An e-commerce transaction system and method based on virtual reality

Abstract

The application provides an e-commerce transaction system and method based on virtual reality, which extracts explicit interaction features and implicit interaction features of each user in a virtual reality shopping scene; associates the explicit interaction features and the implicit interaction features of each user with interaction nodes to obtain interaction attribute information of each user; extracts browsing interest points of other users with the same customer attributes as a target user when the other users shop in the virtual reality, and then predicts behavior preferences of the target user according to all the browsing interest points to obtain a preference prediction value; determines the difference degree of interaction behaviors between the target user and other users through all the interaction attribute information and the preference prediction value; and when the target user shops in the virtual reality using a virtual reality device, the virtual reality device predicts and displays recommended goods according to the current shopping intention of the target user and the difference degree of the interaction behaviors. The scheme of the application can realize personalized recommendation of goods in an e-commerce transaction system.

Description

Technical Field

[0001] This application relates to the field of virtual reality technology, and more specifically, to an e-commerce transaction system and method based on virtual reality. Background Technology

[0002] The application of virtual reality in e-commerce transactions is gradually becoming an emerging trend. With the rapid development of internet technology and virtual reality devices, traditional online shopping methods are gradually shifting towards immersive experiences. Virtual reality provides e-commerce platforms with a brand-new interactive mode. In traditional e-commerce, consumers can only learn about products through text, pictures, and videos, lacking a real perception and interactive experience of the products. However, virtual reality technology creates an immersive virtual shopping environment, allowing consumers to "personally" experience products in a virtual space. Whether it is clothing, furniture, or electronic products, they can view, try on, and use them through virtual models in 3D, thus breaking through the limitations of traditional e-commerce shopping.

[0003] However, existing technologies and traditional recommendation systems rely on static datasets and fixed algorithm strategies, resulting in recommendations that often fail to accurately reflect users' current interests and needs. Furthermore, traditional e-commerce platforms lack a comprehensive understanding of user interactions in virtual reality shopping scenarios, primarily relying on basic display behavior data (such as click counts and purchase history). This data cannot delve into users' latent needs, leading to low personalization of recommendations. By real-time correlation and analysis of users' explicit and implicit behavioral characteristics in virtual reality environments, not only can the accuracy of recommendation systems be improved, but recommendation strategies can also be adjusted based on users' dynamic shopping intentions, adapting more flexibly to changes in user needs, enhancing the shopping experience, and increasing purchase conversion rates. Therefore, how to achieve personalized product recommendations in e-commerce transaction systems has become a challenge for the industry. Summary of the Invention

[0004] This application provides an e-commerce transaction system and method based on virtual reality, which can realize personalized recommendations of products in the e-commerce transaction system.

[0005] In a first aspect, this application provides a product recommendation method based on virtual reality, used for product recommendation in a virtual reality-based e-commerce transaction system, comprising the following steps:

[0006] Collect all user interaction behavior data in the virtual reality mall from virtual reality devices, and then extract the explicit and implicit interaction features of each user in the virtual reality shopping scenario.

[0007] Interaction nodes are associated with the explicit and implicit interaction features of each user to determine the interaction attribute information of each user in the virtual reality shopping scenario.

[0008] The target user's customer attributes are detected, and the browsing interest points of other users with the same customer attributes as the target user are extracted when shopping in virtual reality on virtual reality devices. Then, based on all the browsing interest points, the target user's shopping behavior preferences are confidently predicted to obtain the preference prediction value.

[0009] The degree of difference in interactive behavior between the target user and other users in the virtual reality shopping scenario is determined by using all interactive attribute information and the predicted preference values;

[0010] When a target user uses a virtual reality device to shop in virtual reality, the virtual reality device predicts and displays recommended products based on the difference between the target user's current shopping intention and the interactive behavior.

[0011] Preferably, the extraction of explicit and implicit interaction features for each user in a virtual reality shopping scenario specifically includes:

[0012] For each user, acquire user interaction behavior data in the virtual reality marketplace;

[0013] Filter out explicit user behaviors in each shopping session from interactive behavior data, and determine the explicit interaction characteristics of users in virtual reality shopping scenarios through all explicit behaviors;

[0014] The implicit behaviors of users in each shopping session are filtered out from the interaction behavior data. The implicit interaction characteristics of users in the virtual reality shopping scenario are determined by all implicit behaviors, and then the explicit interaction characteristics and implicit interaction characteristics of each user in the virtual reality shopping scenario are obtained.

