Determination method of exhibition route, electronic equipment, storage medium and program product

Abstract

The embodiment of the invention provides an exhibition viewing route determination method, electronic equipment, a storage medium and a program product. The method comprises the following steps: collecting user information corresponding to a current user and generating a unique identity identifier; analyzing and processing the user information to obtain a target interest type corresponding to the current user; according to the target interest type, determining a plurality of interest exhibition items corresponding to the current user in a plurality of exhibition items of a virtual exhibition hall; generating a plurality of exhibition routes according to the plurality of interest exhibition items and the at least one preset exposure exhibition item; and determining a target exhibition-viewing route in the plurality of exhibition-viewing routes, and recommending the target exhibition-viewing route to the current user through the unique identity identifier. The method is used for improving the effectiveness of exhibition recommendation.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method for determining an exhibition route, electronic equipment, storage medium, and program product. Background Technology

[0003] In related technologies, to enhance the activity of virtual exhibition halls, rewards are triggered based on behavioral metrics (such as user login frequency and online time) to attract user participation. However, when recommending exhibits and planning reward activities, user interests are not considered. As a result, users participate in reward activities merely to complete tasks and obtain rewards, rather than genuinely engaging them in browsing each exhibit. This makes it difficult to establish sustained interaction with users, leading to low effectiveness of exhibition recommendations. Summary of the Invention

[0004] This application provides a method for determining exhibition routes, electronic devices, storage media, and program products to improve the effectiveness of exhibition recommendations.

[0005] In a first aspect, embodiments of this application provide a method for determining exhibition routes, including:

[0006] Collect user information corresponding to the current user and generate a unique identity identifier. The user information includes the current user's basic information and the virtual exhibition hall's viewing information.

[0007] The user information is analyzed and processed to obtain the target interest type corresponding to the current user;

[0008] Based on the target interest type, determine the multiple interest items corresponding to the current user from among the multiple exhibits in the virtual exhibition hall;

[0009] Based on the multiple interest exhibits and the at least one preset exposure exhibit, multiple viewing routes are generated. The viewing routes are used to guide the current user to browse exhibits in the virtual exhibition hall. The viewing routes include multiple exhibits and the required browsing time for each exhibit. The required browsing time is used to limit the dwell time on the corresponding exhibit. The multiple exhibits are a combination of exhibits that conform to the target interest type and the exhibition hall operation goals.

[0010] The target viewing route is determined from the multiple viewing routes, and recommended to the current user through the unique identifier.

[0011] In one possible implementation, based on the target interest type, determining multiple interest items corresponding to the current user from among multiple items in the virtual exhibition hall includes:

[0012] The virtual exhibition hall's multiple exhibits are tagged to obtain multiple theme tags corresponding to each exhibit;

[0013] Word embedding technology is used to convert the current user's target interest type into an interest feature vector;

[0014] The multiple interest exhibits are determined based on the similarity between the multiple topic tags corresponding to each exhibit and the interest feature vector.

[0015] In one possible implementation, the plurality of interest exhibits are determined based on the similarity between the plurality of topic tags corresponding to each exhibit and the interest feature vector, including:

[0016] For any exhibit, determine the similarity between the interest feature vector and each topic tag corresponding to the exhibit, and perform a weighted summation of the similarity and weight of each topic tag to obtain the interest similarity between the interest feature vector and the exhibit.

[0017] The exhibits with an interest similarity greater than a preset similarity are identified as interest exhibits, thus obtaining multiple interest exhibits.

[0018] In one possible implementation, the user information is analyzed and processed to obtain the target interest type corresponding to the current user, including:

[0019] Determine whether the current user's exhibition viewing information is valid information. The exhibition viewing information includes multiple click events of the current user on exhibits in the virtual exhibition hall and the page dwell time corresponding to each click event.

[0020] If so, the target interest type is determined based on the multiple item click events and the page dwell time corresponding to each item click event;

[0021] If not, then the target interest type is determined based on the current user's basic information.

[0022] In one possible implementation, determining whether the current user's viewing information is valid includes:

[0023] The authenticity assessment model is used to evaluate the multiple display item click events and the page dwell time corresponding to each display item click event to obtain the predicted evaluation value of the current user.

[0024] If the predicted evaluation value is greater than or equal to the preset evaluation value, then the current user's exhibition viewing information is valid information;

[0025] If the predicted evaluation value is less than the preset evaluation value, then the current user's viewing information is invalid.

[0026] In one possible implementation, the target interest type is determined based on the plurality of item click events and the page dwell time corresponding to each item click event, including:

[0027] Determine the multiple display item theme types corresponding to the multiple display item click events, and the actual click frequency corresponding to each display item theme type;

[0028] Based on the actual click frequency corresponding to each exhibition item theme type, multiple first interest types are determined among the multiple exhibition item theme types;

[0029] Based on the page dwell time corresponding to each item click event, multiple second interest types are determined among the multiple item theme types;

[0030] The exhibit theme type that overlaps with the plurality of first interest types and the plurality of second interest types is determined as the target interest type.

[0031] In one possible implementation, for any given exhibition route, the exhibition route is generated based on the plurality of exhibits of interest and the at least one preset exposure exhibit, including:

[0032] Determine the number of exhibits corresponding to the exhibition route, wherein the number of exhibits corresponding to the at least one preset exposure exhibit is greater than the number of exposure exhibits corresponding to the at least one preset exposure exhibit.

[0033] The difference between the number of items displayed along the route and the number of items displayed in the exposure section is determined as the number of items of interest.

[0034] The exhibition route is constructed by combining the number of interest items among the multiple interest items with the at least one preset exposure item, and the first item in the exhibition route is the current user's interest item.

[0035] In one possible implementation, recommending a target viewing route to the current user further includes:

[0036] The system recommends reward information corresponding to the target exhibition route to the current user. The reward information includes multiple exhibition visit quantity ranges and the reward recipients corresponding to each exhibition visit quantity range.

[0037] In one possible implementation, it further includes:

[0038] Obtain the multiple exhibition items that the current user has viewed on the target exhibition route, as well as the dwell time for each exhibition item;

[0039] The number of valid exhibits corresponding to the multiple browsing exhibits is determined based on the dwell time corresponding to each browsing exhibit.

[0040] Based on the number of valid exhibits, the target reward recipient for the current user is determined from among the reward recipients corresponding to each exhibit visit range.

[0041] In one possible implementation, after recommending a target viewing route to the user, the process further includes:

[0042] Obtain the current viewing status of the current user, which includes the currently viewed exhibits, the exhibits already viewed, and the viewing duration of each viewed exhibit;

[0043] Based on the current viewing status, determine the current non-interest tag of the current user;

[0044] Identify at least one unviewed exhibit in the target viewing route;

[0045] The browsing position of the unviewed exhibits corresponding to the non-interest tags will be moved to the back, and the corresponding browsing time will be shortened.

[0046] Secondly, embodiments of this application provide a device for determining exhibition routes, including a data acquisition module, an analysis and processing module, a first determination module, a generation module, a second determination module, and a recommendation module:

[0047] The data collection module is used to collect user information corresponding to the current user and generate a unique identity identifier. The user information includes the current user's basic information and the viewing information of the virtual exhibition hall.

