Page preloading decision-making method and device, electronic equipment and storage medium

Abstract

The invention provides a page preloading decision-making method and device, electronic equipment and a storage medium, and belongs to the technical field of Internet. The method comprises the steps of triggering a preloading decision request for at least one predicted page based on an operation behavior of a user on a current page; based on the preloading decision request, obtaining preloading reference data of the prediction page, the preloading reference data comprising a plurality of reference dimensions; based on the preloading reference data, determining a preloading reference feature of each reference dimension; based on the current attribute information corresponding to the reference dimensions, determining the current feature weight of each reference dimension; and based on the preloading reference feature of each reference dimension and the current feature weight, determining a preloading decision result of the prediction page. By adopting the method and the device, the real-time adaptability of the preloading decision can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet, and in particular to a page preloading decision method and device, electronic equipment and storage medium. BACKGROUND

[0002] With the continuous progress of Internet technology, it becomes more and more convenient for users to browse web pages on their terminal devices (such as mobile phones, tablet computers, etc.). However, due to the instability of the network environment and the performance difference of the terminal device, problems such as page loading delay and lag still occur from time to time, which seriously affects the user experience.

[0003] In order to improve the page response speed, preloading strategies are generally used in the prior art, but most of the schemes cannot be adjusted in real time according to the current environment, so that the preloading strategy cannot adapt to the dynamic changes of user behavior and network conditions, resulting in resource waste or delayed preloading.

[0004] Therefore, how to dynamically optimize the preloading decision based on real-time data has become the key to improving user experience. SUMMARY

[0005] Therefore, the embodiments of the present application provide a page preloading decision method and device, electronic equipment and storage medium, which can improve the real-time adaptability of the preloading decision.

[0006] According to an aspect of the present application, a page preloading decision method is provided, the method comprising: triggering a preloading decision request for at least one predicted page based on the user's operation behavior on the current page; obtaining preloading reference data of the predicted page based on the preloading decision request, the preloading reference data comprising a plurality of reference dimensions; determining the preloading reference feature of each reference dimension based on the preloading reference data; determining the current feature weight of each reference dimension based on the current attribute information corresponding to the reference dimension; determining the preloading decision result of the predicted page based on the preloading reference feature and the current feature weight of each reference dimension.

[0007] According to another aspect of the present application, a page preloading decision device is provided, the device comprising: a data acquisition unit configured to trigger a preloading decision request for at least one predicted page based on the user's operation behavior on the current page; and obtain preloading reference data of the predicted page based on the preloading decision request, the preloading reference data comprising a plurality of reference dimensions; a data fusion unit configured to determine a preloading reference feature of each of the reference dimensions based on the preloading reference data, and determine a current feature weight of each of the reference dimensions based on current attribute information corresponding to the reference dimensions; a decision unit configured to determine a preloading decision result of the predicted page based on the preloading reference feature and the current feature weight of each of the reference dimensions.

[0008] According to another aspect of the present application, an electronic device is provided, comprising: a processor; and a memory storing a program, wherein the program comprises instructions which, when executed by the processor, cause the processor to perform the preloading decision method of the page.

[0009] According to another aspect of the present application, a non-transitory computer readable storage medium storing computer instructions is provided, wherein the computer instructions are configured to cause a computer to perform the preloading decision method of the page.

[0010] In the present application, when a user's operation behavior on a current page triggers a preloading decision request for at least one predicted page, preloading reference data of multiple reference dimensions of the predicted page can be obtained, and preloading reference features of each of the reference dimensions can be determined, current feature weights of each of the reference dimensions can be determined based on current attribute information corresponding to the reference dimensions, and a preloading decision result of the predicted page can be determined based on the preloading reference feature and the current feature weight of each of the reference dimensions. Since the feature weights of the reference dimensions can be adjusted in real time according to the current attribute information corresponding to the reference dimensions, the real-time adaptability of the preloading decision is improved, and different application scenarios can be adapted. BRIEF DESCRIPTION OF DRAWINGS

[0011] More details, features and advantages of the present application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which: Figure 1 a preloading decision method flowchart provided by an exemplary embodiment of the present application is shown; Figure 2 a smart preloading system architecture diagram provided by an exemplary embodiment of the present application is shown; Figure 3 a four-dimensional data fusion decision flowchart provided by an exemplary embodiment of the present application is shown; Figure 4 a schematic block diagram of a preloading decision device of a page provided by an exemplary embodiment of the present application is shown; Figure 5 a structural block diagram of an exemplary electronic device which can be used to implement embodiments of the present application is shown. Detailed Implementation

[0012] Embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While some embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the invention. It should be understood that the accompanying drawings and embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the invention.