[0015] Preferably, the interaction node association between the explicit and implicit interaction features of each user, and the determination of the interaction attribute information of each user in the virtual reality shopping scenario, specifically includes:

[0016] For each user, extract all interaction nodes in the virtual reality shopping scenario;

[0017] Based on the user's explicit and implicit interaction characteristics, the interaction behavior of each interaction node is associated and perceived to obtain the associated attribute label of each interaction node.

[0018] By identifying all associated attribute tags, we can determine the user's interaction attribute information in the virtual reality shopping scenario, and thus obtain the interaction attribute information of each user in the virtual reality shopping scenario.

[0019] Preferably, extracting the browsing interest points of other users with the same customer attributes as the target user when shopping in virtual reality on a virtual reality device specifically includes:

[0020] Other users who have the same customer attributes as the target user will be designated as the target user;

[0021] For each specified user, obtain the user's interactive behavior data in the virtual reality mall, and then extract the user's gaze duration and gaze area during the browsing process;

[0022] The browsing interest points of a specified user during virtual reality shopping on a virtual reality device are determined by the gaze duration and the gaze area, thereby obtaining the browsing interest points of each specified user during virtual reality shopping on a virtual reality device.

[0023] Preferably, confidence prediction of the target user's shopping behavior preferences is made based on all browsing interests, and the resulting preference prediction values ​​specifically include:

[0024] Confidence associations are performed on all browsing points of interest to obtain the association factors between them.

[0025] Valid points of interest are selected from all browsing points of interest based on the aforementioned correlation factors;

[0026] By using all valid points of interest, the shopping behavior preferences of the target user are predicted, and the preference prediction value is obtained.

[0027] Preferably, the virtual reality device predicts and recommends products based on the difference between the target user's current shopping intent and the interactive behavior, specifically including:

[0028] Initialize a recommendation model;

[0029] The recommendation strategy for recommending products is adjusted based on the degree of difference in the aforementioned interactive behaviors;

[0030] The recommendation strategy built into the recommendation model is updated using the adjusted recommendation strategy;

[0031] The target user's current shopping intent is used as the input parameter of the recommendation model, and then the recommendation model is combined with the target user's interaction behavior data to predict and generate the target user's recommendation list.

[0032] The products in the recommended list are displayed sequentially using virtual reality devices.

[0033] Preferably, the virtual reality device is a head-mounted display device.

[0034] Secondly, this application provides a virtual reality-based e-commerce transaction system, which includes a product recommendation unit, the product recommendation unit comprising:

[0035] The data acquisition module is used to collect all user interaction behavior data in the virtual reality mall from the virtual reality device, and then extract the explicit and implicit interaction features of each user in the virtual reality shopping scenario.

[0036] The processing module is used to associate the explicit and implicit interaction features of each user with interaction nodes, thereby determining the interaction attribute information of each user in the virtual reality shopping scenario.

[0037] The processing module is also used to detect the customer attributes of the target user, extract the browsing interest points of other users with the same customer attributes as the target user when shopping in virtual reality on virtual reality devices, and then make a confidence prediction of the target user's shopping behavior preferences based on all browsing interest points to obtain the preference prediction value.

[0038] The processing module is also used to determine the degree of difference in interactive behavior between the target user and other users in the virtual reality shopping scenario through all interactive attribute information and the preference prediction value;

[0039] The execution module is used to predict and display recommended products based on the difference between the target user's current shopping intention and the interactive behavior when the target user uses the virtual reality device to shop in virtual reality.

[0040] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described virtual reality-based product recommendation method.

[0041] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described virtual reality-based product recommendation method.

[0042] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0043] In this embodiment, interactive behavior data of all users in a virtual reality shopping mall are collected from a virtual reality device, and then explicit and implicit interactive features of each user in the virtual reality shopping scene are extracted. Interactive nodes are associated with the explicit and implicit interactive features of each user to determine the interactive attribute information of each user in the virtual reality shopping scene. Customer attributes of the target user are detected, and browsing interest points of other users with the same customer attributes are extracted when shopping in the virtual reality device. Based on all browsing interest points, confidence prediction of the target user's shopping behavior preferences is performed to obtain a preference prediction value. The degree of difference in interactive behavior between the target user and other users in the virtual reality shopping scene is determined through all interactive attribute information and the preference prediction value. When the target user uses the virtual reality device for virtual reality shopping, the virtual reality device predicts and displays recommended products based on the target user's current shopping intention and the degree of difference in interactive behavior.