[0048] The analysis and processing module is used to analyze and process the user information to obtain the target interest type corresponding to the current user;

[0049] The first determining module is used to determine, based on the target interest type, multiple interest items corresponding to the current user among multiple exhibits in the virtual exhibition hall;

[0050] The generation module is used to generate multiple viewing routes based on the multiple interest exhibits and the at least one preset exposure exhibit. The viewing routes are used to guide the current user to browse exhibits in the virtual exhibition hall. The viewing routes include multiple exhibits and the required browsing time for each exhibit. The required browsing time is used to limit the dwell time on the corresponding exhibit. The multiple exhibits are a combination of exhibits that meet the target interest type and the exhibition hall operation goals.

[0051] The second determining module is used to determine the target viewing route among the plurality of viewing routes;

[0052] The recommendation module is used to recommend a target exhibition route to the current user based on the unique identifier.

[0053] In one possible implementation, the first determining module is used to:

[0054] The virtual exhibition hall's multiple exhibits are tagged to obtain multiple theme tags corresponding to each exhibit;

[0055] Word embedding technology is used to convert the current user's target interest type into an interest feature vector;

[0056] The multiple interest exhibits are determined based on the similarity between the multiple topic tags corresponding to each exhibit and the interest feature vector.

[0057] In one possible implementation, the first determining module is used to:

[0058] For any exhibit, determine the similarity between the interest feature vector and each topic tag corresponding to the exhibit, and perform a weighted summation of the similarity and weight of each topic tag to obtain the interest similarity between the interest feature vector and the exhibit.

[0059] The exhibits with an interest similarity greater than a preset similarity are identified as interest exhibits, thus obtaining multiple interest exhibits.

[0060] In one possible implementation, the analysis and processing module is specifically used for:

[0061] Determine whether the current user's exhibition viewing information is valid information. The exhibition viewing information includes multiple click events of the current user on exhibits in the virtual exhibition hall and the page dwell time corresponding to each click event.

[0062] If so, the target interest type is determined based on the multiple item click events and the page dwell time corresponding to each item click event;

[0063] If not, then the target interest type is determined based on the current user's basic information.

[0064] In one possible implementation, the analysis and processing module is specifically used for:

[0065] The authenticity assessment model is used to evaluate the multiple display item click events and the page dwell time corresponding to each display item click event to obtain the predicted evaluation value of the current user.

[0066] If the predicted evaluation value is greater than or equal to the preset evaluation value, then the current user's exhibition viewing information is valid information;

[0067] If the predicted evaluation value is less than the preset evaluation value, then the current user's viewing information is invalid.

[0068] In one possible implementation, the analysis and processing module is specifically used for:

[0069] Determine the multiple display item theme types corresponding to the multiple display item click events, and the actual click frequency corresponding to each display item theme type;

[0070] Based on the actual click frequency corresponding to each exhibition item theme type, multiple first interest types are determined among the multiple exhibition item theme types;

[0071] Based on the page dwell time corresponding to each item click event, multiple second interest types are determined among the multiple item theme types;

[0072] The exhibit theme type that overlaps with the plurality of first interest types and the plurality of second interest types is determined as the target interest type.

[0073] In one possible implementation, for any given exhibition route, the generation module is specifically used for:

[0074] Determine the number of exhibits corresponding to the exhibition route, wherein the number of exhibits corresponding to the at least one preset exposure exhibit is greater than the number of exposure exhibits corresponding to the at least one preset exposure exhibit.

[0075] The difference between the number of items displayed along the route and the number of items displayed in the exposure section is determined as the number of items of interest.

[0076] The exhibition route is constructed by combining the number of interest items among the multiple interest items with the at least one preset exposure item, and the first item in the exhibition route is the current user's interest item.

[0077] In one possible implementation, the recommendation module is further configured to:

[0078] The system recommends reward information corresponding to the target exhibition route to the current user. The reward information includes multiple exhibition visit quantity ranges and the reward recipients corresponding to each exhibition visit quantity range.

[0079] In one possible implementation, the device further includes an acquisition module, a third determination module, and a fourth determination module:

[0080] The acquisition module is used to acquire multiple browsing exhibits completed by the current user on the target viewing route, as well as the dwell time corresponding to each browsing exhibit;

[0081] The third determining module is used to determine the number of valid exhibits corresponding to the multiple browsing exhibits based on the dwell time corresponding to each browsing exhibit.

[0082] The fourth determining module is used to determine the target reward object corresponding to the current user among the reward objects corresponding to each range of the number of valid exhibits, based on the number of valid exhibits.

[0083] In one possible implementation, the apparatus further includes a route update module, which is further configured to:

[0084] Obtain the current viewing status of the current user, which includes the currently viewed exhibits, the exhibits already viewed, and the viewing duration of each viewed exhibit;

[0085] Based on the current viewing status, determine the current non-interest tag of the current user;

[0086] Identify at least one unviewed exhibit in the target viewing route;

[0087] The browsing position of the unviewed exhibits corresponding to the non-interest tags will be moved to the back, and the corresponding browsing time will be shortened.

[0088] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0089] The memory stores computer-executed instructions;

[0090] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0091] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0092] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0093] The exhibition route determination method, electronic device, storage medium, and program product provided in this application can analyze and process user information to obtain the target interest type corresponding to the current user. Based on the target interest type and at least one preset exposure exhibit, multiple exhibition routes corresponding to the current user are generated. Among the multiple exhibition routes, a target exhibition route is determined and recommended to the current user through a unique identifier. The target exhibition route can combine the target interest type of each user and the exhibits that the current virtual exhibition hall wants to expose, thereby determining the target exhibition route and recommending it to the user. This allows the target exhibition route to both attract user participation and effectively recommend the exhibits that want to be exposed, thus improving the effectiveness of exhibition recommendation. Attached Figure Description

[0094] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0095] Figure 1 This is a schematic diagram illustrating an application scenario provided by an embodiment of this application;

[0096] Figure 2 A flowchart illustrating a method for determining an exhibition route provided in an embodiment of this application;

[0097] Figure 3 A flowchart illustrating another method for determining an exhibition route provided in this application embodiment;

[0098] Figure 4 A schematic diagram of the structure of a device for determining an exhibition route provided in an embodiment of this application;

[0099] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0100] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0101] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0102] The collection, storage, use, processing, transmission, provision, and disclosure of financial data or user data involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0103] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0104] Figure 1 This diagram illustrates an application scenario provided by an embodiment of this application. Please refer to [link / reference]. Figure 1 The application scenarios include user equipment 101 and server 102. The user equipment 101 can install a business system, which includes a virtual exhibition hall. The virtual exhibition hall includes multiple exhibits, such as exhibits corresponding to various financial businesses, personal business exhibits, popular science exhibits, and marketing activity exhibits.

[0105] Users can browse various exhibits in the virtual exhibition hall through user device 101 to learn about or conduct various business activities. Users can click on a displayed exhibit through user device 101 to display the corresponding business interface. Server 102 can provide user device 101 with the display data corresponding to the business system.

[0106] As the pandemic subsided and offline activities gradually resumed, the activity and user engagement of online virtual exhibition halls significantly declined. To boost activity in virtual exhibition halls, technologies often employ reward-based incentives based on behavioral metrics (such as login frequency and online time) to attract user participation. However, when recommending exhibits and planning reward activities, a lack of consideration for user interests means users participate merely to complete tasks and earn rewards, failing to genuinely engage with the exhibits and foster sustained interaction. This results in low effectiveness of exhibition recommendations.

