A method and apparatus for loading a platform online resource

Abstract

The application relates to a platform online resource loading method and device, and relates to the technical field of computer technology. The method comprises the following steps: in response to a sliding touch screen instruction of a resource display page, determining an identity switching intention and a channel switching intention of the sliding touch screen instruction; if the identity switching intention is large, updating a current account attribute into a preset standard browser attribute, and requesting and loading platform online resources according to the standard browser attribute; if the identity switching intention is small and the channel switching intention is large, determining a target channel to be switched, and requesting and loading platform online resources of the target channel; and if both the identity switching intention and the channel switching intention are small, switching adjacent resources of a current channel or adjusting a display progress of a current resource. By using the application, when a user browses online resources, the contact range of learning, information and planning contents can be expanded, and the effective expansion of a user cognitive boundary and the optimized experience of multi-level content browsing can be realized.

Description

Technical Field

[0004]

[0001] This application relates to the field of computer technology, and particularly to a method and device for loading platform online resources. Background Art

[0002] With the popularization of mobile terminals, video-playing applications have become an important channel for users to obtain information, entertainment, and learning content. Such applications usually provide a continuously scrolling video display interface, and users can achieve resource browsing, content switching, and progress viewing through the sliding operation of the touch screen. Therefore, sliding touch interaction has become an important mode for online video browsing.

[0003] Existing video-playing applications generally adopt methods such as interest model analysis, behavior feature modeling, and content label combination during the process of loading platform online resources. According to the user's historical viewing records and staying behaviors, relevant resource content related to the user's interests is retrieved from the foreground. At the same time, during the resource switching process, the background uses preloading and caching strategies to minimize the display delay caused by network fluctuations and improve browsing fluency.

[0004] However, there may still be certain limitations in the process of browsing platform online resources by existing video-playing applications. For example, when users browse resources of different types or sources, the content presentation path is relatively fixed, and the resource jump method is relatively single, which may affect the user's expansion in the content field to a certain extent, thus limiting the browsing experience. Summary of the Invention

[0005] Based on this, it is necessary to provide a method and device for loading platform online resources for the above technical problems.

[0006] In a first aspect, this application provides a method for loading platform online resources, including: Responding to a sliding touch screen instruction on a resource display page, determining a horizontal sliding component and a vertical sliding component corresponding to the sliding touch screen instruction; Based on the horizontal sliding component and the vertical sliding component, determining an identity switching intention value and a channel switching intention value of the sliding touch screen instruction; If the identity switching intention value is greater than or equal to a first confidence level, updating the current account attribute to a preset standard viewer attribute, and requesting and loading platform online resources according to the standard viewer attribute; If the identity switching intention value is less than the first confidence level and the channel switching intention value is greater than or equal to a second confidence level, determining a target channel to be switched, and requesting and loading platform online resources of the target channel; If the identity switching intention value is less than the first confidence level and the channel switching intention value is less than the second confidence level, then switch to load adjacent resources of the current channel or adjust the display progress of the current resource according to the vertical sliding component and the horizontal sliding component.

[0007] In one embodiment, determining the identity switching intention value and the channel switching intention value based on the horizontal sliding component and the vertical sliding component includes: Construct a horizontal / vertical sliding feature vector based on the sliding angle θ, the sliding distance d, and the sliding speed v corresponding to the horizontal sliding component and the vertical sliding component; Input the horizontal / vertical sliding feature vector into a preset intention recognition model to output the identity switching intention value and the channel switching intention value.

[0008] In one embodiment, the first confidence level and the second confidence level are continuously adjusted according to the user's historical sliding behavior and the resource switching situation after sliding.

[0009] In one embodiment, requesting and loading the platform online resources of the target channel includes: Retain the cached content of the current resource and retrieve the most recently displayed resource of the target channel that has been stored; Request and load the most recently displayed resource, and preload the adjacent resources of the most recently displayed resource.

[0010] In one embodiment, the method further includes: If the display interval duration of the most recently displayed resource of the target channel is greater than the threshold, then delete the most recently displayed resource and extract the resource tag; Request and load the platform online resources of the target channel according to the resource tag.

[0011] In one embodiment, requesting and loading the platform online resources of the target channel includes: Divide the current display page into multiple display areas, and continue to display the current resource in one display area; Request the platform online resources of the target channel and perform preloading, and display the thumbnail preview of the platform online resources in other display areas.

[0012] In one embodiment, updating the current account attribute to a preset standard viewer attribute includes: Among multiple optional standard viewer attributes, update the current account attribute to the standard viewer attribute with the highest degree of association.