[0013] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0014] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0015] It should be noted that the terms "one" and "more" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0016] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0017] This invention provides a page preloading decision method, which can be performed by a terminal, a server, and / or other devices with processing capabilities. The method provided in this embodiment can be performed by any of the aforementioned devices, or by multiple devices working together; this invention does not limit this.

[0018] The following will refer to Figure 1 The flowchart shown illustrates the preloading decision-making method for the page. Figure 2 The diagram shows the architecture of the intelligent preloading system. Figure 3 The flowchart of the four-dimensional data fusion decision-making process is shown to introduce this method.

[0019] As shown in Figure 2 The intelligent preloading system can include a user interaction layer, a data collection layer, a data fusion layer, an intelligent decision-making layer, and an execution management layer. The data collection layer can include a real-time interaction collector, a historical behavior collector, a page structure analyzer, and a device network detector. The data fusion layer can include a feature extraction engine and a data fusion engine. The intelligent decision-making layer can include a dynamic weighted scoring engine and a prediction decision module, and the execution management layer can include a resource scheduling manager, a task execution engine, and a performance monitoring module.

[0020] As shown in Figure 3 In the four-dimensional data fusion decision-making process provided in the embodiment, the four reference dimensions are real-time user interaction data, historical behavior data, page structure data, and device and network status data.

[0021] The method includes the following steps 101-105.

[0022] Step 101, based on the operation behavior of the user on the current page, triggering a preloading decision request for at least one predicted page.

[0023] In one possible implementation, in the user interaction layer, the operation behavior of the user on the current page, such as clicking, sliding, and staying time, can be captured in real time by calling the API interface of the browser end, mobile end, or Web end framework of the user device end or receiving event reporting, and the system can identify the user's intention and trigger a preloading decision request for one or more predicted pages.

[0024] The predicted page refers to a subsequent page that the user is likely to visit. The page that the user is likely to visit next can be predicted based on the user's historical transition probability by Markov chain.

[0025] Step 102, based on the preloading decision request, obtaining preloading reference data of the predicted page.

[0026] The preloading reference data includes multiple reference dimensions. Specifically, in the embodiment, the multiple reference dimensions can include real-time user interaction data, historical behavior data, page structure data, and device and network status data.

[0027] In a possible implementation, the preloading reference data required for decision making can be collected through the data collection layer. Among them, the real-time operation information of the user on the current page can be collected through the real-time interaction collector, including but not limited to mouse trajectory, hovering time, click position, scrolling behavior, keyboard input, etc., so as to obtain real-time user interaction data. Through the historical behavior collector, the past access records and operation patterns of the user are obtained, including but not limited to page access sequence, dwell time, conversion path, preference analysis, etc., to form historical behavior data. Through the page structure parser, the DOM (Document Object Model) structure information, visual layout, element weight, link distribution, content hierarchy, etc. of the current and predicted pages are extracted, and static and dynamic features are formed. Page structure data. Through the device and network detection module, the hardware performance parameters (such as memory size, CPU performance, etc.) of the user terminal and the current network environment parameters (such as network type, bandwidth speed, delay, etc.) are obtained, and device and network status data are generated.

[0028] Whenever a preloading decision request is triggered, the above multi-dimensional preloading reference data corresponding to the predicted page can be obtained from the database and transmitted to the data fusion layer for subsequent processing.

[0029] Step 103, based on the preloading reference data, determine the preloading reference features of each reference dimension.

[0030] In a possible implementation, the feature extraction engine in the data fusion layer can perform normalization processing and feature vector construction on the dimensional data to obtain the preloading reference features of each reference dimension, including real-time interaction features, historical behavior features, page reception features, and device network features. Among them, the real-time interaction features reflect the activity and intention intensity of the current operation of the user, the historical behavior features depict the individual preferences and access rules of the user, the page structure features reflect the loading complexity and resource distribution of the predicted page, and the device network features represent the terminal carrying capacity and transmission efficiency.

