Preloading method of content and related device

Abstract

The present disclosure provides a preloading method of content and related devices. The method is applied to a terminal device, and includes: receiving first data generated by a user browsing a first page of the content, the first data including information related to at least one of a content dimension, a device dimension, and a user dimension; calling a target model pre-trained and stored in the terminal device, inputting the first data into the target model, and outputting predicted information corresponding to the first data; and preloading a second page of the content corresponding to the predicted information.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a content preloading method and related devices. Background Art

[0002] In the related art, when a user is browsing content, subsequent content can be pre-loaded, so that when the user views the subsequent content, the user can quickly obtain the subsequent content without waiting for a network request.

[0003] However, the inventors of the present disclosure have discovered that in the related art, the preloaded content is the same for each user, which easily leads to a waste of resources. Summary of the Invention

[0004] The present disclosure proposes a content preloading method and related devices to solve or partially solve the above problems.

[0005] In a first aspect, the present disclosure provides a content preloading method, which is applied to a terminal device and includes:

[0006] receiving first data generated by a user browsing a first page of the content, the first data including information related to at least one of a content dimension, a device dimension, and a user dimension;

[0007] calling a pre-trained target model stored in the terminal device, inputting the first data into the target model, and outputting prediction information corresponding to the first data;

[0008] A second page of the content corresponding to the prediction information is preloaded.

[0009] In a second aspect of the present disclosure, a content preloading apparatus is provided, which is applied to a terminal device and includes:

[0010] a receiving module configured to: receive first data generated when a user browses a first page of the content, the first data including information related to at least one of a content dimension, a device dimension, and a user dimension;

[0011] a prediction module configured to: call a pre-trained target model stored in the terminal device, input the first data into the target model, and output prediction information corresponding to the first data;

[0012] The preloading module is configured to preload a second page of the content corresponding to the prediction information.

[0013] In a third aspect of the present disclosure, a computer device is provided, comprising one or more processors, a memory; and one or more programs, wherein the one or more programs are stored in the memory and executed by the one or more processors, and the programs include instructions for executing the method according to the first aspect.

[0014] According to a fourth aspect of the present disclosure, a non-volatile computer-readable storage medium containing a computer program is provided. When the computer program is executed by one or more processors, the processors are caused to execute the method described in the first aspect.

[0015] In a fifth aspect of the present disclosure, a computer program product is provided, comprising computer program instructions, which, when executed on a computer, cause the computer to execute the method described in the first aspect.

[0016] The content preloading method and related equipment provided by the present disclosure input the data generated by users browsing content into a target model to obtain prediction information, and then load subsequent pages based on the prediction information, avoiding the waste of resources caused by using a unified preloading method. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present disclosure or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1A A schematic diagram showing an exemplary system provided by an embodiment of the present disclosure is shown.

[0019] Figure 1B A schematic diagram showing an exemplary scenario according to an embodiment of the present disclosure is shown.

[0020] Figure 2A A schematic diagram of an exemplary method provided by an embodiment of the present disclosure is shown.

[0021] Figure 2B An exemplary flow chart for collecting data according to an embodiment of the present disclosure is shown.

[0022] Figure 3 A schematic diagram showing another exemplary method provided by an embodiment of the present disclosure is shown.

[0023] Figure 4 A schematic diagram of the hardware structure of an exemplary computer device provided by an embodiment of the present disclosure is shown.

[0024] Figure 5 A schematic diagram of an exemplary device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0025] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0026] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present disclosure should have the usual meanings understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the embodiments of the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the described object changes, the relative position relationship may also change accordingly.

[0027] Figure 1A FIG. 1 is a schematic diagram of an exemplary system 100 provided by an embodiment of the present disclosure.

[0028] like Figure 1A As shown, the system 100 may include at least one terminal device (e.g., terminal devices 102 and 104), a server 106, and a database server 108. A medium providing a communication link may be included between the terminal devices 102 and 104 and the server 106 and the database server 108, such as a network 110. The network 110 may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0029] User 112 can use terminal devices 102 and 104 to interact with server 106 via network 110 to receive or send messages, etc. Various applications (APPs) can be installed on terminal devices 102 and 104, such as model training applications, reading applications, video applications, social applications, payment applications, web browsers, and instant messaging tools.