[0013] In a second aspect, the present application further provides a device for loading platform online resources, including: A sliding instruction response module, configured to determine a horizontal sliding component and a vertical sliding component corresponding to the sliding touch screen instruction in response to a sliding touch screen instruction on a resource display page; A sliding intention determination module, configured to determine an identity switching intention value and a channel switching intention value of the sliding touch screen instruction based on the horizontal sliding component and the vertical sliding component; An online resource loading module, configured to: If the identity switching intention value is greater than or equal to a first confidence level, update the current account attribute to a preset standard viewer attribute, and request and load platform online resources according to the standard viewer attribute; If the identity switching intention value is less than the first confidence level, and the channel switching intention value is greater than or equal to a second confidence level, determine a target channel to be switched, and request and load the platform online resources of the target channel; If the identity switching intention value is less than the first confidence level, and the channel switching intention value is less than the second confidence level, switch and load adjacent resources of the current channel or adjust the display progress of the current resource according to the vertical sliding component and the horizontal sliding component.

[0014] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the method steps described in the first aspect above are implemented.

[0015] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps described in the first aspect above are implemented.

[0016] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the method steps described in the first aspect above are implemented.

[0017] By using the platform online resource loading method disclosed in the present application, by parsing the sliding touch screen instruction on the resource display page and determining the identity switching intention value and the channel switching intention value based on the horizontal and vertical sliding components, the sliding operation can drive the intelligent switching of the identity level, the channel level, and the resource content level. This mechanism can break through the fixed limitations of interest tag recommendations, enable users to quickly substitute between different identity perspectives, and naturally switch between channels and adjacent resources, thereby expanding the contact scope of learning, information, and planning content. At the same time, through node-based progress adjustment and fast switching of adjacent resources, the resource screening efficiency can be improved while maintaining the interaction fluency, thereby effectively expanding the user's cognitive boundary and optimizing the experience of multi-level content browsing. Description of the Drawings

[0018] Figure 1 A schematic flowchart of a method for loading online resources on a platform in an embodiment; Figure 2 A schematic structural diagram of a device for loading online resources on a platform in an embodiment; Figure 3 An internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0019] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0020] The embodiments of the present application provide a method for loading online resources on a platform. This method can be applied to the clients of various types of resource display platforms. The client can be a mobile terminal installed with the resource display app corresponding to the platform. The resource display platform can be a short video platform, a learning resource platform, a campus service platform, an interest content subscription platform, a comprehensive information browsing platform, etc. When using the resource display app, users usually browse resources, switch content, and jump interfaces by sliding the screen. By improving the resource loading logic and interaction method of the resource display app, the content distribution efficiency can be improved, the user browsing experience can be enhanced, and the diversified exposure of resources can be promoted.

[0021] In the above platform-client structure, the client can be the core execution entity of the resource loading logic, used to complete the acquisition of the sliding instruction, the determination of the sliding component, the identification of the resource category, the loading request for the corresponding resource, and the adjustment of the current resource display progress. After the user issues a sliding behavior, the client can perform a multi-dimensional fusion judgment on the sliding direction and sliding distance, so as to select to load adjacent resources, switch the corresponding channel, or reconstruct the user browsing identity under specific conditions in a more flexible manner, so that the user can obtain a more diversified content perspective within a shorter browsing path.

[0022] The following will take the resource display platform as a college student video platform as an example to Figure 1 explain a method for loading online resources on a platform as shown, which specifically may include the following steps: Step 101, in response to a sliding touch screen instruction on the resource display page, determine the horizontal sliding component and the vertical sliding component corresponding to the sliding touch screen instruction.

[0023] In implementation, the resource display page can be a content presentation interface running on a smart terminal, such as a short video display interface, a graphic and text stream display interface, or a mixed resource browsing interface. Users can perform a swiping operation through the touch screen, and the terminal can capture data such as the trajectory information, moving speed, and moving distance of the contact point through a touch event detection module.

[0024] Furthermore, the terminal can sample the moving path of the contact point within a preset time window, and calculate the displacement components in the horizontal and vertical directions respectively based on coordinate system projection, so as to obtain the corresponding horizontal sliding component and vertical sliding component. In some implementation manners, the terminal can set a minimum trigger threshold, and only when the sliding distance exceeds the threshold is it determined as a valid sliding operation, so as to avoid the influence of accidental touch on the displayed resources.

[0025] Step 102, determine the identity switching intention value and channel switching intention value of the sliding touch screen instruction based on the horizontal sliding component and the vertical sliding component.

[0026] In implementation, an intention judgment model can be configured in the client for semantic-level recognition of the horizontal sliding component and the vertical sliding component. Specifically, the client can be provided with an identity switching function and a channel switching function triggered by the horizontal sliding component and the vertical sliding component. It can be understood that for the college student APP, different from conventional video playback software, its users are mainly college students. Further, different from other user groups, the general needs of college students are learning and exploration. At the same time, considering that college students are at the stage of connecting campus and society and are in the plasticity period of ideological concepts, in addition to the traditional resource recommendation based on interests, more resource recommendations from different fields and different perspectives can be given to reasonably assist and guide the use of the client by the college student user group. Therefore, the identity switching function and the channel switching function can effectively help college student users break through the interest cocoon in a timely manner when "swiping videos and swiping news" through a simple swiping operation. Specifically, adapted to the identity switching function, in addition to the user's personal identity, the platform can set up virtual user identities, and the attributes of the virtual user identities are different from those of the user's personal identity. There can be multiple virtual user identities, and the attributes of each virtual user identity are mutually different. Adapted to the channel switching function, the platform can set up multiple channels with different themes, and preferentially recommend and display the platform's online resources related to the theme of the channel under each channel. For example, in view of the general needs of college students, channels such as knowledge learning, campus activities, entertainment and leisure, hot news, emotional psychology, job hunting and development, etc. can be set up.