[0044] Therefore, this application determines the degree of difference in interactive behavior between the target user and other users in a virtual reality shopping scenario by using interactive attribute information and preference prediction values, and then predicts the recommended products for the target user based on the degree of difference in interactive behavior. First, by associating the user's explicit and implicit interactive features with interactive nodes, the user's behavioral patterns and preferences can be accurately described, thereby determining the user's interactive attribute information in the virtual reality shopping scenario. Among these, the extraction and association of explicit and implicit interactive features helps the system to comprehensively understand the user's shopping habits and interests. Second, based on the browsing interest points of all users, confidence prediction is made on the target user's shopping behavior preferences to obtain preference prediction values. These preference prediction values ​​can reflect the user's potential interest in specific products or product categories, and can also predict the user's potential interest in products or product categories before the user has already expressed interest. When users clearly express interest, the system can predict their potential product interests. This preference prediction not only enhances the foresight of product recommendations but also allows for real-time adjustments to the recommendation strategy to address dynamic changes in user needs. Furthermore, by combining interaction attribute information and preference predictions, the system can analyze the differences in interaction behavior between the target user and other users in a virtual reality shopping scenario. This identifies user groups with similar shopping needs to the target user, making the recommendation strategy more flexible and accurately matching user requirements. Finally, based on the analysis of interaction behavior differences, personalized recommendations can accurately push products that meet user needs, significantly improving recommendation effectiveness, enhancing the user shopping experience, and driving transaction volume growth on e-commerce platforms. In summary, the solution presented in this application can achieve personalized product recommendations in e-commerce transaction systems. Attached Figure Description

[0045] Figure 1 This is an exemplary flowchart of a virtual reality-based product recommendation method according to some embodiments of this application;

[0046] Figure 2 This is a schematic diagram of an e-commerce transaction process according to some embodiments of this application;

[0047] Figure 3 This is a flowchart illustrating the process of determining preference prediction values ​​according to some embodiments of this application;

[0048] Figure 4 This is a schematic diagram of the structure of a product recommendation unit according to some embodiments of this application;

[0049] Figure 5 This is a schematic diagram of the structure of a computer device implementing a virtual reality-based product recommendation method according to some embodiments of this application. Detailed Implementation

[0050] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0051] refer to Figure 1 The figure is an exemplary flowchart of a virtual reality-based product recommendation method 100 according to some embodiments of this application. The virtual reality-based product recommendation method 100 mainly includes the following steps:

[0052] In step 101, all user interaction behavior data in the virtual reality mall are collected from the virtual reality device, and then the explicit interaction features and implicit interaction features of each user in the virtual reality shopping scene are extracted.

[0053] It should be noted that the reference Figure 2 As shown in the figure, this diagram is a schematic diagram of the e-commerce transaction process in some embodiments of this application. First, the distributor submits a distributor qualification application to the e-commerce transaction platform. The e-commerce transaction platform is responsible for reviewing and authorizing the store. Second, the operating customer and the operating store place an order through virtual reality device or client. The order information is transmitted to the store. After the store processes the order, the e-commerce transaction platform settles the payment and delivers the goods to the customer. After receiving the goods, the customer signs for them, completing the entire transaction process. The diagram shows the interaction relationship between each link, including the distributor's qualification application and authorization with the e-commerce transaction platform, the operating customer and the store placing an order through virtual reality device (i.e., client), and the settlement and delivery of goods between the e-commerce transaction platform and the store.

[0054] It should also be noted that the interactive behavior data in this application refers to the human-computer interaction data generated by users during e-commerce transactions using virtual reality devices; it should also be noted that the virtual reality device in this application is a head-mounted display device, but in other embodiments the virtual reality device may be other devices, which are not specifically limited here.

[0055] In specific implementation, collecting all user interaction behavior data in the virtual reality marketplace from the virtual reality device can be achieved in the following way: First, the virtual reality device perceives the user's operation behavior (such as clicks, gestures, and gaze trajectory) in the virtual reality marketplace in real time to obtain user interaction behavior data. Then, the collected interaction behavior data is transmitted to the cloud using a high-speed network protocol (such as WebSocket). A distributed storage system (such as HDFS or a cloud database) is used to store and manage the large-scale interaction behavior data. Finally, the interaction behavior data of all users in the virtual reality marketplace within a specified historical time period can be obtained from the distributed storage system. It should be noted that the specified historical time period in this application is the time period from the current moment to the past six months. In other embodiments, the specified historical time period can also be other time periods, which are not specifically limited here.