[0107] The method for determining exhibition routes provided in this application can collect user information corresponding to the current user and generate a unique identifier, analyze and process the user information to obtain the target interest type corresponding to the current user, determine multiple interest exhibits corresponding to the current user among multiple exhibits in the virtual exhibition hall, generate multiple exhibition routes based on the multiple interest exhibits and at least one preset exposure exhibit, determine the target exhibition route among the multiple exhibition routes, and recommend the target exhibition route to the current user through the unique identifier.

[0108] During the above execution process, the target viewing route can combine the target interest type of each user and the exhibits that the current virtual exhibition hall wants to expose, thereby determining the target viewing route and recommending it to the user. This allows the target viewing route to not only attract users to participate in the activity, but also effectively recommend the exhibits that want to be exposed, thus improving the effectiveness of the viewing recommendation.

[0109] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0110] Figure 2 This is a flowchart illustrating a method for determining an exhibition route, as provided in an embodiment of this application. Please refer to... Figure 2 The method includes:

[0111] S201. Collect the user information corresponding to the current user and generate a unique identity identifier.

[0112] The execution entity in this application embodiment can be a server or a viewing route determination device installed in the server. The viewing route determination device can be implemented by software or by a combination of software and hardware.

[0113] User information corresponding to the current user can be collected through event tracking. Event tracking is a technical means of tracking and recording user behavior data within the virtual exhibition hall by pre-embedding monitoring code (i.e., "event tracking") into the virtual exhibition hall code.

[0114] Performing data collection via event tracking can include the following three steps:

[0115] First, we can clarify which user behaviors need to be collected (such as entry time into the exhibition hall, exhibit click events, dwell time, visit routes, etc.), and define the monitoring dimensions for each behavior. For example: exhibit click events: we need to record the exhibit ID clicked, the click time, and the user's identity at the time of the click; dwell time: we need to record the time of entering the exhibit page and the time of leaving the exhibit page, and the difference between the two is the dwell time.

[0116] Secondly, monitoring code (usually snippets of languages ​​such as JavaScript, Java, and Python) is embedded in the corresponding functional modules of the virtual exhibition hall. This code is activated when the user triggers specific behaviors. For example, when a user opens the virtual exhibition hall, the tracking code automatically records the entry time and sends it to the backend server; when a user clicks on the icon of an exhibit, the tracking code is triggered and reports the exhibit ID and click time; when a user switches from exhibit A to exhibit B, the tracking code calculates and reports the dwell time at exhibit A, and simultaneously records the visitor's route (A→B).

[0117] Finally, the behavioral data collected by the tracking code (such as time, event type, user ID, display item information, etc.) will be sent to the backend server in real time or in batches via network requests (such as HTTP / HTTPS). The server will store the data in a database (such as MySQL, Hadoop, etc.) to form a structured user behavior log.

[0118] A unique identifier can be used to indicate the user's identity.

[0119] User information can include the current user's basic information and viewing information in the virtual exhibition hall. Basic information may include age, gender, and asset status. Viewing information may include the current user's click events on multiple exhibits in the virtual exhibition hall, the duration of page dwell for each click event, the viewing route, and exhibit theme information.

[0120] S202. Analyze and process the user information to obtain the target interest type corresponding to the current user.

[0121] User information can include basic information and exhibition viewing information. However, when browsing exhibits in a virtual exhibition hall, users may complete viewing tasks arbitrarily to meet reward thresholds, which may not effectively reflect their interests. Therefore, it is necessary to determine whether the current user's exhibition viewing information is valid.

[0122] For example, a user can still receive a reward even if they log in and then log out shortly afterward, even though they have met the login requirement but haven't visited any exhibits. Alternatively, a user who is logged in but in an AFK state can also receive a reward; although such a user is online for a long time, they remain in the same location for an extended period, resulting in fewer exhibits visited.

[0123] Specifically, it can determine whether the current user's exhibition viewing information is valid; if so, it can determine the target interest type based on multiple exhibition item click events and the page dwell time corresponding to each exhibition item click event; if not, it can determine the target interest type based on the current user's basic information.

[0124] In this application, the accuracy of determining the target interest type can be improved by determining whether the current user's exhibition viewing information is valid, thereby reducing the impact of invalid information on the determination of the current user's target interest type. Furthermore, the user's basic information can be used to supplement the determination of interest characteristics for users without valid exhibition viewing information, thus improving the reliability of determining the target interest type.

[0125] S203. Based on the target interest type, determine the multiple interest items corresponding to the current user among the multiple exhibits in the virtual exhibition hall.

[0126] In some embodiments, multiple exhibits in the virtual exhibition hall can be tagged to obtain multiple topic tags corresponding to each exhibit; word embedding technology is used to convert the current user's target interest type into an interest feature vector; and multiple interest exhibits are determined based on the similarity between the multiple topic tags corresponding to each exhibit and the interest feature vector.

[0127] In this application, multiple exhibits are tagged, and the target interest type is converted into an interest feature vector. Then, multiple interest exhibits are identified from multiple exhibits by measuring the similarity between multiple topic tags and interest feature vectors. By capturing the matching relationship between the core user interest and the topic of the exhibits, the subjectivity and limitations of traditional manual screening are broken, and intelligent and personalized matching of exhibit recommendations is achieved. This improves the user's browsing efficiency, interest matching, and willingness to participate in interaction in the virtual exhibition hall.

[0128] S204. Generate multiple viewing routes based on multiple interest exhibits and at least one preset exposure exhibit.

[0129] The viewing route can be used to guide the current user to browse the exhibits in the virtual exhibition hall. The viewing route includes multiple exhibits and the corresponding viewing time for each exhibit.

[0130] The browsing time limit is used to restrict the time spent on the corresponding exhibits. Multiple exhibits are combinations of exhibits that match the target interest type and the exhibition hall's operational goals.

[0131] Preset exposure exhibits are those exhibits for which the exhibition hall's operational goals aim to increase exposure.

[0132] Among the various exhibits in the virtual exhibition hall, there are both interest-based exhibits that align with the target audience's interests and pre-selected exposure exhibits designed to increase the hall's visibility and meet the exhibition hall's operational goals. Therefore, it is possible to construct a viewing route that both sparks user interest and promotes the pre-selected exposure exhibits.

[0133] In this application, multiple viewing routes that can arouse user interest and promote the preset exposure items can be generated by targeting interest types and preset exposure items, thereby improving the effectiveness of viewing recommendations.

[0134] S205. Determine the target viewing route from multiple viewing routes and recommend the target viewing route to the current user through a unique identifier.

[0135] In some possible embodiments, the interest matching degree, exposure achievement rate and route flow degree corresponding to each exhibition route can be determined. Based on the interest matching degree, exposure achievement rate and route flow degree corresponding to each exhibition route, the evaluation weight corresponding to each exhibition route can be determined. The exhibition routes with evaluation weights greater than preset weights can be determined as target exhibition routes.

[0136] If there is no viewing route with an evaluation weight greater than the preset weight, then a new target viewing route will be generated, or the viewing route with the highest evaluation weight will be determined as the target viewing route.