[0027] The model can construct two intention values for measuring the identity change tendency and the channel switching tendency according to the sliding direction, sliding amplitude, sliding speed, change of the sliding trajectory curve, sliding inertia characteristics, and the user's historical sliding habits.

[0028] Step 103, if the identity switching intention value is greater than or equal to the first confidence level, update the current account attributes to the preset standard viewer attributes, and request and load the platform online resources according to the standard viewer attributes.

[0029] In implementation, when a strong identity switching intention is detected, the identity switching process can be triggered. The preset standard viewer attributes here can be established in advance by the platform to represent the typical focus points at different stages. Specifically, for the college student APP, since the arrangements in universities are basically the same, corresponding grade identities can be set according to the enrollment time, that is, identities such as freshmen, sophomores, juniors,..., second-year postgraduate students, third-year postgraduate students, etc. can be specifically set. It can be understood that in the university stage, most students have basically experienced or will face the same things in different grades. For example, freshmen generally focus on campus adaptation and interest exploration, sophomores generally focus on professional learning and campus activities, juniors generally focus on social internships and postgraduate entrance examination experiences, seniors generally focus on job hunting guidance and graduation projects, etc. Similarly, the same is true for the postgraduate stage. Considering that the number of doctoral students is small and their cognitive abilities are strong, and the research focus content is difficult to popularize, only the identities corresponding to each grade in the university stage and the postgraduate stage can be set. In this way, the platform can use the viewer attributes of a large number of students in the corresponding academic years as samples to generate the standard viewer attributes for the corresponding academic years to match the student attributes of the corresponding academic years.

[0030] Based on the above processing, when users use the above-mentioned college student app, they can trigger the identity switching function through a sliding touch screen operation, so that the client updates the current account attributes to the preset standard viewer attributes, requests and loads the platform online resources according to the standard viewer attributes. The platform can update the browsing tags bound to the current account to switch the subsequent recommended resources to the content set adapted to this standard identity. Subsequently, the platform can send a resource request instruction to the server according to the updated identity attributes to request to load the platform online resources that conform to this identity perspective. Therefore, users can timely break away from the personal interest cocoon and the current identity limitation, substitute into the viewer perspective of different grade identities, and experience the platform online resources provided by the platform. It is worth mentioning that through the grade identity switching function, users can compare with their own historical experiences when substituting into the lower grade perspective to inspire users to summarize and think about their past experiences, and can also understand the matters that need to be considered in the future in advance when substituting into the higher grade perspective to assist in guiding users to make reasonable plans at present, avoiding users being茫然 about the future and wasting and idling their current energy, thereby gradually guiding college student users at different stages to extend to the upper-level academic planning perspective and enabling them to form a clearer understanding of future responsibilities, course arrangements, and career nodes.

[0031] Step 104, if the identity switching intention value is less than the first confidence level and the channel switching intention value is greater than or equal to the second confidence level, determine the target channel to be switched, request and load the online resources of the platform for the target channel.

[0032] In implementation, when the user's sliding intention is not sufficient to trigger identity reconstruction but shows a strong tendency for channel switching, the client can determine the target channel to be switched according to the channel logical structure. Here, the target channel to be switched can be an adjacent channel of the current channel pre-set by the platform for the client, or a channel that matches the user's sliding operation, or a channel selected by the user himself after the channel list is triggered by the channel switching intention. After that, the client can generate a resource loading instruction after obtaining the channel identifier to request the online resource list of the platform under this channel and complete the interface refresh. This process can also reduce the frequency of interaction by the user through the menu bar and search bar, improving the experience fluency.

[0033] Step 105, if the identity switching intention value is less than the first confidence level and the channel switching intention value is less than the second confidence level, switch and load adjacent resources of the current channel or adjust the display progress of the current resource according to the vertical sliding component and the horizontal sliding component.

[0034] In implementation, if the intention value corresponding to the user's sliding is not sufficient to trigger changes in the identity and channel levels, it can be defaulted as a local browsing behavior. Further, at this time, a larger vertical sliding component can trigger the rapid switching of multiple adjacent resources within the current channel, that is, the client can enter the thumbnail display mode and continuously display the thumbnails of multiple adjacent resources to facilitate the user to quickly select the corresponding resources; a smaller vertical sliding component can trigger the rapid switching of adjacent resources to match the general vertical resource switching method. At the same time, a larger horizontal sliding component can trigger the node-based progress adjustment of the current resource. It can be understood that since college students have relatively abundant time and energy and mostly have an exploratory tendency, the adapted platform online resources are mostly rich in content and sufficient in information, such as knowledge explanation, event analysis, content sharing, etc. The platform can determine multiple progress nodes for these resources according to the resource content to achieve the rapid switching of single-resource content nodes triggered by the user's horizontal sliding component during resource display. And a smaller horizontal sliding component can be used to achieve a small-distance fine-tuning of the progress bar of the current resource to match the general horizontal video progress adjustment method.