[0030] The terminal devices 102 and 104 herein can be either hardware or software. When the terminal devices 102 and 104 are hardware, they can be various electronic devices with display screens, including but not limited to smartphones, tablet computers, e-book readers, MP3 players, laptop computers, and desktop computers (PCs). When the terminal devices 102 and 104 are software, they can be installed in the electronic devices listed above. They can be implemented as multiple software or software modules (for example, to provide distributed services), or as a single software or software module. No specific limitations are given here.

[0031] When the terminal devices 102 and 104 are hardware, they may also be equipped with a video capture device. The video capture device may be any device capable of capturing video, such as a camera, a sensor, etc. The user may use the video capture device on the terminal devices 102 and 104 to capture video.

[0032] The server 106 may be a server that provides various services, such as a backend server that supports various applications displayed on the terminal devices 102 and 104. The server 106 may use the data sent by the terminal devices 102 and 104 as training samples to train the initial model, and may send the training results (such as the target model for preloading content) to the terminal devices 102 and 104. In this way, users can apply the target model to preload content.

[0033] Database server 108 may also be a database server that provides various services. For example, the database server may store a sample set. The sample set includes a large number of samples. These samples may include data generated by browsing content. User 112 may also select training samples from the sample set stored in database server 108 via terminal devices 102 and 104. It is understood that if server 106 can perform the relevant functions of database server 108, database server 108 may not be provided in system 100.

[0034] The server 106 and database server 108 herein can also be hardware or software. When they are hardware, they can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When they are software, they can be implemented as multiple software programs or software modules (for example, to provide distributed services), or as a single software program or software module. No specific limitations are imposed herein.

[0035] It should be noted that the content preloading method provided in the embodiment of the present application is generally executed by the terminal devices 102 and 104, and the training device for training the target model is generally set in the server 106.

[0036] It should be understood that Figure 1A The number of terminal devices, networks, database servers, and servers in the embodiment is merely illustrative. Any number of terminals, networks, database servers, and servers may be provided as required.

[0037] Content-based products generally provide content to users. For example, reading apps typically provide books. To optimize the user experience while using content-based products, several subsequent pieces of content can be pre-downloaded while the user is browsing the current content. This eliminates the need to wait for network requests before browsing subsequent content, significantly increasing the user's desire to continue browsing.

[0038] Figure 1B A schematic diagram showing an exemplary scenario according to an embodiment of the present disclosure is shown.

[0039] like Figure 1B As shown, taking user 112 reading a novel on terminal device 104 as an example, generally, the content data provided to the user (e.g., the novel) can be stored in database server 108. User 112 can use terminal device 104 to send a request 1042 to server 106 to obtain the target content (e.g., by clicking a Start Reading button to trigger the request). Server 106 responds to the request and searches the database server 108 for the target content and returns a target page 1082 (e.g., the first page) of the target content to terminal device 104. During this process, if multiple pages after the target page need to be preloaded, server 106 also needs to return these multiple pages (e.g., pages 2 to N) to terminal device 104. Terminal device 104 also needs to store these multiple pages locally so that when the user browses subsequent pages after the target page (e.g., by clicking a Next Page button), terminal device 104 does not need to send a request to server 106, but can directly retrieve local data for display.

[0040] It can be seen that this preloading method will occupy more of the content provider's server bandwidth because it needs to additionally load content that is not part of the current request. It may also consume more user traffic and occupy more storage space on the terminal device.

[0041] Correspondingly, this approach can also lead to waste. This is because, while a user is browsing the current page, they may not continue to browse subsequent pages. Using the aforementioned preloading method will directly load multiple pages after the target page, ignoring the user's personal wishes. Therefore, this approach may lead to waste of server bandwidth and waste of data and disk space on the user's device.

[0042] In view of this, the embodiment of the present disclosure provides a preloading method of content. First data generated by a user browsing a first page of the content is used to generate corresponding prediction information based on a target model, and then a second page of the content corresponding to the prediction information is preloaded. In this way, data generated by a user browsing content is used to predict the user's intention based on the data using a target model, and then the subsequent page matching the user's intention is loaded, so that the loading of the subsequent page is related to the user's intention, and resource waste caused by using a unified preloading method is avoided. In addition, the target model is a pre-trained target model stored in the terminal device, which can reduce the number of interactions between the terminal device and the server, thereby improving the processing speed and response speed and improving the user experience.