[0137] The proportion of exhibits of interest in the exhibition route can be determined (e.g., the number of exhibits of interest / the total number of exhibits in the route), and the predicted average dwell time of exhibits of interest (calculated based on users' historical behavior data). The higher the proportion and the longer the predicted dwell time, the higher the degree of matching between the exhibition route and users' interests.

[0138] Exposure achievement rate can be used to indicate the display effect of pre-selected exposure exhibits along the exhibition route. Specifically, it can include the location of the pre-selected exposure exhibits (such as whether they are located in the first 3 exhibits of the route to avoid users leaving midway and missing them) and the relevance of the pre-selected exposure exhibits to the preceding and following interest exhibits (such as placing the "green credit exhibit" next to the "personal finance exhibit" that users are interested in, using interest migration to improve the effectiveness of exposure).

[0139] Regarding the flow of the route, we can assess whether the jump logic of the exhibits in the exhibition route is natural (such as whether the connection from "credit card exhibit" to "installment discount exhibit" conforms to the user's decision path) and whether the total number of exhibits is reasonable (to avoid the route being too long and causing user fatigue, usually controlled at 5-8 exhibits).

[0140] For any exhibition route, when determining the evaluation weight based on the interest matching degree, exposure achievement rate and route flow degree corresponding to the exhibition route, the sum of the products of interest matching degree and its corresponding first weight, exposure achievement rate and its corresponding second weight, and route flow degree and its corresponding third weight can be determined as the evaluation weight.

[0141] In this system, the sum of the first, second, and third weights equals 1. Generally, the first weight is greater than the second weight, and the second weight is greater than the third weight. For example, the first weight could be 0.5, the second weight could be 0.3, and the fourth weight could be 0.2.

[0142] After locating the current user through a unique identifier, a pop-up window can be displayed on the virtual exhibition hall homepage showing "A customized exhibition route for you," or a "Next Recommendation" button can be added to the bottom of the exhibition page the user is currently browsing.

[0143] When displaying routes, present them as a visual route map (such as connecting the exhibit icons in sequence, labeling them "Exhibits you may be interested in" and "Hot Recommended Exhibits"), along with a brief description (such as "This route includes financial knowledge and the latest credit card offers that you are interested in").

[0144] You can set rewards for completing a route (e.g., earning points for viewing all exhibits), but the reward threshold is linked to the depth of the exhibit content (e.g., requiring a full minute to complete an exhibit, to avoid skipping over it quickly for the reward).

[0145] The method for determining exhibition routes provided in this application can analyze and process user information to obtain the target interest type corresponding to the current user. Based on the target interest type and at least one preset exposure exhibit, multiple exhibition routes corresponding to the current user are generated. Among the multiple exhibition routes, a target exhibition route is determined and recommended to the current user through a unique identifier. The target exhibition route can combine each user's target interest type and the exhibits that the current virtual exhibition hall wants to expose, thereby determining the target exhibition route and recommending it to the user. This allows the target exhibition route to both attract user participation and effectively recommend the exhibits that want to be exposed, thus improving the effectiveness of exhibition recommendation.

[0146] Figure 3 A flowchart illustrating another method for determining an exhibition route provided in this application embodiment. Please refer to... Figure 3 The method may include:

[0147] S301. Collect the user information corresponding to the current user and generate a unique identity identifier.

[0148] The execution process of S301 can be found in the execution process of S201, and will not be repeated here.

[0149] S302. Determine whether the current user's exhibition viewing information is valid.

[0150] In some embodiments, if the number of item clicks in multiple item click events is greater than a preset number of clicks, then an excessive dwell time number is determined among the multiple item click events. The excessive dwell time number can indicate the number of item click events where the page dwell time is greater than a preset time. If the ratio of the excessive dwell time number to the number of item clicks is less than or equal to a preset ratio, then the current user's viewing information is determined to be valid information.

[0151] For example, assuming a preset click count of 5, a preset duration of 10 minutes, and a preset ratio of 30%, if the current user clicks on an item 8 times (greater than the preset click count of 5), then the number of instances where the page stay exceeds 10 minutes will be counted. If there are 2 instances of excessively long stays, the ratio of excessively long stays to the number of item clicks is 25%, which is less than the preset ratio of 30%, indicating that most of the user's clicks are normal browsing, and the current user's viewed information can be determined as valid information. If there are 3 instances of excessively long stays, the ratio is 37.5%, which is greater than the preset ratio of 30%, indicating that the user has engaged in a lot of abnormal behavior of staying on the page for extended periods (such as idling), and the viewed information will be determined as invalid information.

[0152] If the number of clicks on multiple exhibits is less than the preset number, the current user may have accidentally clicked or there may be other accidental circumstances. This does not reflect the current user's interest characteristics, and the current user's viewing information can be determined as invalid information.

[0153] When browsing web pages, users typically allocate their time based on content relevance, avoiding spending too much time on irrelevant pages or too little on pages of interest. If a user only briefly clicks on an item without viewing the content, or stays on a single item page for an extended period (e.g., while the computer is idle), it doesn't accurately reflect their true interest.

[0154] If a user stays on a page for a longer than the preset duration after clicking on a certain item, based on the actual browsing scenario (such as no scrolling or subsequent operations), the click event may be invalid and does not reflect the user's interest.

[0155] If a user's multiple click events on exhibits result in a large number of prolonged pauses, it indicates that the user is likely in a non-active browsing state. The exhibit information does not reflect the user's current interests, and therefore, the exhibit information can be determined to be invalid.

[0156] In some embodiments, an authenticity assessment model can be used to evaluate multiple display item click events and the page dwell time corresponding to each display item click event to obtain the current user's predicted evaluation value; if the predicted evaluation value is greater than or equal to the preset evaluation value, the current user's viewing information is valid information; if the predicted evaluation value is less than the preset evaluation value, the current user's viewing information is invalid information.

[0157] The authenticity assessment model is a machine learning model trained on sample data, such as logistic regression model, random forest model, lightweight neural network model, etc.

[0158] By inputting multiple display item click events and the corresponding page dwell time for each display item click event into the authenticity assessment model, a predicted assessment value can be obtained.

[0159] Predictive evaluation values ​​can be used to quantify the effectiveness of exhibition information. The value range is usually 0-100 points or 0-1. The higher the score, the more accurately the exhibition information reflects the user's interests.

[0160] For example, assuming the authenticity assessment model is a random forest model, the preset assessment value is 60 points (out of 100). If the current user clicks on the exhibits 7 times, spends 2-5 minutes on each exhibit, and follows a browsing route of "Homepage, Personal Finance, Mortgage Consultation, Credit Card Offers" (logically coherent, conforming to the user's browsing habits from "Asset Planning" to "Specific Needs"), inputting this data into the model yields a predicted assessment value of 78 points. This score is greater than the preset assessment value of 60 points, and the viewed information is deemed valid. However, if the user clicks on the exhibits 5 times, spends more than 8 minutes on 3 of them, and follows a browsing route of "Homepage, ..., Corporate Loans, Homepage, Green Finance" (a chaotic route with frequent returns to the homepage), the model outputs a predicted assessment value of 42 points, which is less than the preset assessment value of 60 points, and the viewed information is deemed invalid.

[0161] In this application, the predicted evaluation value is quantitatively output through the authenticity assessment model, and invalid information is accurately filtered by comparing the predicted evaluation value with the preset evaluation value, which can improve the reliability of determining valid information.

