Data loading method, device, target server, target terminal and storage medium

Abstract

The embodiment of the present invention discloses a data loading method, device, target server, target terminal and storage medium. The method includes: obtaining historical item information of historical items corresponding to the historical operation behavior of the target client, and candidate item information of each candidate item in the item pool; determining the target item from each candidate item according to the similarity between the historical item information and each candidate item information, and determining the inventory unit information of the target item; sending the inventory unit information to the target client, so that the target client determines the web page address of the hypertext markup language web page for displaying the item detail data of the target item according to the inventory unit information, and loads the item detail data according to the web page address based on the web page view component embedded in the target client. The technical solution of the embodiment of the present invention improves the data loading speed of the web page view component, while ensuring a high cache hit rate and resource utilization.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of computer technology, and in particular to a data loading method, device, target server, target terminal and storage medium. Background Art

[0002] Hyper Text Markup Language 5 (HTML5) can be abbreviated as H5. It has the characteristics of short development cycle and good flexibility. Therefore, most Android apps are embedded with Android's web view (Webview) component for app development. The Webview component helps the app to better display H5 web pages. The app can be understood as a client (i.e. software) integrated on the terminal.

[0003] In the process of implementing the present invention, the inventors found that the following technical problems exist in the prior art: the data loading speed is slow when loading data based on the Webview component, which affects the smoothness of the H5 web page display, and further affects the user experience when browsing the H5 web page. Summary of the invention

[0004] The embodiments of the present invention provide a data loading method, device, target server, target terminal and storage medium, which solve the problem of slow data loading speed of Webview component.

[0005] In a first aspect, an embodiment of the present invention provides a data loading method, which may include:

[0006] Acquire historical item information of historical items corresponding to historical operation behaviors of the target client, and candidate item information of each candidate item in the item pool;

[0007] According to the similarity between the historical item information and the information of each candidate item, a target item is determined from each candidate item, and the inventory unit information of the target item is determined;

[0008] The inventory unit information is sent to the target client, so that the target client determines the web address of the hypertext markup language web page for displaying the item detail data of the target item according to the inventory unit information, and loads the item detail data according to the web address based on the web view component embedded in the target client.

[0009] In a second aspect, an embodiment of the present invention further provides a data loading method, which may include:

[0010] Acquire the inventory unit information of the target item corresponding to the target client, wherein the target item is an item determined by the target server from each candidate item according to the similarity between the historical item information of the historical item corresponding to the historical operation behavior of the target client and the candidate item information of each candidate item in the item pool;

[0011] The webpage address of a hypertext markup language webpage for displaying item detail data of the target item is determined according to the inventory unit information, and the item detail data is loaded according to the webpage address based on a webpage view component embedded in the target client.

[0012] In a third aspect, an embodiment of the present invention further provides a data loading device, which may include:

[0013] An item information acquisition module, used to acquire historical item information of historical items corresponding to historical operation behaviors of a target client, and candidate item information of each candidate item in an item pool;

[0014] A stock quantity unit information determination module is used to determine a target item from each candidate item according to the similarity between the historical item information and each candidate item information, and determine the stock quantity unit information of the target item;

[0015] The first item detail data loading module is used to send the inventory unit information to the target client, so that the target client determines the web address of the hypertext markup language web page used to display the item detail data of the target item based on the inventory unit information, and loads the item detail data according to the web address based on the web view component embedded in the target client.

[0016] In a fourth aspect, an embodiment of the present invention further provides a data loading device, which may include:

[0017] a stock keeping unit information acquisition module, used to acquire the stock keeping unit information of a target item corresponding to a target client, wherein the target item is an item determined by a target server from among candidate items according to the similarity between historical item information of historical items corresponding to historical operation behaviors of the target client and candidate item information of each candidate item in an item pool;

[0018] The second item detail data loading module is used to determine the web page address of the hypertext markup language web page for displaying the item detail data of the target item according to the inventory unit information, and load the item detail data according to the web page address based on the web page view component embedded in the target client.

[0019] In a fifth aspect, an embodiment of the present invention further provides a target server, which may include:

[0020] one or more processors;

[0021] A memory for storing one or more programs;

[0022] When one or more programs are executed by one or more processors, the one or more processors implement the data loading method provided in the first or second embodiment of the present invention.