[0043] Figure 2A A schematic diagram of an exemplary method 200 provided by the embodiment of the present disclosure is shown. As shown in the method 200 can further include the following steps. Figure 2A

[0044] As shown in the method 200 can further include the following steps. Figure 2A As shown in the initial state, it is assumed that the user 112 uses the terminal device 102 to send a request to the server 106 to browse the target content (for example, a certain book). The server 106 can return the target page (for example, the first page of the book) of the target content to the terminal device 102 based on the request.

[0045] Then, the user 112 can use the terminal device 102 to browse the received target page. In the process of browsing, the terminal device 102 can collect data 202 generated by browsing the target page, respectively.

[0046] As an optional embodiment, the data 202 can further include information related to the content dimension and / or the device dimension. Optionally, the data 202 can include device information and / or content information. Since the data 202 can be used to train the initial model to generate the target model for predicting the user's intention, the device information and / or the content information included in the data 202 can improve the prediction effect of the target model trained based on the data 202.

[0047] ​The content information may further include the type of content browsed by the user 112 (e.g., book categories), the number of browsed contents (e.g., how many books), whether to view content details (e.g., detailed introduction of the book), interaction information with the content (e.g., whether to comment or like), duration of content browsing, etc. As an optional embodiment, when the displayed content is a book, the type of browsed content may further include the category of the browsed books, the number of browsed contents may further include the number of browsed books, whether to view content details may further include whether to view the book details page, the interaction information with the content may further include comments, likes, and collections made on the book, and the duration of content browsing may further include the duration of reading the book.

[0048] Device information may further include device battery level, network type (e.g., 4G, 5G, Wi-Fi, etc.), device model, and device location information (e.g., city). Device information is used to reflect the context of the user browsing content and can express the context of data 202, thereby better reflecting the user's content browsing intention.

[0049] Figure 2B A schematic diagram illustrating an exemplary process 220 of collecting data 202 according to an embodiment of the present disclosure is shown.

[0050] In some embodiments, the data 202 can be collected at a predetermined location where the user 112 browses to the target page. In other words, when the user 112 browses to the predetermined location of the target page, the terminal device 102 can generate the data 202 as a data sample. As an optional embodiment, Figure 2B As shown, the predetermined position can be the end of the target page. In this way, when the user 112 browses one of the pages of a certain content to the end, the data at the current moment can be collected to generate a data sample accordingly.

[0051] After generating data 202, in some embodiments, terminal device 102 does not immediately upload data 202 to server 106. Instead, it may label data 202 based on the user's subsequent browsing behavior. As an optional embodiment, the model used to train data 202 may be a decision tree model (e.g., XGBoost, LightGBM), and the labels of data 202 may be positive sample labels or negative sample labels.

[0052] In some embodiments, as Figure 2B As shown, the positive sample label or the negative sample label can be marked in the following manner:

[0053] If the user 112 continues to browse the next page of the current page that generates the data 202, the data 202 can be marked as a positive sample;

[0054] If the user 112 does not continue to browse the next page of the current page that generates the data 202, the data 202 can be marked as a negative sample.

[0055] In this way, by analyzing the subsequent browsing data, it can be determined whether the data 202 is a positive sample or a negative sample, and accordingly labeled, for training the model.

[0056] As an optional embodiment, the above steps of generating the data 202 and labeling the data 202 can be implemented by using the point embedding technology, so as to automatically generate and label the data 202, thereby saving the step of manual labeling and improving the running efficiency.

[0057] After obtaining the data 202 labeled with the label, as shown in Figure 2A and Figure 2B The terminal device 102 can upload the data 202 with the label to the server 106. As an optional embodiment, the server 106 can store the data 202 into the database server 108 for subsequent training of the model.

[0058] After running the above process for a period of time (for example, one day, one week, one month, etc.), a training sample set including a plurality of data 202 can be obtained. It can be understood that the data 202 in the training sample set can be data 202 generated by the user 112 browsing any content page. For example, it can be data generated by reading different pages of different books in the same reading APP.

[0059] Then, as shown in Figure 2A The training sample set can be used to train the initial model to obtain the target model 204.

[0060] As an optional embodiment, in the selection of the model, a decision tree model can be used. It can be understood that the target model 204 can be used to predict the browsing intention of the user, which can actually be converted into binary classification (i.e., yes or no) according to the user data. Therefore, in addition to the decision tree model, other classification models can also be used, such as neural network model, support vector machine, Bayesian classification model, etc.