[0056] In some embodiments, extracting the explicit and implicit interaction features of each user in a virtual reality shopping scenario can be achieved through the following steps:

[0057] For each user, acquire user interaction behavior data in the virtual reality marketplace;

[0058] Filter out explicit user behaviors in each shopping session from interactive behavior data, and determine the explicit interaction characteristics of users in virtual reality shopping scenarios through all explicit behaviors;

[0059] The implicit behaviors of users in each shopping session are filtered out from the interaction behavior data. The implicit interaction characteristics of users in the virtual reality shopping scenario are determined by all implicit behaviors, and then the explicit interaction characteristics and implicit interaction characteristics of each user in the virtual reality shopping scenario are obtained.

[0060] It should be noted that explicit behavior in this application refers to actions by which a user explicitly expresses their intentions and needs within a virtual reality device. These are typically user-initiated, directly observable, and quantifiable operations, such as clicking on a product or making a purchase. Implicit behavior in this application refers to user actions within a virtual reality device that do not directly express intentions but indirectly reflect interests and needs through their behavioral patterns. Furthermore, the explicit interaction features in this application...

[0061] In specific implementation, the explicit behaviors of users in each shopping session are filtered from the interaction behavior data. Determining the explicit interaction characteristics of users in the virtual reality shopping scene through all explicit behaviors can be achieved in the following way: Filtering feature data that reflects users' product click and purchase behaviors in each shopping session from the interaction behavior data, extracting the click frequency and purchase frequency during the shopping process from the feature data, and then converting all click frequencies and purchase frequencies into feature vectors according to the chronological order of recording time. These feature vectors are then used as the explicit interaction characteristics of users in the virtual reality shopping scene. Similarly, the implicit behaviors of users in each shopping session are filtered from the interaction behavior data. Determining the implicit interaction characteristics of users in the virtual reality shopping scene through all implicit behaviors can be achieved in the following way: Filtering feature data that reflects users' browsing behavior in each shopping session from the interaction behavior data, extracting the browsing duration and field-of-view density during the shopping process from the feature data, and then converting all browsing duration and field-of-view density into feature vectors according to the chronological order of recording time. These feature vectors are then used as the implicit interaction characteristics of users in the virtual reality shopping scene.

[0062] In step 102, interactive nodes are associated with the explicit and implicit interactive features of each user to determine the interactive attribute information of each user in the virtual reality shopping scenario.

[0063] In some embodiments, associating explicit and implicit interaction features of each user with interaction nodes to determine the interaction attribute information of each user in a virtual reality shopping scenario can be achieved through the following steps:

[0064] For each user, extract all interaction nodes in the virtual reality shopping scenario;

[0065] Based on the user's explicit and implicit interaction characteristics, the interaction behavior of each interaction node is associated and perceived to obtain the associated attribute label of each interaction node.

[0066] By identifying all associated attribute tags, we can determine the user's interaction attribute information in the virtual reality shopping scenario, and thus obtain the interaction attribute information of each user in the virtual reality shopping scenario.

[0067] It should be noted that the interaction node in this application refers to the key node corresponding to each interaction behavior of the user in the virtual reality scene. Each interaction node carries the time, location and related attribute information of the behavior. The associated attribute tag in this application refers to the attribute identifier assigned to each interaction node in the virtual reality scene. In addition, the interaction attribute information in this application refers to the characteristic information that can reflect the user's interaction behavior in the virtual reality scene, such as the user's behavior pattern and operating habits in a specific scene.

[0068] In specific implementation, extracting all user interaction nodes in a virtual reality shopping scenario can be achieved in the following way: The time node of each operation can be obtained from the user's operation record, and this time node can be defined as an interaction node, thus obtaining all user interaction nodes in the virtual reality shopping scenario. Associating the interactive behavior of each interaction node with the user's explicit and implicit interaction characteristics to obtain the association attribute label of each interaction node can be achieved in the following way: Existing association rule mining algorithms (such as the Apriori algorithm) can be used to map explicit and implicit interaction characteristics to each specific interaction node, and then the association rule mining algorithm can be used to... The algorithm learns the association information between explicit and implicit interaction features at interaction nodes, and then outputs the association value of each interaction node through association rule mining algorithm, and uses the association value of each interaction node as the association attribute label of the corresponding interaction node. The user's interaction attribute information in the virtual reality shopping scene can be determined by the following method: all association attribute labels can be grouped according to the category of interaction attribute, and the association attribute labels of each category group can be statistically analyzed. Then, the statistical value of each category group can be used as the attribute value of the corresponding interaction attribute category. Finally, all attribute values ​​are combined into a set as the user's interaction attribute information in the virtual reality shopping scene.