[0031] Further, the data fusion engine can align the preloading reference features of each reference dimension to form a fusion feature vector.

[0032] Step 104, based on the current attribute information corresponding to the reference dimension, determine the current feature weight of each reference dimension.

[0033] Among them, corresponding to the four reference dimensions of the above real-time user interaction data, historical behavior data, page structure data, and device and network status data, the current attribute information used can include active user attribute information, new and old user attribute information, and network attribute information.

[0034] In a possible implementation, the current attribute information dynamically affects the weight distribution of each reference dimension in different scenarios. For example, the weight of the real-time interaction feature can be dynamically adjusted according to the active user attribute information, the weight of the historical behavior feature can be adjusted according to the new and old user attribute information, and the network attribute information can be used to dynamically adjust the weight of the device network feature.

[0035] Optionally, the specific processing of the step 104 can include: determining whether to adjust the basic feature weight of the reference dimension based on the current attribute information corresponding to the reference dimension; if yes, adjusting the basic feature weight based on the adjustment factor of the reference dimension to obtain the current feature weight of the reference dimension; if no, taking the basic feature weight of the reference dimension as the current feature weight of the reference dimension.

[0036] As a specific example, the basic feature weights of the real-time user interaction data, the historical behavior data, the page structure data, and the device and network status data can be [0.4, 0.3, 0.2, 0.1] respectively. If the active user attribute information indicates that the user is in a high active state, the weight of the real-time interaction feature can be increased, and the adjustment factor of the real-time interaction feature can be set to 1.5, so that the current feature weight of the real-time interaction feature is adjusted to 0.4x1.5=0.6. If the new and old user attribute information indicates that the user is a new user, the weight of the historical behavior feature can be reduced, and the adjustment factor of the historical behavior feature can be set to 0.3, so that the current feature weight of the historical behavior feature is adjusted to 0.3x0.3=0.09. If the network attribute information shows that the current network environment is weak, the weight of the device network feature can be increased, and the adjustment factor of the device network feature can be set to 2.0, so that the current feature weight of the device network feature is adjusted to 0.1x2.0=0.2.

[0037] After the judgment for each reference dimension is completed, the current feature weights of the reference dimensions can be normalized to ensure that the sum of the weights is 1, thereby ensuring the rationality of the fusion calculation.

[0038] In scenario 1, it is assumed that the user is a new user using a high-performance device, and the network is a 5G network. At this time, the current feature weights of the adjusted reference dimensions can be [0.6, 0.1, 0.2, 0.1] respectively. That is, the weight of the real-time interaction feature is increased to strengthen the response to the current behavior, and the weight of the historical behavior feature is reduced to adapt to the unstable characteristics of the new user behavior mode.

[0039] In scenario 2, it is assumed that the user is an old user using a low-performance device, and the network is in a weak network environment. At this time, the current feature weights of each reference dimension after adjustment can be [0.2, 0.4, 0.1, 0.3] respectively. That is, the weight of the real-time interaction feature is reduced to alleviate the pressure of device performance, the weight of the historical behavior feature is increased to rely on mature behavior patterns, and the weight of the device network feature is increased to cope with the stability problem caused by weak network.

[0040] In scenario 3, it is assumed that the user is an active interaction user, and the network is in a medium network environment. At this time, the current feature weights of each reference dimension after adjustment can be [0.5, 0.3, 0.1, 0.1] respectively. That is, the weight of the real-time interaction feature is increased to respond to the high-frequency operation behavior of the user, the stable contribution of the historical behavior feature is maintained, and the weight of the page structure feature is moderately reduced under the premise of considering the adaptability of the device and the network.

[0041] After determining the current feature weights of each reference dimension, they can be input into the prediction decision module of the intelligent decision layer for processing in step 105.

[0042] Step 105, determining the preloading decision result of the predicted page based on the preloaded reference features and the current feature weights of each reference dimension.