[0035] It is worth mentioning that the resource loading guidance logic in the embodiments of this application is more adapted to the college student user group, that is, more adapted to college student APPs. On the one hand, due to the group consistency of the basic attribute characteristics of college students, there is an execution basis for resource guidance and recommendation. On the other hand, since college students are in a period of rapid absorption of diverse information content, there is a necessity for guidance and recommendation.

[0036] Optionally, the processing in step 102 can be specifically as follows: construct a horizontal / vertical sliding feature vector based on the sliding angles θ, sliding distances d, and sliding speeds v corresponding to the horizontal sliding component and the vertical sliding component; input the horizontal / vertical sliding feature vector into a preset intention recognition model to output an identity switching intention value and a channel switching intention value.

[0037] In implementation, in a college student video platform, to more accurately identify the user's identity switching or channel switching intention from the sliding operation, the multi-dimensional features of the sliding behavior can be quantified into vectors recognizable by the model, and the intention value can be output through the preset model. Specifically, based on the horizontal sliding component ΔX and the vertical sliding component ΔY of the sliding touch screen instruction, three core features, namely the sliding angle θ, the sliding distance d, and the sliding speed v, can be extracted to construct a feature vector. Among them, the sliding angle θ is the angle between the sliding trajectory and the screen horizontal line, reflecting the dominant tendency of the sliding direction; the sliding distance d is the straight-line length of the sliding trajectory, used to reflect the intention intensity of the user's operation. The greater the distance, the clearer the intention; the sliding speed v is the ratio of the sliding distance to the sliding time, which can be used to reflect the urgency of the user's operation.

[0038] Meanwhile, an intention recognition model can be preset in the client. This model takes the horizontal / vertical sliding feature vector as the input and outputs an identity switching intention value and a channel switching intention value. This model can use the sliding behaviors of a large number of users on the platform as training materials to generate initial model parameters. Then, during each user's use of the client, the sliding behavior of the user and the final corresponding switching result can be continuously extracted to continuously optimize the model parameters to more accurately identify the user's sliding intention. Further, in the initial stage of the user's use of the client, a confirmation step for the switching intention can be provided for the user's sliding touch screen operation. That is, when the model outputs an identity / channel switching intention value based on the user's sliding behavior and triggers the identity switching or channel switching logic, a switching confirmation box pops up for the user to confirm or cancel the switching intention, so as to avoid the client misjudging the user's sliding switching intention with a high probability in the initial stage of the user's use. Of course, the client can also reverse-train the user's sliding habits by displaying sliding indication information. For example, when detecting the start of the user's sliding, the angle range and sliding distance for triggering identity switching or channel switching can be displayed on the screen, or a sliding animation can be shown through a floating window to define the minimum sliding speed for triggering the switching. Based on this, the user can consciously adjust the sliding behavior according to the content displayed on the screen according to actual needs. It can be understood that after the model is continuously optimized and the recognition accuracy of the identity switching intention and the channel switching intention is greater than a certain threshold, the above-mentioned auxiliary processing of switching confirmation or guiding sliding can be stopped.

[0039] Optionally, the first confidence level and the second confidence level can be continuously adjusted according to the user's historical sliding behavior and the resource switching situation after sliding. That is, the client can record the user's historical sliding behavior data and the actual operation results after sliding as the basis for confidence level adjustment. Further, in addition to identifying whether the user performs a fallback operation after switching, the residence time after switching can also be additionally combined to estimate the user's degree of recognition of the switching result. If the user repeatedly performs a fallback on the switching result or the residence time is generally short, the confidence level can be appropriately increased to avoid frequently identifying the user's sliding behavior as a switching intention.

[0040] It is worth mentioning that when predicting the switching result corresponding to the user's sliding behavior, this embodiment provides two methods for optimizing the intention recognition model and adjusting the confidence level. The client can select any one of the adjustment methods or choose both methods simultaneously. Preferably, since the method of adjusting the confidence level is relatively simple and fast and has a greater impact, it can be used in the initial stage of the user using the client to quickly set a confidence level that relatively matches the user's operation habits and timely improve the misidentification of the switching intention of the sliding behavior. During the subsequent long-term use of the client by the user, the intention recognition model can be continuously optimized to more accurately identify the user's sliding operation intention and form a model that matches the user's operation habits.