[0162] S303. If so, determine the target interest type based on multiple item click events and the page dwell time corresponding to each item click event.

[0163] In some embodiments, multiple exhibit theme types corresponding to multiple exhibit click events and the actual click frequency corresponding to each exhibit theme type are determined; based on the actual click frequency corresponding to each exhibit theme type, multiple first interest types are determined among the multiple exhibit theme types; based on the page dwell time corresponding to each exhibit click event, multiple second interest types are determined among the multiple exhibit theme types; exhibit theme types that overlap with multiple first interest types and multiple second interest types are determined as target interest types.

[0164] The first interest type indicates the type of exhibits that users tend to follow based on their active clicking behavior; the second interest type indicates the type of exhibits that users tend to follow based on their deep browsing behavior (stay time); the target interest type indicates the type of exhibits that simultaneously conforms to users' clicking and deep browsing tendencies and can truly reflect users' core interests.

[0165] Specifically, the first interest type can be determined as follows: determine the exposure base corresponding to each exhibit theme type; for any exhibit theme type, determine the corrected click frequency corresponding to that exhibit theme type by the ratio of the exposure base corresponding to that exhibit theme type to the actual click frequency; sort the exhibit theme types according to the corrected click frequency, and determine the top N exhibit theme types as the first interest type.

[0166] Where N is an integer greater than or equal to 1. For example, N can be 3.

[0167] Exposure base is used to indicate the total number of times a certain type of exhibit in a virtual exhibition hall is seen by users within a statistical period (e.g., 1 day / 1 visit). For example, the exposure base for the "Credit Card Offers" exhibit recommended on the homepage is 500 times, while the exposure base for the "Corporate Finance" exhibit on the secondary page is 100 times.

[0168] If a certain type of exhibit has a high exposure rate in the virtual exhibition hall (such as a homepage recommendation), the weight of its click frequency should be reduced (to avoid passive clicks caused by strong exposure interfering with interest judgment); conversely, if a certain type of exhibit is hidden on a secondary page (requiring active search to find), the weight of its click frequency should be increased (considered as being driven by user active interest).

[0169] For example, suppose the display topic types are A (credit card offers, exposure base 500), B (personal finance, exposure base 200), and C (business loans, exposure base 100). The actual click frequency for user A is 50 times, for B it's 30 times, and for C it's 20 times. Calculate the corrected click frequency: A is 50 / 500 = 0.1, B is 30 / 200 = 0.15, and C is 20 / 100 = 0.2. Sorted by corrected click frequency, C (0.2) > B (0.15) > A (0.1). If N = 2, then the first interest type is C, then B.

[0170] Specifically, the second interest type can be determined as follows: Based on the page dwell time corresponding to each item click event, determine the relative dwell time corresponding to each item click event; for any item theme type, determine the average of the relative dwell times of the item click events corresponding to that item theme type as the aggregate dwell time of that item theme type; based on the aggregate dwell time, sort the multiple item theme types, and determine the top M item theme types as the second interest type.

[0171] Where M is an integer greater than or equal to 1. For example, M can be 3. M and N can be the same or different.

[0172] When determining the relative dwell time of a click event for a certain display item, the dwell time of each display item can be compared with the average reasonable browsing time of that display item (pre-calculated based on full user data, such as complex financial product display items requiring an average of 4 minutes and simple coupon display items requiring an average of 1 minute) to calculate the relative dwell time (e.g., actual user dwell time / average reasonable time).

[0173] For example, suppose the display topic types are A (credit card offers, average reasonable duration 1 minute), B (personal finance, average reasonable duration 4 minutes), and C (business loans, average reasonable duration 3 minutes). The user's three clicks on A resulted in dwell times of 0.8, 1.2, and 1.0 minutes respectively, with relative dwell times of 0.8, 1.2, and 1.0, and an aggregate dwell time (average) of 1.0. The two clicks on B resulted in dwell times of 5.0 and 3.8 minutes respectively, with relative dwell times of 1.25 and 0.95, and an aggregate dwell time of 1.1. The one click on C resulted in a dwell time of 4.5 minutes, with a relative dwell time of 1.5 and an aggregate dwell time of 1.5. The aggregate dwell time ranking is C (1.5) > B (1.1) > A (1.0). If M=2, then the second interest type is C and B.

[0174] In this application, by using the actual click frequency corresponding to each exhibition item theme type and the page dwell time corresponding to each exhibition item click event, the target interest type can be determined among multiple exhibition item theme types, which can improve the accuracy of determining the target interest type and thus improve the effectiveness of exhibition recommendation.

[0175] S304. If not, determine the target interest type based on the current user's basic information.

[0176] Based on the current user's basic information, the target user type of the current user can be determined, and the target interest type corresponding to the current user can be located through the target user type.

[0177] Based on the key dimensions of basic information, multi-dimensional user types can be identified. For example, if the user type is students and general customers, the corresponding interest types are campus credit cards (no annual fee), fixed-deposit savings, and low-risk money market funds; if the user type is married homeowners and general customers, the corresponding interest types are mortgage interest rate discounts, home renovation loans, and comprehensive family financial management.

[0178] After obtaining the basic information of the current user, it is compared with multiple user types one by one to determine the unique or most suitable target user type.

[0179] Then, based on the matched target user type, the corresponding interest type is retrieved from the "user type-interest type" mapping library and used as the target interest type for the current user.

[0180] S305. Tag multiple exhibits in the virtual exhibition hall to obtain multiple theme tags corresponding to each exhibit.

[0181] Multiple topic tags can be core topic tags, secondary topic tags, and feature tags.

[0182] For example, the tags for a certain exhibit, "Supermarket Co-branded Credit Card Cashback," are set as follows: core theme tag "Financial Services - Credit Card Offers," secondary theme tag "Cashback - Supermarket Shopping," and feature tag "Offer Calculation - Card Application Guide."

[0183] All labels can be included in the exhibition hall label database and maintained regularly based on the exhibits.

[0184] S306. Using word embedding technology, the target interest type of the current user is converted into an interest feature vector.

[0185] Word embedding technology can be used to convert users' target interest types into high-dimensional interest feature vectors.

[0186] For example, after the target interest type "supermarket credit card - high frequency cashback" is converted into a vector, it contains numerical representations of core features such as "supermarket credit card", "cashback", "daily purchases", "high cashback ratio" and "co-branded benefits", ensuring the computability of interest types.

[0187] S307. Based on the similarity between the multiple theme tags and interest feature vectors corresponding to each exhibit, determine multiple interest exhibits.

[0188] The cosine similarity algorithm can be used as the core matching algorithm (it is suitable for similarity calculation of high-dimensional vectors, has high computational efficiency and stable results), and the target interest type feature vector and the topic label feature vector of each exhibit can be associated and matched.

[0189] In some embodiments, for any given exhibit, the similarity between the interest feature vector and each topic tag corresponding to the exhibit is determined, and the similarity and weight of each topic tag are weighted and summed to obtain the interest similarity between the interest feature vector and the exhibit; exhibits with interest similarity greater than a preset similarity are determined as interest exhibits, so as to obtain multiple interest exhibits.