[0061] The decision tree model may be selected, for example, XGBoost or LightGBM. Among them, the XGBoost model is a relatively mature framework in the industry, which is pre-trained using the existing feature data set. Therefore, on the basis of the pre-trained model, the target model 204 can be quickly and conveniently trained using the aforementioned training sample set with less data.

[0062] After the target model 204 is trained, as shown in Figure 2A , the server 106 can distribute the data package or software package (for example, SDK) of the target model 204 to the terminal device 102 for predicting the subsequent browsing intention of the user 112.

[0063] After the terminal device 102 receives the data package or software package of the target model 204, it can be stored locally or used to update the corresponding content APP (for example, a certain reading APP).

[0064] Then, the user 112 can continue to browse the content. For example, continue to read the current book or open a new book. It can be understood that at this time, the terminal device 102 can send a new request to the server 106 according to the operation of the user 112 (for example, click next page or click a new book), and the server 106 returns the page corresponding to the request to it.

[0065] In the process of browsing the page by the user 112, as shown in Figure 2A , the terminal device 102 can continue to generate data 206.

[0066] As an optional embodiment, the data 206 can further include device information and / or content information. Alternatively, the data 206 can include device information and / or content information, so that the prediction effect based on the data 206 is better.

[0067] Among them, the content information can further include the type of content (for example, the classification of the book) browsed by the user 112, the number of content (for example, how many books) browsed, whether to view the details of the current content (for example, the detailed introduction of the book), the interaction information with the current content (for example, whether to comment, like), the duration of browsing the current content, etc.

[0068] The device information can further include device power, network type (for example, 4G, 5G, WIFI, etc.), device model, device positioning information (for example, the city where it is located). The device information is used to reflect the scene information when the user browses the current content, and can express the context of the data 202, so as to better reflect the browsing intention of the user to the current content.

[0069] When the user 112 browses to a predetermined position of the current page (for example, the end of the page), a mechanism for predicting the user's 112 intention is triggered, such as Figure 2A As shown, a software package containing the target model stored locally on the terminal device 102 can be called, data 206 can be input into the target model 204, and prediction information corresponding to the data 206 can be output. Then, subsequent pages of the current content can be preloaded based on the prediction information. In this way, by calling the locally stored target model, the number of interactions between the terminal device and the server can be reduced, thereby improving processing speed and response speed, and enhancing the user experience.

[0070] In some embodiments, rather than directly performing a binary classification based on the prediction information, i.e., dividing the user into willing or not willing, the user's willingness can be assigned a corresponding level based on the specific value of the prediction information. Therefore, as an optional embodiment, when target model 204 is a decision tree model, the prediction information is the probability of predicting data 206 as a positive sample; accordingly, subsequent pages of the current content can be preloaded based on this probability.

[0071] In some embodiments, user willingness can be divided into three levels: high, medium, and low. Accordingly, two thresholds can be set, such that when the probability is greater than the first threshold (e.g., 0.8), indicating that the user's willingness to continue browsing is high, the first number of second pages of the current content (e.g., the pages of the next 20 chapters) can be preloaded; when the probability is greater than the second threshold (e.g., 0.6) and less than or equal to the first threshold, indicating that the user's willingness to continue browsing is medium, the second number of second pages of the current content (e.g., the pages of the next 10 chapters) can be preloaded; when the probability is less than or equal to the second threshold, indicating that the user's willingness to continue browsing is low, the preloading of the current content can be stopped.

[0072] As an optional embodiment, the preload request 210 corresponding to the aforementioned preload operation may be triggered when the user clicks the next page and sent to the server 106. The preload request 210 may include the number of pages to be loaded or the user's willingness level. Thus, upon receiving the preload request 210, the server 106 determines the data to be returned based on the information about the number of pages to be loaded in the preload request 210, or determines the data to be returned based on the user's willingness level in the preload request 210. The server 106 then returns the page 210 to be preloaded to the terminal device 102.

[0073] In some embodiments, the terminal device 102 can continuously upload data generated during the content browsing process to the server 106 for continuing to train or retrain the target model 204, thereby updating the target model 204 so that the predicted information can be more in line with current habits.