[0043] In one possible implementation, the preloaded reference features and the current feature weights of each reference dimension can be weighted and fused to obtain a comprehensive score for predicting the feasibility of preloading based on a target prediction algorithm in the prediction decision module. If it is determined through the comprehensive score that the user is likely to click the predicted page, the preloading mechanism is triggered; if it is determined through the comprehensive score that the user is unlikely to click the predicted page, the preloading is not triggered to save system resources and network overhead.

[0044] Optionally, different preloading granularities can be selected according to the network state, and the corresponding processing can be as follows: determining a comprehensive score of the predicted page based on the preloaded reference features and the current feature weights of each reference dimension; if the comprehensive score of the predicted page is greater than or equal to a preset threshold, determining a preloading strategy of the predicted page based on the network state of the user device, wherein different network states have corresponding preloading granularities in the preloading strategy.

[0045] The preset threshold of the comprehensive score can be dynamically adjusted according to different business scenarios to optimize resource utilization rate while ensuring user experience.

[0046] The specific preloading strategy determination process can be as follows: If the network state of the user equipment satisfies the high bandwidth condition, the preloading strategy of the predicted page is determined as complete page preloading; If the network state of the user equipment satisfies the medium bandwidth condition, the preloading strategy of the predicted page is determined as key component preloading; If the network state of the user equipment satisfies the low bandwidth condition, the preloading strategy of the predicted page is determined as only interface data preloading.

[0047] In a possible implementation, the prediction decision module can obtain the real-time network bandwidth of the user equipment, and determine whether the bandwidth interval in which the current network is located satisfies the high bandwidth condition, the medium bandwidth condition or the low bandwidth condition, and then select the corresponding preloading strategy. When in the high bandwidth environment, the system preferentially loads complete page content, to ensure that the page is instantly opened after being clicked by the user; in the medium bandwidth environment, only key components are preloaded to balance the speed and resource consumption; and in the low bandwidth weak network condition, only the minimum interface data is loaded to ensure that the basic interactive function is available, and to avoid the decline of user experience caused by resource loading failure.

[0048] Meanwhile, the system can monitor the network state change of the user equipment in real time, dynamically adjust the preloading strategy, ensure that the preloading granularity is timely degraded or upgraded when the bandwidth fluctuates, and avoid redundant requests or loading delay. For example, when the network state recovers to the high bandwidth, the system can automatically complete the preloading of non-core resources that are not previously loaded, to improve the fluency of subsequent response of the page. Or, when the network state decreases from the high bandwidth to the low bandwidth, the system can suspend the preloading of complete page content, and switch to the light mode of only loading interface data, to realize graceful degradation, and guarantee the continuity and stability of user operation.

[0049] After the intelligent decision layer determines the preloading decision result of the predicted page, if the preloading mechanism is triggered, a corresponding preloading instruction can be formed, and the preloading instruction is transmitted to the execution management layer to execute the corresponding preloading task. The resource scheduling manager can be used to manage the life cycle of the preloading task, including the creation, priority sorting, concurrency control and resource release of the task.

[0050] Optionally, after the task execution engine executes the preloading, the preloading result can also be used for feedback optimization. The corresponding processing can be as follows: After the preloading is executed based on the preloading decision result, preloading effect data is obtained; Based on the preloading effect data, the basic feature weight and / or the adjustment factor are updated.

[0051] The preloading effect data can include prediction accuracy and performance improvement data. The prediction accuracy is used to represent the probability of hitting the actual target page. The performance improvement data includes response time improvement data. The actual target page refers to the page finally accessed by the user. The response time refers to the time interval from the user initiating a request to receiving a complete response.

[0052] In a possible implementation, the page response time after preloading is completed, the resource loading time consumption and the user click behavior data can be collected in real time by the performance monitoring module, and the actual access path is compared with the predicted target page to calculate the prediction accuracy. The system can feed back the preloading effect data to the intelligent decision layer, and update the basic feature weight and / or adjustment factor based on the preloading effect data through the reinforcement learning algorithm, so as to continuously optimize the decision accuracy of the subsequent preloading strategy.