[0069] In step 103, the target user's customer attributes are detected, and the browsing interest points of other users with the same customer attributes as the target user when shopping in virtual reality on virtual reality devices are extracted. Then, based on all the browsing interest points, a confidence prediction of the target user's shopping behavior preferences is made to obtain the preference prediction value.

[0070] It should be noted that customer attributes refer to information describing the individual characteristics of a customer. In this application, customer attributes are the basic attribute information of a user, including information such as age, gender, and geographical location. In other embodiments, customer attributes may also include other information of the user, which is not specifically limited here.

[0071] In some embodiments, extracting the browsing interest points of other users with the same customer attributes as the target user when shopping in a virtual reality device can be achieved through the following steps:

[0072] Other users who have the same customer attributes as the target user will be designated as the target user;

[0073] For each specified user, obtain the user's interactive behavior data in the virtual reality mall, and then extract the user's gaze duration and gaze area during the browsing process;

[0074] The browsing interest points of a specified user during virtual reality shopping on a virtual reality device are determined by the gaze duration and the gaze area, thereby obtaining the browsing interest points of each specified user during virtual reality shopping on a virtual reality device.

[0075] It should be noted that, in this application, gaze duration refers to the duration for which a user's gaze is focused on a specific target (such as a product) in a virtual reality environment; gaze area refers to the specific spatial area where a user's gaze lingers in a virtual reality environment; and browsing points of interest refers to information points that a user shows high attention to based on their interactive behavior in a virtual reality shopping scenario. Browsing points of interest can reflect a user's immediate needs and potential preferences.

[0076] In practical implementation, designating other users with the same customer attributes as the target user as designated users can be achieved in the following way: First, filter out a group of users with the same customer attributes as the target user, and designate all users in this group as designated users. For each designated user, obtain their interaction behavior data in the virtual reality marketplace, and then extract their gaze duration and gaze area during product browsing. This can be achieved in the following way: Obtain the designated user's interaction behavior data in the virtual reality marketplace from a distributed storage system, and then extract the designated user's gaze duration and gaze area from the obtained interaction behavior data. The gaze duration can be obtained by analyzing the length of time the designated user's gaze lingers on a specific product or interface element, while the gaze area is captured in real-time using an eye tracker in the virtual reality device to capture the user's gaze trajectory. The process involves obtaining the gaze coordinates of the eyes and mapping them to the displayed content (such as products, interface elements, etc.) in the virtual reality environment to obtain the user's gaze focus. The virtual reality environment is then divided into several virtual areas (e.g., product display areas, navigation buttons, advertising banners). Based on the position of the gaze focus, the user's gaze focus is matched with these areas to obtain the gaze area. Determining the browsing interest points of a specific user during virtual reality shopping using the gaze duration and gaze area can be achieved by combining gaze duration and gaze area and using statistical analysis methods (such as heatmap analysis) to identify the browsing interest points of the specified user. These interest points can represent products, areas, or information that the user pays close attention to, thus obtaining the browsing interest points of each specified user during virtual reality shopping.

[0077] In some embodiments, reference Figure 3 As shown in the figure, this is a flowchart illustrating the process of determining preference prediction values ​​in some embodiments of this application. In this embodiment, confidence prediction of the target user's shopping behavior preferences is made based on all browsing interest points, and the preference prediction values ​​can be obtained by the following steps:

[0078] In step 1031, confidence association is performed on all browsing points of interest to obtain the association factors between browsing points of interest;

[0079] In step 1032, valid points of interest are selected from all browsing points of interest based on the association factors;

[0080] In step 1033, the shopping behavior preferences of the target user are predicted using all valid points of interest to obtain the preference prediction value.

[0081] It should be noted that the correlation factor in this application is an indicator that measures the degree of correlation between browsing points of interest; the effective points of interest in this application refer to information points that can truly reflect user behavioral preferences; and the preference prediction value in this application is an indicator that measures the degree of potential user preference for predicted goods or content.