[0190] Specifically, for any exhibit, the following steps can be taken: first, determine the first similarity between the target interest type vector and the exhibit's core topic tag vector; second, determine the second similarity between the target interest type vector and the exhibit's secondary topic tag vector; third, determine the third similarity between the target interest type vector and the exhibit's feature tag vector; first, determine the first product of the first similarity and its corresponding fourth weight; second, determine the second product of the second similarity and its corresponding fifth weight; third, determine the third product of the third similarity and its corresponding sixth weight; and finally, determine the interest similarity of the exhibit by summing the first, second, and third products.

[0191] The fourth weight is the weight of the core topic tag, the fifth weight is the weight of the secondary topic tag, and the sixth weight is the weight of the feature tag. The sum of the fourth, fifth, and sixth weights is 1.

[0192] For example, the fourth weight can be 70%, the fifth weight can be 20%, and the sixth weight can be 10%.

[0193] The preset similarity can be 80%, which can be adjusted by the exhibition hall operator according to business needs. For example, it can be reduced to 70% for niche interest users and increased to 85% for precise recommendation scenarios.

[0194] Items with an interest similarity that reaches or exceeds the preset similarity can be officially identified as the current user's interest items.

[0195] In some embodiments, if the number of interest-related exhibits matched by interest similarity is less than the minimum recommended threshold (e.g., 3), the preset threshold is automatically reduced by 5%, and the selection is re-filtered until the quantity requirement is met; if the number of matched interest-related exhibits exceeds the maximum recommended threshold (e.g., 20), the exhibits are sorted from largest to smallest according to interest similarity, and the exhibits with the highest recommended threshold number are taken as the final set of interest-related exhibits.

[0196] In this application, similarity can be used to filter the current user's interest items, which can improve the accuracy and suitability of interest item recommendations. At the same time, by dynamically adjusting the threshold and controlling the number threshold, the effectiveness (avoiding too few or too many) and stability of the recommendation results are taken into account, thereby improving the user's recognition of the virtual exhibition hall's recommended content, browsing depth and retention time, and optimizing the efficiency of precise reach of exhibition hall resources.

[0197] S308. Generate multiple viewing routes based on multiple interest exhibits and at least one preset exposure exhibit.

[0198] In some embodiments, for any given exhibition route, the number of route exhibits is determined, where the number of route exhibits is greater than the number of exposure exhibits corresponding to at least one preset exposure exhibit. The difference between the number of route exhibits and the number of exposure exhibits is determined as the number of interest exhibits. The number of interest exhibits among the multiple interest exhibits is combined with at least one preset exposure exhibit to construct the exhibition route. The first exhibit on the exhibition route is the current user's interest exhibit.

[0199] The number of exhibits on a route refers to the total number of exhibits included in a single viewing route. This number needs to be preset based on the average viewing time of users and the complexity of the exhibit content. For example, it can be set to 5-8 to avoid routes that are too long, causing users to quit midway, or routes that are too short, failing to cover the core exhibits.

[0200] The number of exhibits to be exposed refers to the number of pre-set exhibits that must be included in a single exhibition route, determined by the exhibition hall's operational goals. For example, if two pre-set exhibits are to be promoted in a given month, then the number of exhibits to be exposed is set to 2.

[0201] Furthermore, when constructing exhibition routes, the principle of exhibit relevance must be followed: when combining exhibits, it is necessary to ensure that the themes of adjacent exhibits are logically related (e.g., from "personal finance interest exhibits to retirement finance preset exposure exhibits" or "credit card interest exhibits to installment discount preset exposure exhibits"), to avoid confusing exhibit transitions that could lead to a decline in user experience. At the same time, by "interspersing preset exposure exhibits with interest exhibits" (e.g., interest exhibits, preset exposure exhibits, interest exhibits, interest exhibits, preset exposure exhibits), user interests and operational goals can be balanced, further improving route acceptance.

[0202] In this application, by first guiding users with interest-based exhibits and then combining interest-based and exposure-based exhibits according to their relevance, it is possible to ensure that the viewing route matches the user's interests to improve the browsing completion rate, while also ensuring that the preset exposure exhibits effectively reach the user, thereby increasing the user acceptance of the viewing route and the exposure conversion rate of the preset exposure exhibits.

[0203] S309. Determine the target viewing route from multiple viewing routes and recommend the target viewing route to the current user through a unique identifier.

[0204] The execution process of S309 can be found in the execution process of S205, and will not be repeated here.

[0205] S310. Recommend reward information corresponding to the target exhibition route to the current user.

[0206] The reward information may include multiple ranges of the number of exhibition visits, as well as the reward recipients for each range of the number of exhibition visits.

[0207] The exhibition visit range can be divided according to the total number of exhibits in the target exhibition route, which is the range of the number of valid exhibits that the user needs to visit. For example, if the route contains 5 exhibits, it can be divided into three ranges: 1-2 valid exhibits, 3-4 valid exhibits, and 5 valid exhibits.

[0208] The reward recipients refer to the incentive content within the corresponding range, which needs to be set in combination with user needs and operational goals. Common forms include financial benefits (such as wealth management interest rate coupons, credit card points), service benefits (such as free account diagnosis, priority processing channels), and physical / virtual gifts (such as mobile phone top-ups, bank-related products). In addition, the reward value should follow the tiered principle that the higher the range, the higher the reward value, to guide users to browse more deeply.

[0209] After the current user completes the exhibition viewing along the target route, the target reward recipients need to be determined according to the following steps. The specific execution process is as follows: obtain the multiple exhibition items that the current user has viewed along the target route, as well as the dwell time corresponding to each exhibition item; determine the number of valid exhibition items corresponding to the multiple exhibition items based on the dwell time corresponding to each exhibition item; and determine the target reward recipients corresponding to the current user from the reward recipients corresponding to each exhibition viewing range based on the number of valid exhibition items.

[0210] To validate the validity of each browsing item, two conditions must typically be met simultaneously: First, the user's actual dwell time must be greater than or equal to the minimum valid browsing time for that item; second, the user must engage in active interaction on the item's page, such as clicking the "View Details" button or swiping through the text and images, thus excluding idle browsing.

[0211] Exhibits that meet the criteria can be recorded as valid exhibits, and the total number of all valid exhibits can be counted, i.e., the number of valid exhibits.

[0212] The system matches the number of valid exhibits with the range of exhibit visits in the reward information. Once the user's range is determined, the corresponding reward object for that range is retrieved, which is the target reward object for the current user.

[0213] Simultaneously, the system must push the target reward recipients to the user's device in real time (e.g., through a virtual exhibition hall pop-up or bank app notification) using the user's unique identifier (such as an account ID), and clearly define the reward's claim method and usage rules. Claim methods can include automatic deposit or click-to-claim. Usage rules can specify the validity period of interest rate coupons, points redemption paths, etc.

[0214] In this application, a tiered reward mechanism centered on the number of valid exhibits can be used to avoid users' ineffective behaviors such as quickly refreshing exhibits or idling around in order to obtain rewards. This guides users to browse exhibits in depth along the target viewing route, while high-value rewards can enhance users' enthusiasm for participating in the viewing activities, thereby increasing the completion rate of the target viewing route and the effective reach rate of the preset exposure exhibits.

[0215] In some embodiments, after recommending a target viewing route to the user, the current viewing status of the user can be obtained. The current viewing status includes the currently viewed exhibits, the exhibits already viewed, and the viewing time of each viewed exhibit. Based on the current viewing status, the current non-interest tags of the user are determined. At least one unviewed exhibit in the target viewing route is determined. The viewing position of the unviewed exhibits corresponding to the non-interest tags is moved to the back, and the corresponding viewing time is shortened.