[0074] As an optional embodiment, the time node for triggering the next preloading can be before the preloaded content is about to be browsed (for example, the last page is left to be browsed), and then, for example, when the user clicks on the next page, the preloading mechanism is triggered again to complete the data preloading, thereby avoiding frequent interactions between the terminal device 102 and the server 106 to save resources.

[0075] As can be seen from the above embodiments, the system 100 provided by the disclosed embodiments selectively preloads unviewed content by predicting user intent. For example, preloading is not performed for users with a very low willingness to continue browsing; only a small amount of content is preloaded for users with a moderate willingness to continue browsing; and more content is preloaded for users with a strong willingness to continue browsing. This saves both server bandwidth costs and user device resources.

[0076] The intelligent preloading method provided by the disclosed embodiments balances user experience with resource costs and offers a solution for personalized preloading behavior for different users. In poor network conditions, such as weak connections, this method reduces waiting time for users who wish to browse to the next page of data. Furthermore, for users who do not wish to browse to the next page of data, this method reduces server resource waste and device disk usage.

[0077] The embodiment of the present disclosure also provides a method for preloading content. Figure 3 The flow chart of the exemplary method 300 provided by the embodiment of the present disclosure is shown. The method 300 may be Figure 1A The terminal device 102 or 104 is implemented. Figure 3 As shown, the method 300 may further include the following steps.

[0078] In step 302, a user (eg, Figure 2A The first data (eg, the first page of the content (eg, a book)) generated by the user 112 browsing the first page (eg, a page of the book) is Figure 2A Optionally, the first data includes information related to content dimension and / or device dimension.

[0079] In step 304, a pre-trained target model (e.g., Figure 2A model 204), and input the first data into the target model, and output prediction information corresponding to the first data.

[0080] In some embodiments, the target model is second data (for example, a third page of the content or other content) generated by the user browsing the content or other content (for example, other pages of the book currently being browsed, or pages of other books). Figure 2A In this way, the model trained based on the data generated by the user 112 browsing other pages of the current content or pages of other content can better reflect the user's wishes.

[0081] In some embodiments, the target model is pre-trained on a server and stored in the terminal device as a software package. Thus, by invoking the locally stored target model, the number of interactions between the terminal device and the server can be reduced, thereby increasing processing speed and response speed, and enhancing the user experience.

[0082] In some embodiments, the target model is a decision tree model, and the second data is labeled with a positive sample label or a negative sample label.

[0083] In some embodiments, the positive sample label or the negative sample label is labeled in the following manner: in response to the user having browsed the next page of the third page, the second data is labeled as a positive sample; or in response to the user not browsing the next page of the third page, the second data is labeled as a negative sample. In this way, automatic labeling based on browsing information eliminates the need for manual labeling and improves operational efficiency.

[0084] In some embodiments, inputting the first data into a pre-trained target model and outputting prediction information corresponding to the first data includes: in response to the user browsing to a predetermined location on the first page, inputting the first data into the pre-trained target model and outputting prediction information corresponding to the first data. By setting a trigger mechanism, operational efficiency can be improved.

[0085] In some embodiments, the first data or the second data includes device information and / or content information, so as to better reflect browsing intentions.

[0086] In some embodiments, the device information includes at least one of device battery level, network type, device model, and device location information. The device information is used to reflect the context of the user browsing content and can express the context of the data 202, thereby better reflecting the user's content browsing intention.

[0087] In some embodiments, the content information includes at least one of the type of content viewed by the user, the amount of content viewed, whether the user viewed the details of the content, interaction information with the content, and the duration of the content viewing.

[0088] In some embodiments, the content is a book, the type of content viewed includes the category of the book viewed, the number of content viewed includes the number of books viewed, whether to view content details includes whether to view the book details page, the interaction information with the content includes comments, likes, and favorites on the book, and the content viewing time includes the time spent reading the book. In this way, the problem of preloading book pages in the book reading scenario can be effectively solved.

[0089] In step 306 , a second page of the content corresponding to the prediction information is preloaded.

[0090] In some embodiments, the prediction information is a probability of predicting the first data as a positive sample; and preloading the second page corresponding to the prediction information includes preloading the second page corresponding to the probability. In this way, user preferences can be divided into multiple levels based on probability, and the number of pages corresponding to the level can be preloaded, thereby achieving more efficient resource utilization.