[0082] In practice, the association analysis of all browsing points of interest (OPIs) to obtain the association factors between them can be achieved as follows: Select a designated user as the chosen user. Existing association rule algorithms (such as the Apriori algorithm) can be used to analyze the association between the chosen user and all other designated users' OPIs. The association rule algorithm then outputs an association value describing the OPIs between the chosen user and all other designated users. This association value is used as the association factor between the chosen user and the corresponding OPIs of all other designated users. Repeating the above steps yields the association factors between the remaining designated users and the corresponding OPIs of all other designated users. Valid OPIs can be selected from all OPIs based on these association factors as follows: A pre-set association threshold is used, which reflects the association relationship between OPIs. The average level is then calculated, and each correlation factor is compared with the correlation threshold. When the correlation factor is greater than or equal to the correlation threshold, the browsing interest point of the selected user corresponding to the correlation factor is determined to be a valid browsing interest point, and the valid browsing interest point is used as a valid interest point. When the correlation factor is less than the correlation threshold, the browsing interest point of the selected user corresponding to the correlation factor is determined to be an invalid browsing interest point. All valid interest points can be obtained in the above way. The shopping behavior preference of the target user can be predicted by using all valid interest points to obtain the preference prediction value. That is, a preference prediction model based on a neural network is initialized, all valid interest points are input into the preference prediction model, the preference prediction value is used as the output label in the preference prediction model, and the preference prediction model is used to predict the shopping behavior preference of the target user. Thus, the prediction result of the preference prediction model can be used as the preference prediction value.

[0083] It should be noted that the training steps of the preference prediction model include the following stages: Data collection and preprocessing: Collect user behavior data (such as clicks, browsing, and purchases), clean and format the data, remove noise data, and standardize it, such as filling in missing values ​​or removing invalid behaviors; Feature extraction: Extract users' browsing interest points from user interaction behavior data as input features for the model; Model selection and initialization: Select a suitable preference prediction model, such as collaborative filtering, matrix factorization, neural networks, etc., and initialize the model's structure and parameters. If a deep learning model is used, initialize the weights and biases of the neural network; Model training and optimization: Train the model using training data, and adjust the model parameters through optimization algorithms (such as gradient descent) to make the predicted values ​​as close as possible to the user's actual preferences. During training, cross-validation can be used to prevent overfitting; Model evaluation and tuning: Evaluate the model's performance on the validation set, using evaluation metrics such as precision, recall, and F1 score to measure the accuracy of the predictions. If the model performs poorly, it may be necessary to adjust the model architecture, learning rate, or other hyperparameters, and retrain and optimize.

[0084] In step 104, the degree of difference in interactive behavior between the target user and other users in the virtual reality shopping scenario is determined by using all interactive attribute information and the preference prediction value.

[0085] In some embodiments, determining the degree of difference in interactive behavior between a target user and other users in a virtual reality shopping scenario using all interactive attribute information and the preference prediction value can be achieved through the following steps:

[0086] Based on the predicted preference values, users with the same preference behaviors as the target users are selected;

[0087] Filter out the interaction attribute information of all tagged users;

[0088] By performing a differential analysis between the interaction attribute information of the target user and the interaction attribute information of all marked users, the degree of difference in the interactive behavior between the target user and other users in the virtual reality shopping scenario can be obtained.

[0089] In specific implementation, the following methods can be used to select labeled users with the same preference behavior as the target user based on the preference prediction value: The existing cosine similarity can be used to compare the preference prediction values ​​of other users with those of the target user, selecting all users with similar preference behaviors, and all selected users are designated as labeled users. Labeled users refer to the group of users whose preference prediction values ​​are highly similar to those of the target user. The following methods can be used to select the interaction attribute information of all labeled users: The interaction attribute information of all labeled users can be selected from all interaction attribute information. The following methods can be used to perform a differential analysis between the interaction attribute information of the target user and the interaction attribute information of all labeled users to obtain the degree of difference in interaction behavior between the target user and other users in the virtual reality shopping scenario: The average Euclidean distance between the interaction attribute information of the target user and each labeled user can be used as the degree of difference in interaction behavior between the target user and other users in the virtual reality shopping scenario.

[0090] It should be noted that the degree of difference in interactive behavior in this application refers to the degree of difference in interactive attributes between the target user and other users in a virtual reality shopping scenario.

[0091] In step 105, when the target user uses the virtual reality device to shop in virtual reality, the virtual reality device predicts and displays recommended products based on the difference between the target user's current shopping intention and the interactive behavior.

[0092] In some embodiments, the virtual reality device predicts and displays recommended products based on the difference between the target user's current shopping intent and the interactive behavior, which can be achieved through the following steps:

[0093] Initialize a recommendation model;

[0094] The recommendation strategy for recommending products is adjusted based on the degree of difference in the aforementioned interactive behaviors;

[0095] The recommendation strategy built into the recommendation model is updated using the adjusted recommendation strategy;

[0096] The target user's current shopping intent is used as the input parameter of the recommendation model, and then the recommendation model is combined with the target user's interaction behavior data to predict and generate the target user's recommendation list.