[0216] During the exhibition, the viewing route can be dynamically adjusted in real time using reinforcement learning models (such as the DQN algorithm).

[0217] The reinforcement learning model will adjust the order of the next recommended exhibits and change the required duration of the user's stay on the current exhibit based on the user's current viewing status.

[0218] When the system detects that a user spends a short time in front of a credit card display item and shows signs of disinterest, the reinforcement learning model will use the DQN algorithm, combined with previous experience and current environmental information, to re-evaluate the priority of subsequent display items.

[0219] If the model detects that users show less interest in installment payment options but more interest in points redemption options, it will adjust the viewing route, recommending points redemption options to users in advance, and appropriately shortening the required time users spend in the installment payment options area to improve user engagement and satisfaction.

[0220] If users follow the adjusted route, spend more time on subsequent exhibits, and engage in more interactive activities, the reinforcement learning model will provide a positive reward; conversely, if users are not interested in the adjusted route and their participation decreases, the system will provide a negative reward.

[0221] In this application, through continuous trial and error and learning, the reinforcement learning model gradually finds the optimal strategy, namely the order of exhibits and dwell time requirements that can maximize user engagement and the achievement rate of exhibition hall operation goals.

[0222] The method for determining exhibition routes provided in this application can analyze and process user information to obtain the target interest type corresponding to the current user. Based on the target interest type and at least one preset exposure exhibit, multiple exhibition routes corresponding to the current user are generated. Among the multiple exhibition routes, a target exhibition route is determined and recommended to the current user through a unique identifier. The target exhibition route can combine each user's target interest type and the exhibits that the current virtual exhibition hall wants to expose, thereby determining the target exhibition route and recommending it to the user. This allows the target exhibition route to both attract user participation and effectively recommend the exhibits that want to be exposed, thus improving the effectiveness of exhibition recommendation.

[0223] Figure 4 This is a schematic diagram of a device for determining an exhibition route, provided in an embodiment of this application. Please refer to [link / reference]. Figure 4 The device 400 for determining the exhibition route may include a data acquisition module 401, an analysis and processing module 402, a first determination module 403, a generation module 404, a second determination module 405, and a recommendation module 406.

[0224] The data collection module 401 is used to collect user information corresponding to the current user and generate a unique identity identifier. The user information includes the current user's basic information and the viewing information of the virtual exhibition hall.

[0225] The analysis and processing module 402 is used to analyze and process user information to obtain the target interest type corresponding to the current user;

[0226] The first determining module 403 is used to determine, based on the target interest type, multiple interest items corresponding to the current user among multiple exhibits in the virtual exhibition hall;

[0227] The generation module 404 is used to generate multiple viewing routes based on multiple interest exhibits and at least one preset exposure exhibit. The viewing routes are used to guide the current user to browse exhibits in the virtual exhibition hall. The viewing routes include multiple exhibits and the required browsing time for each exhibit. The required browsing time is used to limit the dwell time on the corresponding exhibit. The multiple exhibits are a combination of exhibits that meet the target interest type and the exhibition hall operation goals.

[0228] The first determining module 405 is used to determine the target viewing route among multiple viewing routes;

[0229] The recommendation module 406 is used to recommend a target exhibition route to the current user through a unique identifier.

[0230] In one possible implementation, the first determining module 403 is used to:

[0231] The virtual exhibition hall's multiple exhibits are tagged to obtain multiple theme tags corresponding to each exhibit;

[0232] Word embedding technology is used to convert the current user's target interest type into an interest feature vector;

[0233] Multiple interest exhibits are determined based on the similarity between the multiple theme tags and interest feature vectors corresponding to each exhibit.

[0234] In one possible implementation, the first determining module 403 is used to:

[0235] For any exhibit, determine the similarity between the interest feature vector and each topic tag corresponding to the exhibit, and then perform a weighted sum of the similarity and weight of each topic tag to obtain the interest similarity between the interest feature vector and the exhibit.

[0236] Items with an interest similarity greater than a preset similarity are identified as interest items, thus obtaining multiple interest items.

[0237] In one possible implementation, the analysis and processing module 402 is specifically used for:

[0238] Determine whether the current user's exhibition viewing information is valid. The exhibition viewing information includes the current user's click events on multiple exhibits in the virtual exhibition hall and the page dwell time corresponding to each click event.

[0239] If so, the target interest type is determined based on multiple item click events and the page dwell time corresponding to each item click event;

[0240] If not, determine the target interest type based on the current user's basic information.

[0241] In one possible implementation, the analysis and processing module 402 is specifically used for:

[0242] The authenticity assessment model is used to evaluate multiple display item click events and the page dwell time corresponding to each display item click event to obtain the predicted evaluation value of the current user.

[0243] If the predicted evaluation value is greater than or equal to the preset evaluation value, then the current user's exhibition viewing information is valid.

[0244] If the predicted evaluation value is less than the preset evaluation value, then the current user's exhibition viewing information is invalid.

[0245] In one possible implementation, the analysis and processing module 402 is specifically used for:

[0246] Determine the multiple display item theme types corresponding to multiple display item click events, and the actual click frequency corresponding to each display item theme type;

[0247] Based on the actual click frequency corresponding to each exhibition item theme type, several primary interest types are determined among multiple exhibition item theme types;

[0248] Based on the page dwell time corresponding to each exhibit click event, multiple secondary interest types are determined among multiple exhibit theme types;

[0249] For exhibits whose themes overlap with multiple primary interest types and multiple secondary interest types, the target interest type is determined.

[0250] In one possible implementation, for any given exhibition route, the generation module 404 is specifically used for:

[0251] Determine the number of exhibits corresponding to the exhibition route, ensuring that the number of exhibits along the route is greater than the number of exhibits corresponding to at least one preset exposure exhibit.

[0252] The difference between the number of exhibits along the route and the number of exhibits that are exposed is determined as the number of exhibits of interest.

[0253] The exhibition route is constructed by combining the number of interest-related exhibits from multiple interest-related exhibits with at least one preset exposure exhibit. The first exhibit in the exhibition route is the current user's interest-related exhibit.

[0254] In one possible implementation, the recommendation module 406 is further configured to:

[0255] Recommend reward information corresponding to the target exhibition route to the current user. The reward information includes multiple exhibition visit ranges and the reward recipients for each exhibition visit range.

[0256] In one possible implementation, the apparatus further includes an acquisition module, a third determination module, and a fourth determination module:

[0257] The acquisition module is used to acquire the multiple exhibition items that the current user has completed for the target exhibition route, as well as the dwell time for each exhibition item.

[0258] The third determination module is used to determine the number of valid exhibits corresponding to multiple exhibits based on the dwell time corresponding to each exhibit.

[0259] The fourth determination module is used to determine the target reward recipient for the current user among the reward recipients corresponding to each range of valid exhibits.

[0260] In one possible implementation, the device 400 further includes a route update module, which is also used for:

[0261] Get the current user's current viewing status, which includes the currently viewed exhibits, the exhibits already viewed, and the viewing time for each of the viewed exhibits;

[0262] Based on the current viewing status, determine the current user's current non-interest tags;

[0263] Identify at least one unviewed exhibit in the target exhibition route;

[0264] Move the browsing position of unviewed exhibits corresponding to non-interest tags to the back and shorten the browsing time required for each.