[0041] In addition, since the adjustment of the confidence level method is relatively simple, the client can actively adjust the confidence level according to the user's potential usage needs, so as to give reasonable guidance to college students without the user's perception or with weak perception. For example, when approaching the exam week and detecting that the user frequently switches between the "review channel" and the "true question channel", and the client finds through historical data that the vertical sliding matching rate has increased, the second confidence level can be automatically reduced to make the switching easier to trigger; when the user only browses with a single identity for a long time, the platform can slightly reduce the first confidence level to make the horizontal sliding easier to trigger the switching of other grade identities and promote the user's access to diverse platform online resources.

[0042] Optionally, the specific processing of requesting and loading the platform online resources of the target channel in step 104 can be as follows: retain the cached content of the current resource, retrieve the most recently displayed resource of the target channel that has been stored; request and load the most recently displayed resource, and preload the adjacent resources of the most recently displayed resource.

[0043] In implementation, when it is determined that it is necessary to request and load the online resources of the target channel, the method of retrieving resources can also be optimized to reduce the resource loading delay caused by channel switching. When the client completes channel switching, the data cache of the current resources is retained to support the user's possible return behavior in the future. For example, the user may hope to return to the original channel after browsing a new channel. At this time, retaining the cache can achieve a quick page restoration, avoiding re-requesting the online resources of the platform, thereby reducing network consumption and shortening the loading time.

[0044] Meanwhile, the client can retrieve the most recently displayed resources corresponding to the target channel from the local cache, session records, or user logs on the platform side. For example, if the user has browsed the target channel before, the client can preferentially retrieve the resources that the user has most recently viewed in the target channel and use them as the display object for the first item after channel switching to provide a continuous experience. If the user enters the target channel for the first time, the recommended priority resources within the target channel can be selected as the display starting point.

[0045] In addition, when the client displays the most recently displayed resources, it can also send a parallel request to the platform to obtain the adjacent resource data of the resource for preloading. The adjacent resources here can be determined according to different dimensions of the resources. For example, in video resources, the adjacent resources can be video entries at adjacent positions in the recommendation sequence; in learning resources, the adjacent resources can be adjacent chapters or material entries under the same theme; in campus news resources, the adjacent resources can be context entries of similar activities or announcements of the same type.

[0046] In another optional embodiment, after the client completes the identification of the target channel, it can also determine whether the most recently displayed resources of the channel still have display value based on the user's historical display behavior. That is, if the display interval duration of the most recently displayed resources of the target channel is greater than the threshold, the most recently displayed resources are deleted, and resource tags are extracted; the platform online resources of the target channel are requested and loaded according to the resource tags.

[0047] Specifically, the client can retain a historical display record associated with each channel in the local storage space. This record at least includes the display time of the most recently displayed resources within each channel. The client can calculate the display interval duration between the current time and this display time and compare the display interval duration with a preset threshold.

[0048] When it is detected that the display interval duration is greater than the preset threshold, it means that the resource has not been accessed for a long time, and continuing to display this resource may reduce the interest relevance of the user. Therefore, the client can perform a deletion operation to remove the local cache content of this resource to release the cache space and avoid reloading inappropriate content.

[0049] Subsequently, the client can use the resource tags associated with the resource as a basis for requesting to load new resources. Specifically, the client can request the platform to obtain the target channel resource content that conforms to the tag characteristics based on the resource tags, thereby ensuring the continuity of the user's browsing behavior within the target channel and reducing the impact of fragmented attention of the college student user group. It can be understood that especially in the knowledge-related channels popular among the college student group (i.e., the channel resources contain more information content), users can switch to other channels to relax moderately or browse other resource content in parallel during the learning process. After that, when switching back to the knowledge-related channel, the recently displayed resources can serve as the user's cognitive anchor points to help the user enter the learning state faster.

[0050] At the same time, for the platform's recommendation algorithm, since the channel switching function is introduced in this embodiment, it may break the user interest clues relied on by the recommendation algorithm, increasing the probability of misjudging the interest by the recommendation model and making it impossible to guarantee the recommendation accuracy. By introducing the mechanism of the recently displayed resources, it can assist the execution of the recommendation algorithm, which is beneficial to ensuring the weight stability of the user interest model and avoiding large deviations in recommendations caused by channel switching.

[0051] Optionally, when the client requests and loads the platform online resources of the target channel, it divides the current display page into multiple display areas and continues to display the current resource in one display area; requests the platform online resources of the target channel and performs preloading, and displays the thumbnail previews of the platform online resources in other display areas.

[0052] In implementation, when the client determines the target channel based on the user's sliding touch screen instruction and triggers the loading of platform online resources, the client can perform dynamic layout division on the current display page so that the user can perceive the resource content of the target channel in advance without interrupting the browsing of the current resource. Specifically, after receiving the instruction to trigger channel switching, the client can call the interface rendering module to divide the current display page into areas, forming a main display area for continuing to display the current resource and at least one thumbnail display area as an auxiliary display space. At this time, the current resource still maintains normal playback or scrolling display, and its local cache, rendering thread, and progress status are all retained.