[0097] The products in the recommended list are displayed sequentially using virtual reality devices.

[0098] It should be noted that the recommendation model in this application refers to a machine learning model used for personalized product recommendations, which is a neural network-based recommendation model.

[0099] In specific implementation, initializing a recommendation model can be achieved as follows: a recommendation model can be trained based on a neural network and then initialized. Adjusting the product recommendation strategy based on the difference in interaction behavior can be achieved as follows: an existing reinforcement learning model can be used to automatically adjust the product recommendation strategy. If the difference in interaction behavior is less than a first threshold, the recommendation strategy may tend to recommend products similar to the labeled user; if the difference is greater than a second threshold, the recommendation strategy needs to focus on the unique interests of the target user and improve the user's personalized needs. Updating the built-in recommendation strategy in the recommendation model using the adjusted recommendation strategy can be achieved as follows: the adjusted recommendation strategy replaces and updates the built-in recommendation strategy in the recommendation model. During the update process, an online learning mechanism can be used for real-time updates. The recommendation model, which uses the target user's current shopping intent as input parameters and then combines this with the target user's interaction behavior data to predict and generate a recommendation list for the target user, can be implemented as follows: The target user's current shopping intent is input into the recommendation model. This intent can be inferred from the user's specific behavior in the virtual reality scene (such as search keywords). The recommendation model combines shopping intent information and the user's historical interaction behavior to predict the user's purchasing needs, and the output list predicted by the model is used as the target user's recommendation list. The products in the recommendation list are then displayed sequentially using a virtual reality device. This can be achieved by using an interactive interface, such as a virtual shelf or 3D product display, to display recommended products, ensuring that the user can clearly see product information and details.

[0100] On the other hand, in some embodiments, this application provides a virtual reality-based e-commerce transaction system, which includes a product recommendation unit, as referenced. Figure 4 The figure is a schematic diagram of the structure of a product recommendation unit according to some embodiments of this application. The product recommendation unit 400 includes: a collection module 401, a processing module 402, and an execution module 403, which are described below:

[0101] The acquisition module 401 in this application is mainly used to collect all users' interactive behavior data in the virtual reality mall from the virtual reality device, and then extract the explicit interaction features and implicit interaction features of each user in the virtual reality shopping scene.

[0102] Processing module 402, in this application, is used to associate the explicit and implicit interaction features of each user with interaction nodes, thereby determining the interaction attribute information of each user in the virtual reality shopping scene;

[0103] In this application, the processing module 402 is also used to detect the customer attributes of the target user, extract the browsing interest points of other users with the same customer attributes as the target user when shopping in virtual reality on virtual reality devices, and then make a confidence prediction of the target user's shopping behavior preferences based on all browsing interest points to obtain the preference prediction value.

[0104] In this application, the processing module 402 is also used to determine the degree of difference in the interactive behavior between the target user and other users in the virtual reality shopping scenario through all interactive attribute information and the preference prediction value;

[0105] The execution module 403 in this application is mainly used to predict and display recommended products based on the difference between the target user's current shopping intention and the interactive behavior when the target user uses the virtual reality device to make virtual reality shopping.

[0106] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described virtual reality-based product recommendation method.

[0107] In some embodiments, reference Figure 5 The figure is a schematic diagram of the structure of a computer device implementing a virtual reality-based product recommendation method according to some embodiments of this application. The virtual reality-based product recommendation method in the above embodiments can be implemented through... Figure 5 The computer device shown is used to implement this, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0108] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0109] The communication bus 502 can be used to transmit information between the aforementioned components.

[0110] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0111] The memory 503 stores program code for executing the solution of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. In the above embodiments, the virtual reality-based product recommendation method can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0112] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0113] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0114] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0115] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described virtual reality-based product recommendation method.

[0116] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0117] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A product recommendation method based on virtual reality, used for product recommendation in a virtual reality-based e-commerce transaction system, characterized in that, Includes the following steps: Collect all user interaction behavior data in the virtual reality mall from virtual reality devices, and then extract the explicit and implicit interaction features of each user in the virtual reality shopping scenario. Interaction nodes are associated with the explicit and implicit interaction features of each user to determine the interaction attribute information of each user in the virtual reality shopping scenario. The target user's customer attributes are detected, and the browsing interest points of other users with the same customer attributes as the target user are extracted when shopping in virtual reality on virtual reality devices. Then, based on all the browsing interest points, the target user's shopping behavior preferences are confidently predicted to obtain the preference prediction value. The degree of difference in interactive behavior between the target user and other users in the virtual reality shopping scenario is determined by using all interactive attribute information and the predicted preference values. When a target user uses a virtual reality device to shop in virtual reality, the virtual reality device predicts and displays recommended products based on the difference between the target user's current shopping intention and the interactive behavior.

