A plurality of components in the electronic device 700 are connected to the I / O interface 705, including: an input unit 706, an output unit 707, a storage unit 708, and a communication unit 709. The input unit 706 may be any type of device capable of inputting information to the electronic device 700, and the input unit 706 may receive input digital or character information, and generate key signal inputs related to user settings and / or function control of the electronic device. The output unit 707 may be any type of device capable of presenting information, and may include but is not limited to a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 708 may include but is not limited to a disk, an optical disk. The communication unit 709 allows the electronic device 700 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and may include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth™ device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0188] The computing unit 701 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 701 performs the various methods and processes described above. Each of the methods described above may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 700 via the ROM 702 and / or the communication unit 709.

[0189] Figure 8 This is a schematic diagram of an exemplary computer program product provided by the present disclosure. The exemplary embodiment of the present disclosure also provides a computer program product 800, including a computer program 801, wherein the computer program 801 is used to enable the computer to perform a method according to an embodiment of the present disclosure when executed by a processor of the computer.

[0190] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0191] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0192] As used in this disclosure, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and / or device (e.g., disk, optical disk, memory, programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0193] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0194] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0195] A computer system may include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other.

[0196] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instruction is loaded and executed on a computer, the process or function described in the embodiment of the present disclosure is executed in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, a terminal, a user device or other programmable device. The computer program or instruction may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program or instruction may be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server, data center, etc. that integrates one or more available media. The available medium may be a magnetic medium, for example, a floppy disk, a hard disk, a tape; it may also be an optical medium, for example, a digital video disc (DVD); it may also be a semiconductor medium, for example, a solid state drive (SSD).

[0197] Although the present disclosure has been described in conjunction with specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present disclosure. Accordingly, this specification and the drawings are merely exemplary illustrations of the present disclosure as defined by the appended claims, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present disclosure. Obviously, those skilled in the art may make various modifications and variations to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure is also intended to include these modifications and variations.

Claims

1. A method for recommending user preferences, characterized in that: The method comprises: Acquire user characteristic information of a target user and virtual characteristic information of a virtual character associated with the target user; Performing cluster calculations on the user characteristic information and the virtual characteristic information respectively to determine the real user group and the virtual character group to which the target user belongs; Determine common users between the real user group and the virtual character group, and form a target group based on the common users; the number of the common users is greater than or equal to two; Based on the preference information respectively corresponding to each of the common users in the target group, a user preference recommendation result of the target user is generated.

2. The method according to claim 1, characterized in that The method further comprises: Acquiring affiliation information; wherein the affiliation information represents the association relationship between the real user and the virtual character; Based on the affiliation information, first real users respectively associated with each virtual character in the virtual character group are determined.

3. The method according to claim 2, characterized in that The determining a plurality of common users between the real user group and the virtual character group, and forming a target group based on the plurality of common users, comprises: Acquire a corresponding second real user in the real user group; A plurality of common users between the first real user and the second real user are determined, and the plurality of common users are grouped into the target group.

4. The method according to claim 1, characterized in that The performing clustering calculations on the user characteristic information and the virtual characteristic information respectively to determine the real user group and the virtual character group corresponding to the target user includes: Acquire a trained group classification model; the training process of the group classification model includes: taking user feature training information and virtual feature training information as training samples, inputting the training samples into a pre-constructed group classification model, and obtaining user grouping prediction data; updating the parameters of the group classification model based on the user grouping prediction data, and obtaining the trained group classification model; Inputting the user characteristic information into the group classification model to obtain the actual user group corresponding to the target user; The virtual feature information is input into the group classification model to obtain the virtual character group corresponding to the target user.

5. The method according to claim 1, characterized in that The performing clustering calculations on the user characteristic information and the virtual characteristic information respectively to determine the real user group and the virtual character group corresponding to the target user includes: A real user graph is constructed based on the user characteristic information; the nodes in the real user graph represent real users, the edges in the real user graph represent the relationships between real users, and the weights of the edges are set based on the user characteristic information of the real users; Constructing a virtual character graph based on the virtual feature information; the nodes in the virtual character graph represent virtual characters, the edges in the virtual character graph represent the relationship between the virtual characters, and the weights of the edges are set based on the virtual feature information of the virtual characters; Using a community detection algorithm to process the real user graph, find other real users with similar user feature information, and form the real user group; The virtual character graph is processed using a community detection algorithm to find other virtual characters with similar virtual feature information to form the virtual character group.

6. The method according to claim 1, characterized in that The generating a user preference recommendation result of the target user based on the preference information corresponding to each common user in the target group includes: Obtaining preference information corresponding to each common user in the target group to form initial preference information; Acquire first preference information corresponding to the target user; Based on the initial preference information and the first preference information, a user preference recommendation result of the target user is generated.

7. The method according to claim 1, characterized in that The acquiring of user characteristic information of a target user and virtual characteristic information of a virtual character associated with the target user includes: Acquire user portrait information of the target user, and perform feature extraction on the user portrait information to obtain user feature information; the user feature information includes user portrait features, terminal attribute features, or behavior attribute features; Acquire the attribute information of the virtual character, and perform feature extraction on the attribute information to obtain virtual feature information; the virtual feature information includes the style, dress, behavior, or length of stay in the virtual space of the virtual character.

8. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the method of claim 1.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the method of claim 1 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method of claim 1 is implemented.

Images