[0053] Meanwhile, the client can request the corresponding platform online resource list according to the identifier of the target channel, and select the platform online resources with higher adaptability as the display objects according to the local cache policy, display priority policy and user's historical interaction records of the client. After the preloading of the platform online resources of the target channel is completed, the client can display the content of the platform online resources in the form of thumbnail previews in the divided auxiliary display area. In this way, the channel switching is visualized, the cost of channel switching is reduced, enabling users to quickly learn about the content of other channels while maintaining the immersive experience of the current content, reducing the sense of fragmentation in resource browsing caused by channel switching, and allowing users to choose whether to switch channels or whether to reload other resources of the target channel according to the thumbnail previews.

[0054] Optionally, the process of updating the viewer attribute in step 103 can be specifically: among multiple optional standard viewer attributes, update the current account attribute to the standard viewer attribute with the highest correlation.

[0055] In implementation, when the client calculates the identity switching intention value based on the user's touch screen sliding instruction and determines that it is greater than or equal to the first confidence level, it is considered that the user is currently in the active browsing identity switching state. For this reason, the client can access multiple preset standard viewer attribute sets locally, and these standard viewer attributes can be abstracted and summarized according to the user's learning stage, usage habits or content preferences. In the usage scenario of the college student App, the standard viewer attributes can include, but are not limited to, viewer models corresponding to different grade stages of freshman, sophomore, junior, senior and postgraduate, and each attribute contains different information requirement tags and corresponding resource weight configurations.

[0056] Before updating the account attribute, the client can first call the attribute matching module to perform a correlation calculation on multiple optional standard viewer attributes by comparing the attribute indicators such as the behavior characteristic data, channel access frequency, resource stay duration, and sliding access path stored in the local cache of the current account. After that, the client can automatically select the target standard viewer attribute with the highest correlation according to the calculated correlation result for updating, so as to avoid that when switching the viewer identity, due to the excessive fragmentation between the viewer attribute and the current account attribute, the user cannot effectively browse the platform recommended resources after the identity switch, greatly reducing the user's interest in resource browsing, and thus unable to achieve the guiding and planning value of the expected identity switching function for the college student user group. After the attribute update is completed, the client will initiate a resource request instruction to the resource side based on the new standard viewer attribute to obtain content resources that are more matched with this identity in real time.

[0057] In one embodiment, during the execution of resource request and loading by the client, it can further combine multi-dimensional schedule information stored locally or authorized for synchronization by the user, including class schedules, work and rest times, positioning information, etc., to implement targeted resource recommendation logic. Specifically, in the scenario of the college student App, the client can obtain the activity type of the user in the current time period from the local calendar file, course scheduling data, or the data synchronization interface of the school course schedule system, and adjust the label weights of various resources according to the pre-set resource recommendation strategy. For example, during class, the client can increase the push probability of knowledge point explanation resources and after-class tutoring resources related to the course; during the bedtime period, the client can reduce the push ratio of academic-depth resources and instead display light-reading content; when it detects that the user's location is in the classroom or library, it can appropriately increase the priority of resources such as learning plans and exercise analysis, while on campus roads or in the dormitory, it is more inclined to present lightweight browsing resources. By combining the user's time, space, and activity context for content scheduling, the client can achieve dynamic optimization during the resource loading stage, effectively improving the rationality and targeting of the user's browsing experience.

[0058] In one embodiment, during the process of the client displaying online resources on the platform, it can monitor in real time the duration of the user's stay at each progress node within the resources. For example, for video courses, the nodes can correspond to chapter switches, knowledge point transitions, or key prompt paragraphs; for graphic resources, the nodes can correspond to different paragraphs, structural headings, or branches of the knowledge graph. When the client detects that the user's stay duration at a certain node is significantly higher than the average level, it can mark the content corresponding to this node as a potential difficulty point or a high-interest point, and increase the caching priority of this content in the cache management module to avoid loading delays when the user looks back later due to background cleaning. On the contrary, for the content that the user quickly skips over, the client can reduce its priority and clean it first when the device storage space is tight or the cache cycle is full. Through the above method, the client can selectively retain the content that is more valuable to the user under limited cache space, enabling the user to obtain a higher response speed when reviewing, looking back, or jumping to browse, which is especially suitable for college students' learning behaviors such as repeatedly looking back at difficult knowledge and focusing on reviewing wrong questions.

[0059] In one embodiment, the client can also automatically classify and switch resource types according to the user's swiping speed on the resource display page to match the user's real-time attention investment state. Specifically, when the client detects that the user is swiping quickly at a high speed, it predicts that the user is in a low-energy browsing mode and can preferentially load resources that can be quickly consumed, such as entertainment, campus funny topics, and easy club information. When detecting a medium-speed swipe, the client can push resources that require a short stay to understand, such as experience sharing, competition activity notifications, and internship interview experiences. When the swipe speed is close to slow or even pauses, it can be considered that the user is in the focused browsing stage, and the client then increases the recommendation weight of academic courses, scientific research discussions, and in-depth lecture content. Through the dynamic mapping relationship between the swipe speed and the resource type, the college student user group can obtain resource content with a higher matching degree when the state changes, thereby reducing the push of invalid resources and improving the user's browsing experience.