[0091] In some embodiments, the preloading of the second page of the content corresponding to the probability includes: in response to the probability being greater than a first threshold, preloading a first number of second pages of the content; in response to the probability being greater than a second threshold and less than or equal to the first threshold, preloading a second number of second pages of the content; or in response to the probability being less than or equal to the second threshold, stopping preloading the content.

[0092] It should be noted that the method of the embodiments of the present disclosure can be performed by a single device, such as a computer or server. The method of the embodiments of the present disclosure can also be applied in a distributed scenario, where multiple devices cooperate to perform the method. In such a distributed scenario, one of the multiple devices may only perform one or more steps of the method of the embodiments of the present disclosure, and the multiple devices will interact with each other to complete the method.

[0093] It should be noted that the above description is limited to some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in an order different from that described in the above embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0094] The embodiment of the present disclosure further provides a computer device for implementing the above method 200 or 300. Figure 4FIG. 4 shows a hardware structure diagram of an exemplary computer device 400 provided in an embodiment of the present disclosure. The computer device 400 can be used to implement Figure 1A The terminal device 102 or 104 can also be used to implement Figure 1A In some scenarios, the computer device 400 can also be used to implement Figure 1A The database server 108.

[0095] like Figure 4 As shown, computer device 400 may include: processor 402, memory 404, network module 406, peripheral interface 408 and bus 410. Processor 402, memory 404, network module 406 and peripheral interface 408 are communicatively connected to each other within computer device 400 via bus 410.

[0096] The processor 402 may be a central processing unit (CPU), an image processor, a neural network processor (NPU), a microcontroller (MCU), a programmable logic device, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), or one or more integrated circuits. The processor 402 may be used to perform functions related to the technology described in this disclosure. In some embodiments, the processor 402 may also include multiple processors integrated into a single logical component. For example, Figure 4 As shown, processor 402 may include multiple processors 402a, 402b, and 402c.

[0097] The memory 404 may be configured to store data (eg, instructions, computer code, etc.). Figure 4 As shown, the data stored in the memory 404 may include program instructions (for example, program instructions for implementing the method 200 or 300 of the embodiment of the present disclosure) and data to be processed (for example, the memory may store configuration files of other modules, etc.). The processor 402 may also access the program instructions and data stored in the memory 404 and execute the program instructions to operate on the data to be processed. The memory 404 may include a volatile storage device or a non-volatile storage device. In some embodiments, the memory 404 may include a random access memory (RAM), a read-only memory (ROM), an optical disk, a magnetic disk, a hard disk, a solid-state drive (SSD), a flash memory, a memory stick, etc.

[0098] The network interface 406 can be configured to provide the computer device 400 with communication with other external devices via a network. The network can be any wired or wireless network capable of transmitting and receiving data. For example, the network can be a wired network, a local wireless network (e.g., Bluetooth, WiFi, near field communication (NFC)), a cellular network, the Internet, or a combination thereof. It will be understood that the type of network is not limited to the specific examples above.

[0099] The peripheral interface 408 can be configured to connect the computer device 400 to one or more peripheral devices to enable information input and output. For example, the peripheral devices can include input devices such as a keyboard, a mouse, a touchpad, a touch screen, a microphone, and various sensors, as well as output devices such as a display, a speaker, a vibrator, and an indicator light.

[0100] The bus 410 may be configured to transmit information between various components of the computer device 400 (e.g., the processor 402, the memory 404, the network interface 406, and the peripheral interface 408), such as an internal bus (e.g., a processor-memory bus), an external bus (USB port, PCI-E bus), etc.

[0101] It should be noted that although the architecture of the computer device 400 shown above only includes the processor 402, memory 404, network interface 406, peripheral interface 408, and bus 410, in a specific implementation, the architecture of the computer device 400 may also include other components necessary for normal operation. In addition, those skilled in the art will understand that the architecture of the computer device 400 may only include the components necessary to implement the embodiments of the present disclosure, and does not necessarily include all the components shown in the figure.

[0102] The embodiment of the present disclosure also provides a content preloading device. Figure 5 FIG. 5 shows a schematic diagram of an exemplary device 500 provided by an embodiment of the present disclosure. Figure 5 As shown, the apparatus 500 may be used to implement the method 200 or 300, applied to the terminal device 102 or 104, and may further include the following modules.