[0053] As a specific example, after an e-commerce website applies the page preloading decision method provided by the application, when the user browses the commodity list page, the system detects that the user's mouse hovers over a certain commodity for 2 seconds, and then the detail page of the commodity can be preloaded as a predicted page. Historical data shows that the access probability of this type of commodity detail page is 85%. Page structure analysis shows that the commodity is located in the center of the page and has high visual weight. Network detection shows that the user uses a 5G network. Then, the comprehensive score of the commodity is calculated by the above method to reach a preset threshold, and the system triggers the preloading instruction to start rendering the commodity detail page. When the user clicks, the page is displayed instantly, and the loading time is reduced from 1.2 seconds to 0.1 seconds.

[0054] As another specific example, in a news website application scenario, the user is reading an article, and the system analyzes the user's reading speed and scrolling position to predict that the user will soon finish reading the current article and start analyzing the next article preloading candidate. Combined with the user's historical reading preferences (technology articles) and the relevance of the current article, the next recommended article is determined as the predicted page. The preloading of the recommended article is triggered by calculating the comprehensive score of the recommended article to reach a preset threshold by the above method. The network state of the current user device is in a medium bandwidth network environment, and the system dynamically adjusts the priority and granularity of the preloaded resources, and selects to preload only the HTML (HyperText Markup Language, HyperText Markup Language) and key pictures of the recommended article. When the user slides to the end of the article and clicks the recommended link, the page has completed the core content rendering in advance, and the content can be displayed quickly, and the user's overall perception is improved.

[0055] The embodiment can achieve the following beneficial effects: (1) When the user's operation behavior on the current page triggers a preloading decision request for at least one predicted page, the preloading reference data of multiple reference dimensions of the predicted page can be obtained, and the preloading reference features of each reference dimension are determined, the current feature weight of each reference dimension is determined based on the current attribute information corresponding to the reference dimension, and then the preloading decision result of the predicted page is determined based on the preloading reference features and the current feature weight of each reference dimension. Since the feature weight of each reference dimension can be adjusted in real time according to the current attribute information corresponding to the reference dimension, the real-time adaptability of the preloading decision is improved, which can adapt to different application scenarios.

[0056] (2) Through multi-dimensional data fusion, the prediction accuracy is greatly improved compared with single index. Moreover, through the collaborative analysis of historical data and real-time data, the long-term mode of the user is integrated with the short-term intention, and the preloading hit rate is further improved.

[0057] (3) Through fine preloading granularity control, the complete page, key component or only interface data can be selected for preloading, precise resource utilization is realized, invalid preloading is avoided, and resource waste is reduced.

[0058] (4) Precise preloading reduces user waiting time, response speed is faster, interaction is smoother, and user experience is significantly improved.

[0059] (5) The improvement of page loading speed can also reduce the user churn rate, and better user experience can promote user conversion.

[0060] The embodiment of the application provides a preloading decision device of a page, which is used for realizing the preloading decision method of the page. Figure 4 As shown in the figure, the preloading decision device 400 of the page comprises a data acquisition unit 401, a data fusion unit 402 and a decision unit 403.

[0061] The data acquisition unit 401 is used for triggering a preloading decision request for at least one predicted page based on the user's operation behavior on the current page; and obtaining preloading reference data of the predicted page based on the preloading decision request, wherein the preloading reference data comprises multiple reference dimensions. The data fusion unit 402 is used for determining the preloading reference features of each reference dimension based on the preloading reference data; and determining the current feature weight of each reference dimension based on the current attribute information corresponding to the reference dimension. The decision unit 403 is used for determining the preloading decision result of the predicted page based on the preloading reference features and the current feature weight of each reference dimension.

[0062] Optionally, the data fusion unit 402 is used for: determine whether to adjust the base feature weight of the reference dimension based on the current attribute information corresponding to the reference dimension; If yes, adjust the base feature weight based on the adjustment factor of the reference dimension to obtain the current feature weight of the reference dimension. If no, take the base feature weight of the reference dimension as the current feature weight of the reference dimension.

[0063] Optionally, the data fusion unit 402 is further configured to: acquire preloading effect data after preloading based on the preloading decision result is performed; update the base feature weight and / or the adjustment factor based on the preloading effect data.

[0064] Optionally, the preloading effect data includes prediction accuracy and performance improvement data, the prediction accuracy is used to represent the probability of hitting the actual target page, and the performance improvement data includes response time improvement data.