[0060] In one embodiment, to meet the needs of college students for content coherence during the learning process, after the client identifies that the current resource belongs to the learning channel, it can be set not to adopt the random adjacent resource loading strategy, but to associate relevant resource content according to the knowledge structure diagram of the resources. For example, when the current resource corresponds to the knowledge points of the third chapter of a certain course, after the user performs a downward swipe action, the client can automatically load the exercise explanation video, extended reading materials, or the knowledge content of the next chapter of this chapter. Further, the client can generate a resource association path based on the course sequence, knowledge point dependency, or wrong question bank data, and preferentially load along the path when a swipe event is triggered, so that the content forms a topologically coherent knowledge chain. By executing this processing logic, the user can naturally advance the learning progress during the swipe operation and avoid the problem of learning disconnection caused by random recommendations.

[0061] In the embodiments of this application, by responding to the swipe touch screen instruction on the resource display page and respectively determining the horizontal swipe component and the vertical swipe component, the client can accurately identify the user's swipe behaviors in different directions and amplitudes. On this basis, the swipe behavior is mapped to an identity switching intention value and a channel switching intention value, expanding the swipe operation from the traditional single content page turning to the identity level and the channel level, which is different from the existing fixed recommendation logic based on interest tags. Especially in the usage scenario of the college student APP, users are often more likely to fall into the content loop guided by interest and it is difficult to actively contact cross-topic content such as learning plans and information cognition. By introducing the intention value determination mechanism, higher-level browsing significance can be given to the swipe behavior, facilitating users to get out of the low-value interest content loop in a timely manner.

[0062] When it is detected that the identity switching intention value reaches the first confidence level, the client updates the current account attributes to the preset standard viewer attributes, and requests the platform online resources based on these attributes, enabling the user to quickly switch perspectives among different academic levels. This helps the college student group enhance their awareness of future planning and avoid cognitive limitations caused by long-term immersion in single-interest videos. When the identity switching intention is insufficient but the channel switching intention reaches the second confidence level, the client can automatically switch to the target channel and load the corresponding platform online resources, facilitating the user's planned channel-level switching of resources.

[0063] For sliding behaviors that are not sufficient to trigger identity reconstruction and channel switching, the switching of adjacent resources or the display progress of the current resource can be controlled according to the vertical sliding component and the horizontal sliding component. This not only preserves the general browsing experience but also adapts to the needs of college students to quickly screen information during fragmented time.

[0064] In summary, through the combination of the two-dimensional components of sliding touch control and multi-level browsing semantics in the embodiments of this application, the browsing process of platform online resources is closely linked at the identity level, channel level, and content level, changing the limitation that the content presentation of existing video applications is locked by interest weights. This helps users actively expand their cognitive boundaries when browsing platform online resources, achieve more reasonable resource push and content loading, and thus obtain a user experience and browsing value significantly superior to existing resource display applications.

[0065] It should be understood that although the steps in the flowcharts involved in the above embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless clearly stated in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps in other steps.

[0066] Based on the same inventive concept, as Figure 2 shown, the embodiments of this application also provide a loading device 200 for platform online resources. The device 200 includes: A sliding instruction response module 201, configured to determine the horizontal sliding component and the vertical sliding component corresponding to the sliding touch screen instruction in response to the sliding touch screen instruction on the resource display page; A sliding intention determination module 202, configured to determine the identity switching intention value and the channel switching intention value of the sliding touch screen instruction based on the horizontal sliding component and the vertical sliding component; The online resource loading module 203 is used for: If the identity switching intention value is greater than or equal to the first confidence level, update the current account attribute to a preset standard viewer attribute, and request and load the platform online resources according to the standard viewer attribute; If the identity switching intention value is less than the first confidence level and the channel switching intention value is greater than or equal to the second confidence level, determine the target channel to be switched, and request and load the platform online resources of the target channel; If the identity switching intention value is less than the first confidence level and the channel switching intention value is less than the second confidence level, switch and load adjacent resources of the current channel or adjust the display progress of the current resource according to the vertical sliding component and the horizontal sliding component.

[0067] In one embodiment, the sliding intention determination module 202 is specifically used for: Construct a horizontal / vertical sliding feature vector based on the sliding angle θ, sliding distance d, and sliding speed v corresponding to the horizontal sliding component and the vertical sliding component; Input the horizontal / vertical sliding feature vector into a preset intention recognition model, and output an identity switching intention value and a channel switching intention value.

[0068] In one embodiment, the first confidence level and the second confidence level are continuously adjusted according to the user's historical sliding behavior and the resource switching situation after sliding.

[0069] In one embodiment, the online resource loading module 203 is specifically used for: Retain the cached content of the current resource, and retrieve the most recently displayed resource of the target channel that has been stored; Request and load the most recently displayed resource, and preload the adjacent resources of the most recently displayed resource.