[0103] The receiving module 502 is configured to: receive first data generated when a user browses a first page of the content, wherein the first data includes information related to at least one of a content dimension, a device dimension, and a user dimension;

[0104] The prediction module 504 is configured to: call a pre-trained target model stored in the terminal device, input the first data into the target model, and output prediction information corresponding to the first data;

[0105] The preloading module 506 is configured to preload a second page of the content corresponding to the prediction information.

[0106] In some embodiments, the prediction module 504 is configured to: in response to the user browsing to a predetermined location of the first page, input the first data into a pre-trained target model, and output prediction information corresponding to the first data.

[0107] In some embodiments, the target model is a decision tree model, and the prediction information is the probability of predicting the first data as a positive sample; the preloading module 506 is configured to: preload a second page of the content corresponding to the probability.

[0108] In some embodiments, the preloading module 506 is configured to: in response to the probability being greater than a first threshold, preload a first quantity of a second page of the content; in response to the probability being greater than a second threshold and less than or equal to the first threshold, preload a second quantity of a second page of the content; or in response to the probability being less than or equal to the second threshold, stop preloading the content.

[0109] In some embodiments, the target model is a model trained using second data generated when the user browses a third page of the content or other content.

[0110] In some embodiments, the target model is pre-trained in a server, and the target model is stored in the terminal device in the form of a software package.

[0111] In some embodiments, the second data is labeled with a positive sample label or a negative sample label.

[0112] In some embodiments, the positive sample label or the negative sample label is marked in the following manner: in response to the user having browsed the next page of the third page, the second data is marked as a positive sample; or in response to the user not browsing the next page of the third page, the second data is marked as a negative sample.

[0113] In some embodiments, the first data or the second data includes device information and / or content information.

[0114] In some embodiments, the device information includes at least one of device power level, network type, device model, and device location information.

[0115] In some embodiments, the content information includes at least one of the type of content viewed by the user, the amount of content viewed, whether the user viewed the details of the content, interaction information with the content, and the duration of the content viewing.

[0116] In some embodiments, the content is a book, the type of the browsed content comprises a category of the browsed book, the number of the browsed content comprises a number of the browsed book, the whether to view the details of the content comprises whether to view a detail page of the book, the interaction information with the content comprises a comment, a like, a collection made on the book, and the time length of browsing the content comprises a time length of reading the book.

[0117] For ease of description, the above apparatus is described in various modules in terms of functions. Of course, the functions of the modules can be implemented in one or more software and / or hardware when implementing the present disclosure.

[0118] The apparatus of the above embodiments is used to implement the corresponding method 200 or 300 in any of the preceding embodiments, and has the beneficial effects of the corresponding method embodiments, which are not repeated here.

[0119] Based on the same inventive concept, the present disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the method 200 or 300 of any of the above embodiments.

[0120] The computer-readable medium of the present embodiment includes permanent and non-permanent, removable and non-removable media, which can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0121] The storage medium of the above embodiments stores computer instructions for causing the computer to perform the method 200 or 300 of any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which are not repeated here.

[0122] Based on the same inventive concept, the disclosure also provides a computer program product corresponding to any of the above-mentioned embodiment methods 200 or 300, comprising a computer program. In some embodiments, the computer program is executable by one or more processors to cause the processor to perform the method 200 or 300. Corresponding to the execution subject of each step in each embodiment of the method 200 or 300, the processor performing the corresponding step can belong to the corresponding execution subject.

[0123] The computer program product of the above-mentioned embodiments is used to make the processor perform the method 200 or 300 as described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which are not described here again.

[0124] It should be understood by those of ordinary skill in the art that the discussion of any of the above embodiments is only exemplary and is not intended to imply that the scope (including claims) of the disclosure is limited to these examples; the above embodiments or technical features between different embodiments can also be combined, the steps can be implemented in any order, and there are many other changes of different aspects of the embodiments of the disclosure as described above. In order to be brief, they are not provided in detail.

[0125] In addition, in order to simplify the description and discussion, and so as not to make the embodiments of the disclosure difficult to understand, the well-known power / ground connections of integrated circuit (IC) chips and other components can or can not be shown in the provided drawings. In addition, the devices can be shown in the form of block diagrams in order to avoid making the embodiments of the disclosure difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform to be implemented the embodiments of the disclosure (i.e. these details should be fully within the understanding of those skilled in the art). Where specific details (e.g. circuits) are set forth in order to describe an exemplary embodiment of the disclosure, it will be apparent to those skilled in the art that the embodiments of the disclosure can be implemented without these specific details or with variations on these specific details. Therefore, these descriptions should be considered illustrative rather than limiting.