2. The method as described in claim 1, characterized in that, Extracting the explicit and implicit interaction features of each user in a virtual reality shopping scenario specifically includes: For each user, acquire user interaction behavior data in the virtual reality marketplace; Filter out explicit user behaviors in each shopping session from interactive behavior data, and determine the explicit interaction characteristics of users in virtual reality shopping scenarios through all explicit behaviors; The implicit behaviors of users in each shopping session are filtered out from the interaction behavior data. The implicit interaction characteristics of users in the virtual reality shopping scenario are determined by all implicit behaviors, and then the explicit interaction characteristics and implicit interaction characteristics of each user in the virtual reality shopping scenario are obtained.

3. The method as described in claim 1, characterized in that, The interaction node association is performed on the explicit and implicit interaction features of each user to determine the interaction attribute information of each user in the virtual reality shopping scenario. Specifically, this includes: For each user, extract all interaction nodes in the virtual reality shopping scenario; Based on the user's explicit and implicit interaction characteristics, the interaction behavior of each interaction node is associated and perceived to obtain the associated attribute label of each interaction node. By identifying all associated attribute tags, we can determine the user's interaction attribute information in the virtual reality shopping scenario, and thus obtain the interaction attribute information of each user in the virtual reality shopping scenario.

4. The method as described in claim 1, characterized in that, Extracting the browsing interests of other users with the same customer attributes as the target user during virtual reality shopping on a virtual reality device specifically includes: Other users who have the same customer attributes as the target user will be designated as the target user; For each specified user, obtain the user's interactive behavior data in the virtual reality mall, and then extract the user's gaze duration and gaze area during the browsing process; The browsing interest points of a specified user during virtual reality shopping on a virtual reality device are determined by the gaze duration and the gaze area, thereby obtaining the browsing interest points of each specified user during virtual reality shopping on a virtual reality device.

5. The method as described in claim 1, characterized in that, Based on all browsing interests, a confidence prediction of the target user's shopping behavior preferences is made, resulting in the following preference prediction values: Confidence associations are performed on all browsing points of interest to obtain the association factors between them. Valid points of interest are selected from all browsing points of interest based on the aforementioned correlation factors; By using all valid points of interest, the shopping behavior preferences of the target user are predicted, and the preference prediction value is obtained.

6. The method as described in claim 1, characterized in that, Virtual reality devices predict and recommend products based on the differences between the target user's current shopping intent and the interactive behavior, specifically including: Initialize a recommendation model; The recommendation strategy for recommending products is adjusted based on the degree of difference in the aforementioned interactive behaviors; The recommendation strategy built into the recommendation model is updated using the adjusted recommendation strategy; The target user's current shopping intent is used as the input parameter of the recommendation model, and then the recommendation model is combined with the target user's interaction behavior data to predict and generate the target user's recommendation list. The products in the recommended list are displayed sequentially using virtual reality devices.

7. The method as described in claim 1, characterized in that, The virtual reality device is a head-mounted display device.

8. A virtual reality-based e-commerce transaction system, comprising a product recommendation unit, characterized in that, The product recommendation unit includes: The data acquisition module is used to collect all user interaction behavior data in the virtual reality mall from the virtual reality device, and then extract the explicit and implicit interaction features of each user in the virtual reality shopping scenario. The processing module is used to associate the explicit and implicit interaction features of each user with interaction nodes, thereby determining the interaction attribute information of each user in the virtual reality shopping scenario. The processing module is also used to detect the customer attributes of the target user, extract the browsing interest points of other users with the same customer attributes as the target user when shopping in virtual reality on virtual reality devices, and then make a confidence prediction of the target user's shopping behavior preferences based on all browsing interest points to obtain the preference prediction value. The processing module is also used to determine the degree of difference in interactive behavior between the target user and other users in the virtual reality shopping scenario through all interactive attribute information and the preference prediction value; The execution module is used to predict and display recommended products based on the difference between the target user's current shopping intention and the interactive behavior when the target user uses the virtual reality device to shop in virtual reality.

9. A computer device comprising a memory and a processor, the memory storing code, characterized in that, The processor is configured to acquire the code and execute the virtual reality-based product recommendation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the virtual reality-based product recommendation method as described in any one of claims 1 to 7.

Images