[0065] Optionally, the decision unit 403 is configured to: determine the comprehensive score of the predicted page based on the preloading reference feature and the current feature weight of each reference dimension; If the comprehensive score of the predicted page is greater than or equal to a preset threshold, determine the preloading strategy of the predicted page based on the network state of the user equipment, wherein different network states have corresponding preloading granularities in the preloading strategy.

[0066] Optionally, the decision unit 403 is configured to: If the network state of the user equipment meets the high bandwidth condition, determine that the preloading strategy of the predicted page is complete page preloading. If the network state of the user equipment meets the medium bandwidth condition, determine that the preloading strategy of the predicted page is key component preloading. If the network state of the user equipment meets the low bandwidth condition, determine that the preloading strategy of the predicted page is only interface data preloading.

[0067] Optionally, the plurality of reference dimensions of the preloading reference data include real-time user interaction data, historical behavior data, page structure data, and device and network condition data. The current attribute information includes active user attribute information, new and old user attribute information, and network attribute information.

[0068] The present embodiment can achieve the following beneficial effects: (1) When the user's operation behavior on the current page triggers a preloading decision request for at least one predicted page, the preloading reference data of multiple reference dimensions of the predicted page can be obtained, the preloading reference features of each reference dimension are determined, the current feature weight of each reference dimension is determined based on the current attribute information corresponding to the reference dimension, and then the preloading decision result of the predicted page is determined based on the preloading reference features and the current feature weight of each reference dimension. Since the feature weight of each reference dimension can be adjusted in real time according to the current attribute information corresponding to the reference dimension, the real-time adaptability of the preloading decision is improved, which can adapt to different application scenarios.

[0069] (2) Through multi-dimensional data fusion, the prediction accuracy is greatly improved compared with single index. Moreover, through the collaborative analysis of historical data and real-time data, the long-term mode of the user is integrated with the short-term intention, and the preloading hit rate is further improved.

[0070] (3) Through fine preloading granularity control, the complete page, key component or only interface data can be selected for preloading, the accurate use of resources is realized, invalid preloading is avoided, and resource waste is reduced.

[0071] (4) Precise preloading reduces user waiting time, response speed is faster, interaction is smoother, and user experience is significantly improved.

[0072] (5) The improvement of page loading speed can also reduce the user churn rate, and better user experience can promote user conversion.

[0073] The exemplary embodiments of the present application also provide an electronic device, comprising: at least one processor; and a memory connected with the at least one processor in communication. The memory stores a computer program capable of being executed by the at least one processor, and the computer program is used to make the electronic device execute the method according to the embodiments of the present application when executed by the at least one processor.

[0074] The exemplary embodiments of the present application also provide a non-transitory computer readable storage medium storing a computer program, wherein the computer program is used to make the computer execute the method according to the embodiments of the present application when executed by the processor of the computer.

[0075] The exemplary embodiments of the present application also provide a computer program product comprising a computer program, wherein the computer program is used to make the computer execute the method according to the embodiments of the present application when executed by the processor of the computer.

[0076] Reference Figure 5The present invention will now be described in the form of a structural block diagram of an electronic device 500 that can serve as a server or client of the present invention, which is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0077] like Figure 5 As shown, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. The RAM 503 may also store various programs and data required for the operation of the device 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0078] Multiple components in electronic device 500 are connected to I / O interface 505, including: input unit 506, output unit 507, storage unit 508, and communication unit 509. Input unit 506 can be any type of device capable of inputting information to electronic device 500. Input unit 506 can receive input digital or text information and generate key signal inputs related to user settings and / or function control of electronic device. Output unit 507 can be any type of device capable of presenting information and may include, but is not limited to, a display, speaker, video / audio output terminal, vibrator, and / or printer. Storage unit 508 may include, but is not limited to, disks and optical discs. Communication unit 509 allows electronic device 500 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth devices, Wi-Fi devices, WiMax devices, cellular communication devices, and / or the like.

[0079] The computing unit 501 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs various methods and processes described above. For example, in some embodiments, the page preloading decision method described above can be implemented as a computer software program, which is tangibly embodied in a machine-readable medium, such as the storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 500 via the ROM 502 and / or the communication unit 509. In some embodiments, the computing unit 501 can be configured to perform the page preloading decision method described above by any other suitable means, such as by means of firmware.