[0126] Although the disclosure has been described in conjunction with specific embodiments thereof, many alternatives, modifications and variations will be apparent to those skilled in the art in light of the foregoing description. For example, other memory architectures (e.g. dynamic RAM (DRAM)) can use the embodiments discussed.

[0127] The embodiments of the present disclosure are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A content preloading method, applied to a terminal device, comprising: receiving first data generated by a user browsing a first page of the content, where the content includes a book, and the first data includes information related to a content dimension and / or a device dimension; calling a pre-trained target model stored in the terminal device, inputting the first data into the target model, and outputting prediction information corresponding to the first data; preloading a second page of the content corresponding to the prediction information; The prediction information is the probability of predicting the first data as a positive sample; The preloading of the second page corresponding to the prediction information of the content includes: preloading a number of second pages of the content corresponding to the probability; The number is positively correlated with the probability, the second pages corresponding to the probability are consecutive pages following the first page in the content, and the second pages corresponding to the probability and the first page are consecutive pages in the content.

2. The method according to claim 1, wherein Inputting the first data into the target model and outputting prediction information corresponding to the first data includes: In response to the user browsing to a predetermined position of the first page, the first data is input into the target model, and prediction information corresponding to the first data is output.

3. The method according to claim 1, wherein The target model is a decision tree model.

4. The method according to claim 3, wherein: The preloading of the second page corresponding to the probability of the content includes: In response to the probability being greater than a first threshold, preloading a first number of second pages of the content; In response to the probability being greater than a second threshold and less than or equal to the first threshold, preloading a second number of second pages of the content; or In response to the probability being less than or equal to the second threshold, preloading the content is stopped.

5. The method according to claim 1, wherein The target model is a model trained using second data generated when the user browses the content or a third page of other content.

6. The method according to claim 5, wherein: The target model is pre-trained in the server, and the target model is stored in the terminal device in the form of a software package.

7. The method according to claim 5, wherein: The second data is labeled with a positive sample label or a negative sample label.

8. The method of claim 7, wherein: The positive sample label or the negative sample label is marked in the following manner: In response to the user browsing the next page of the third page, marking the second data as a positive sample; or In response to the user not browsing the next page of the third page, marking the second data as a negative sample.

9. The method of claim 1, wherein: The first data includes device information and / or content information.

10. The method of claim 5, wherein: The second data includes device information and / or content information.

11. The method according to claim 9 or 10, wherein: The device information includes at least one of the device power level, network type, device model, and device location information; the content information includes at least one of the type of content browsed by the user, the amount of content browsed, whether the details of the content were viewed, interaction information with the content, and the length of time the content was browsed.

12. The method of claim 11, wherein: The content is a book, the type of the browsed content includes the classification of the browsed books, the number of browsed contents includes the number of browsed books, whether to view the details of the content includes whether to view the details page of the book, the interaction information with the content includes comments, likes, and collections of the book, and the duration of browsing the content includes the duration of reading the book.

13. A content preloading device, applied to a terminal device, comprising: a receiving module configured to: receive first data generated when a user browses a first page of the content, wherein the content includes a book, and the first data includes information related to at least one of a content dimension, a device dimension, and a user dimension; a prediction module configured to: call a pre-trained target model stored in the terminal device, input the first data into the target model, and output prediction information corresponding to the first data; a preloading module, configured to: preload a second page of the content corresponding to the prediction information; The prediction information is the probability of predicting the first data as a positive sample; The preloading module is configured to: preload a number of second pages of the content corresponding to the probability; The number is positively correlated with the probability, the second pages corresponding to the probability are consecutive pages following the first page in the content, and the second pages corresponding to the probability and the first page are consecutive pages in the content.

14. A computer device comprising one or more processors, a memory; and one or more programs, wherein the one or more programs are stored in the memory and executed by the one or more processors, the programs comprising instructions for executing the method according to any one of claims 1 to 12.

15. A non-transitory computer-readable storage medium containing a computer program, which, when executed by one or more processors, causes the processors to perform the method according to any one of claims 1 to 12.

16. A computer program product comprising computer program instructions, which, when executed on a computer, cause the computer to perform the method according to any one of claims 1 to 12.

Images