[0265] The device for determining the exhibition route provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0266] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Please refer to... Figure 5 The electronic device 500 may include a processor 501 and a memory 502. Exemplarily, the processor 501 and the memory 502 are interconnected via a bus 503.

[0267] Memory 502 stores instructions executed by the computer;

[0268] The processor 501 executes computer execution instructions stored in the memory 502, causing the processor 501 to perform the off-table information update method as shown in the above method embodiment.

[0269] Accordingly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method for determining the exhibition route in the above-described method embodiments.

[0270] Accordingly, embodiments of this application may also provide a computer program product, including a computer program, which, when executed by a processor, can implement the method for determining the exhibition route shown in the above method embodiments.

[0271] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0272] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0273] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0274] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0275] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0276] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0277] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0278] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0279] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for determining an exhibition route, characterized in that, include: Collect user information corresponding to the current user and generate a unique identity identifier. The user information includes the current user's basic information and the virtual exhibition hall's viewing information. The user information is analyzed and processed to obtain the target interest type corresponding to the current user; Based on the target interest type, determine the multiple interest items corresponding to the current user from among the multiple exhibits in the virtual exhibition hall; Based on the multiple interest exhibits and at least one preset exposure exhibit, multiple viewing routes are generated. The viewing routes are used to guide the current user to browse exhibits in the virtual exhibition hall. The viewing routes include multiple exhibits and the required browsing time for each exhibit. The required browsing time is used to limit the dwell time on the corresponding exhibit. The multiple exhibits are a combination of exhibits that meet the target interest type and the exhibition hall operation goals. The target viewing route is determined from the multiple viewing routes, and recommended to the current user through the unique identifier.

2. The method according to claim 1, characterized in that, Based on the target interest type, from the multiple exhibits in the virtual exhibition hall, determine multiple interest exhibits corresponding to the current user, including: The virtual exhibition hall's multiple exhibits are tagged to obtain multiple theme tags corresponding to each exhibit; Word embedding technology is used to convert the current user's target interest type into an interest feature vector; The multiple interest exhibits are determined based on the similarity between the multiple topic tags corresponding to each exhibit and the interest feature vector.

3. The method according to claim 2, characterized in that, The multiple interest exhibits are determined based on the similarity between the multiple topic tags corresponding to each exhibit and the interest feature vector, including: For any exhibit, determine the similarity between the interest feature vector and each topic tag corresponding to the exhibit, and perform a weighted summation of the similarity and weight of each topic tag to obtain the interest similarity between the interest feature vector and the exhibit. The exhibits with an interest similarity greater than a preset similarity are identified as interest exhibits, thus obtaining multiple interest exhibits.

4. The method according to claim 1, characterized in that, The user information is analyzed and processed to obtain the target interest type corresponding to the current user, including: Determine whether the current user's exhibition viewing information is valid information. The exhibition viewing information includes multiple click events of the current user on exhibits in the virtual exhibition hall and the page dwell time corresponding to each click event. If so, the target interest type is determined based on the multiple item click events and the page dwell time corresponding to each item click event; If not, then the target interest type is determined based on the current user's basic information.

5. The method according to claim 4, characterized in that, Determining whether the current user's viewing information is valid includes: The authenticity assessment model is used to evaluate the multiple display item click events and the page dwell time corresponding to each display item click event to obtain the predicted evaluation value of the current user. If the predicted evaluation value is greater than or equal to the preset evaluation value, then the current user's exhibition viewing information is valid information; If the predicted evaluation value is less than the preset evaluation value, then the current user's viewing information is invalid.

6. The method according to claim 4, characterized in that, Based on the multiple item click events and the page dwell time corresponding to each item click event, the target interest type is determined, including: Determine the multiple display item theme types corresponding to the multiple display item click events, and the actual click frequency corresponding to each display item theme type; Based on the actual click frequency corresponding to each exhibition item theme type, multiple first interest types are determined among the multiple exhibition item theme types; Based on the page dwell time corresponding to each item click event, multiple second interest types are determined among the multiple item theme types; The exhibit theme type that overlaps with the plurality of first interest types and the plurality of second interest types is determined as the target interest type.

7. The method according to claim 1, characterized in that, For any given exhibition route; Based on the multiple interest exhibits and at least one preset exposure exhibit, the viewing route is generated, including: Determine the number of exhibits corresponding to the exhibition route, wherein the number of exhibits corresponding to the at least one preset exposure exhibit is greater than the number of exposure exhibits corresponding to the at least one preset exposure exhibit. The difference between the number of items displayed along the route and the number of items displayed in the exposure section is determined as the number of items of interest. The exhibition route is constructed by combining the number of interest items among the multiple interest items with the at least one preset exposure item, and the first item in the exhibition route is the current user's interest item.

8. The method according to claim 1, characterized in that, When recommending a target exhibition route to the current user, it also includes: The system recommends reward information corresponding to the target exhibition route to the current user. The reward information includes multiple exhibition visit quantity ranges and the reward recipients corresponding to each exhibition visit quantity range.

9. The method according to claim 7, characterized in that, This also includes: Obtain the multiple exhibition items that the current user has viewed on the target exhibition route, as well as the dwell time for each exhibition item; The number of valid exhibits corresponding to the multiple browsing exhibits is determined based on the dwell time corresponding to each browsing exhibit. Based on the number of valid exhibits, the target reward recipient for the current user is determined from among the reward recipients corresponding to each exhibit visit range.

10. The method according to claim 1, characterized in that, After recommending the target exhibition route to the user, it also includes: Obtain the current viewing status of the current user, which includes the currently viewed exhibits, the exhibits already viewed, and the viewing duration of each viewed exhibit; Based on the current viewing status, determine the current non-interest tag of the current user; Identify at least one unviewed exhibit in the target viewing route; The browsing position of the unviewed exhibits corresponding to the non-interest tags will be moved to the back, and the corresponding browsing time will be shortened.

11. A device for determining an exhibition route, characterized in that, It includes a data acquisition module, an analysis and processing module, a first determination module, a generation module, a second determination module, and a recommendation module. The data collection module is used to collect user information corresponding to the current user and generate a unique identity identifier. The user information includes the current user's basic information and the viewing information of the virtual exhibition hall. The analysis and processing module is used to analyze and process the user information to obtain the target interest type corresponding to the current user; The first determining module is used to determine, based on the target interest type, multiple interest items corresponding to the current user among multiple exhibits in the virtual exhibition hall; The generation module is used to generate multiple viewing routes based on the multiple interest exhibits and the at least one preset exposure exhibit. The viewing routes guide the current user to browse exhibits in the virtual exhibition hall. Each viewing route includes multiple exhibits and a required browsing time for each exhibit. The required browsing time limits the dwell time on the corresponding exhibit. The multiple exhibits are a combination of exhibits that conform to the target interest type and the exhibition hall's operational goals. The second determining module is used to determine the target viewing route among the plurality of viewing routes; The recommendation module is used to recommend a target exhibition route to the current user based on the unique identifier.

12. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as claimed in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 10.

14. A computer program product comprising a computer program that, when executed by a processor, implements the method of any one of claims 1 to 10.

Images