[0080] Program code for carrying out methods of the present application can written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces a function / operation specified in the flowchart and / or block diagram. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, and partially on a remote machine or entirely on a remote machine or server.

[0081] In the context of the present application, a machine-readable medium can be a tangible medium that can contain or store program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable storage media can include, without limitation, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media can include one or more lines of electrical connections, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical storage devices, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0082] As used in this description, the terms "machine-readable medium," "computer-readable medium," and "computer-readable media" refer to any computer program product, apparatus and / or device (e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and / or data to a programmable processor.

[0083] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0084] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0085] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

Claims

1. A method for making preloading decisions for a page, characterized in that, The method includes: Based on the user's actions on the current page, a preloading decision request for at least one predicted page is triggered; Based on the preloading decision request, preloading reference data for the prediction page is obtained, and the preloading reference data includes multiple reference dimensions; Based on the preloaded reference data, determine the preloaded reference features for each reference dimension; Based on the current attribute information corresponding to the reference dimension, determine the current feature weight of each reference dimension; The preloading decision result of the prediction page is determined based on the preloaded reference features and the current feature weights for each reference dimension.

2. The method according to claim 1, characterized in that, The step of determining the current feature weight of each reference dimension based on the current attribute information corresponding to the reference dimension includes: Based on the current attribute information corresponding to the reference dimension, determine whether to adjust the basic feature weights of the reference dimension; If so, the basic feature weights are adjusted based on the adjustment factor of the reference dimension to obtain the current feature weights of the reference dimension; If not, then the basic feature weights of the reference dimension will be used as the current feature weights of the reference dimension.

3. The method according to claim 2, characterized in that, The method further includes: After performing preloading based on the preloading decision results, obtain the preloading effect data; Based on the preloaded effect data, the basic feature weights and / or the adjustment factors are updated.

4. The method according to claim 3, characterized in that, The preloading effect data includes prediction accuracy and performance improvement data. The prediction accuracy is used to represent the probability of hitting the actual target page, and the performance improvement data includes response time improvement data.

5. The method according to claim 1, characterized in that, The step of determining the preloading decision result of the predicted page based on the preloaded reference features and the current feature weights for each reference dimension includes: The comprehensive score of the predicted page is determined based on the preloaded reference features and the current feature weights for each reference dimension. If the overall score of the prediction page is greater than or equal to a preset threshold, then the preloading strategy of the prediction page is determined based on the network status of the user device. In the preloading strategy, different network statuses have corresponding preloading granularities.

6. The method according to claim 5, characterized in that, The method for determining the preloading strategy of the predicted page based on the network status of the user equipment includes: If the network status of the user device meets the high bandwidth condition, then the preloading strategy of the predicted page is determined to be full page preloading; If the network status of the user equipment meets the medium bandwidth condition, then the preloading strategy of the predicted page is determined to be the preloading of key components. If the network status of the user device meets the low bandwidth condition, then the preloading strategy of the prediction page is determined to be interface data preloading only.

7. The method according to claim 1, characterized in that, The multiple reference dimensions of the preloaded reference data include: real-time user interaction data, historical behavior data, page structure data, and device and network status data; The current attribute information includes active user attribute information, new and old user attribute information, and network attribute information.

8. A page preloading decision-making device, characterized in that, The device includes: The data acquisition unit is used to trigger a preloading decision request for at least one prediction page based on the user's operation behavior on the current page; and to obtain preloading reference data for the prediction page based on the preloading decision request, wherein the preloading reference data includes multiple reference dimensions. The data fusion unit is used to determine the preloaded reference features of each reference dimension based on the preloaded reference data; and to determine the current feature weight of each reference dimension based on the current attribute information corresponding to the reference dimension. A decision unit is configured to determine the preloading decision result of the prediction page based on the preloaded reference features and the current feature weights for each of the reference dimensions.

9. An electronic device, comprising: processor; as well as Stored program memory, The program includes instructions that, when executed by the processor, cause the processor to perform the method according to any one of claims 1-7.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-7.

Images