[0070] In one embodiment, the online resource loading module 203 is specifically used for: If the display interval duration of the most recently displayed resource of the target channel is greater than the threshold, delete the most recently displayed resource, and extract the resource tag; Request and load the platform online resources of the target channel according to the resource tag.

[0071] In one embodiment, the online resource loading module 203 is specifically used for: Divide the current display page into multiple display areas, and continue to display the current resource in one display area; Request the platform online resources of the target channel and perform preloading, and display the thumbnail preview of the platform online resources in other display areas.

[0072] In one embodiment, the online resource loading module 203 is specifically configured to: Among multiple optional standard viewer attributes, update the current account attribute to the standard viewer attribute with the highest degree of association.

[0073] In one embodiment, a computer device is provided, and its internal structure diagram can be as Figure 3 shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange data between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it implements a method for loading online resources of a platform.

[0074] Those skilled in the art can understand that Figure 3 the structure shown in

[0075] is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0076] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps in the above method embodiments.

[0077] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0078] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The processors involved in the embodiments provided in this application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0079] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0080] The above-described embodiments only represent several implementation manners of this application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of this application. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can still be made, and these all belong to the protection scope of this application. Therefore, the protection scope of this application should be subject to the appended claims.

Claims

1. A method for loading online resources on a platform, characterized in that, The method includes: In response to a swipe touch command on the resource display page, determine the horizontal swipe component and the vertical swipe component corresponding to the swipe touch command; The identity switching intent value and channel switching intent value of the swipe touch screen command are determined based on the horizontal swipe component and the vertical swipe component. If the identity switching intent value is greater than or equal to the first confidence level, the current account attributes are updated to the preset standard visitor attributes, and online resources on the platform are requested and loaded according to the standard visitor attributes. If the identity switching intent value is less than the first confidence level and the channel switching intent value is greater than or equal to the second confidence level, then the target channel to be switched to is determined, and the platform online resources of the target channel are requested and loaded. If the identity switching intent value is less than the first confidence level and the channel switching intent value is less than the second confidence level, then based on the vertical sliding component and the horizontal sliding component, switch to loading the adjacent resources of the current channel, or adjust the display progress of the current resource.

2. The method according to claim 1, characterized in that, The determination of identity switching intent value and channel switching intent value based on horizontal sliding component and vertical sliding component includes: Based on the sliding angle θ, sliding distance d, and sliding speed v corresponding to the lateral and longitudinal sliding components, construct lateral / longitudinal sliding feature vectors; The horizontal / vertical sliding feature vectors are input into a preset intent recognition model, which outputs the identity switching intent value and the channel switching intent value.

3. The method according to claim 1, characterized in that, The first confidence level and the second confidence level are continuously adjusted based on the user's historical swiping behavior and the resource switching situation after swiping.

4. The method according to claim 1, characterized in that, The request and loading of the platform's online resources for the target channel includes: Retain the cached content of the current resource and retrieve the most recently displayed resource of the target channel that has been stored. Request and load the recently displayed resource, and preload the adjacent resources of the recently displayed resource.

5. The method according to claim 1, characterized in that, The method further includes: If the display interval of the most recently displayed resource in the target channel exceeds the threshold, then the most recently displayed resource is deleted and the resource tag is extracted; The platform's online resources for the target channel are requested and loaded according to the resource tag.

6. The method according to claim 1, characterized in that, The request and loading of the platform's online resources for the target channel includes: Divide the currently displayed page into multiple display areas, and continue to display the current resource in one of the display areas; The platform requests online resources from the target channel and performs preloading, then displays a thumbnail preview of the online resources in other display areas.

7. The method according to claim 1, characterized in that, The step of updating the current account attributes to preset standard browser attributes includes: Among several optional standard visitor attributes, update the current account attribute to the standard visitor attribute with the highest relevance.

8. A device for loading online resources on a platform, characterized in that, The device includes: The swipe command response module is used to respond to swipe touch commands on the resource display page and determine the horizontal swipe component and the vertical swipe component corresponding to the swipe touch command. The swipe intent determination module is used to determine the identity switching intent value and channel switching intent value of the swipe touch screen command based on the horizontal swipe component and the vertical swipe component. The online resource loading module is used for: If the identity switching intent value is greater than or equal to the first confidence level, the current account attributes are updated to the preset standard visitor attributes, and online resources on the platform are requested and loaded according to the standard visitor attributes. If the identity switching intent value is less than the first confidence level and the channel switching intent value is greater than or equal to the second confidence level, then the target channel to be switched to is determined, and the platform online resources of the target channel are requested and loaded. If the identity switching intent value is less than the first confidence level and the channel switching intent value is less than the second confidence level, then based on the vertical sliding component and the horizontal sliding component, switch to loading the adjacent resources of the current channel, or adjust the display progress of the current resource.

9. A computer device comprising a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the method for loading online resources of a platform according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, the computer program, when executed by a processor, implementing the method for loading online resources of a platform according to any one of claims 1-7.

Images