[0023] In a sixth aspect, an embodiment of the present invention further provides a target terminal, which may include:

[0024] one or more processors;

[0025] A memory for storing one or more programs;

[0026] When one or more programs are executed by one or more processors, the one or more processors implement the data loading method provided in the third embodiment of the present invention.

[0027] In a seventh aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the data loading method provided by any embodiment of the present invention.

[0028] The technical solution of the embodiment of the present invention can determine the target item whose item details data is likely to be displayed by the target client from among the candidate items by acquiring the historical item information of the historical items corresponding to the historical operation behavior of the target client and the similarity between the candidate item information of each candidate item in the item pool, and determine the inventory unit information of the target item, and there is a one-to-one correspondence between the inventory unit information and the web page address of the hypertext markup language web page used to display the item details data of the target item; then, the inventory unit information is sent to the target client, so that the target client determines the corresponding web page address according to the inventory unit information, and pre-loads (i.e., caches) the item details data of the target item according to the web page address based on the web page view component embedded in itself, so that when the target client has a loading demand for the cached item details data, it does not need to obtain data from the network, and can directly load the data, thereby improving the data loading speed of the web page view component, and because the cached item details data is the item details data that is likely to be loaded, a high cache hit rate and resource utilization rate are guaranteed. The above technical solution achieves the effect of improving the data loading speed of the web view component and ensures a high cache hit rate and resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a flow chart of a data loading method in Embodiment 1 of the present invention;

[0030] Figure 2is a flow chart of a data loading method in Embodiment 2 of the present invention;

[0031] Figure 3 is a flow chart of a data loading method in Embodiment 3 of the present invention;

[0032] Figure 4 is a schematic diagram of an optional example of a data loading method in Embodiment 3 of the present invention;

[0033] Figure 5 is a structural block diagram of a data loading device in Embodiment 4 of the present invention;

[0034] Figure 6 is a structural block diagram of a data loading device in Embodiment 5 of the present invention;

[0035] Figure 7 This is a schematic diagram of the structure of a target server in Embodiment 6 of the present invention.

[0036] Figure 8 It is a structural diagram of a target terminal in Embodiment 7 of the present invention. DETAILED DESCRIPTION

[0037] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It is also necessary to explain that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0038] Before introducing the embodiment of the present invention, the application scenario of the embodiment of the present invention is first exemplified: In order to improve the data loading speed of the Webview component, an optional solution is to pre-load the data required for the H5 web page from the network to the hardware memory of the terminal. In this way, the Webview component can load the data directly from the hardware memory when loading these data, without requesting data from the network, thereby improving the data loading speed. It should be noted that the number of items in the e-commerce field is often in the tens of millions. In the case of limited storage capacity of the hardware memory, only the data required for the H5 web page of some items can be pre-loaded (i.e. cached), but these cached data are not necessarily the data that the user needs to load later. At this time, it is easy to have problems such as low cache hit rate and low resource utilization.

[0039] Embodiment 1

[0040] Figure 1It is a flow chart of a data loading method provided in the first embodiment of the present invention. This embodiment can be applied to the case of accelerating the data loading speed of the Webview component, and is particularly applicable to the case of improving the cache hit rate and resource utilization while accelerating the data loading speed of the Webview component. The method can be executed by the data loading device provided in the embodiment of the present invention, and the device can be implemented by software and / or hardware, and the device can be integrated on the target server.

[0041] See also Figure 1 The method of the embodiment of the present invention specifically includes the following steps:

[0042] S110: Acquire historical item information of historical items corresponding to historical operation behaviors of the target client and candidate item information of each candidate item in the item pool.

[0043] The target client may be an application (ie, software, App) deployed on a target terminal, and the target terminal may be a mobile terminal such as a mobile phone, a tablet computer, a notebook, etc., or a fixed terminal such as a computer, etc. The historical operation behavior may be some operation behaviors of the target user on a historical item displayed by the target client within the historical time period, such as historical click behavior, historical cart addition behavior, historical search behavior, historical collection behavior, historical attention behavior, historical interest classification behavior, etc. The number of target users may be one, two or more, because there may be one, two or more target users operating the target client installed on the target terminal; the historical time period may be a preset time period in the past, or a preset time period in the past determined according to the number of preset historical operation behaviors, that is, the time period in the past in which a preset number of historical operation behaviors occurred is taken as the historical time period, etc.; the historical item may be an item displayed on the target client, specifically, the item detail data of the historical item is displayed on the target client in the form of an H5 web page, and the item detail data may be data used to introduce the historical item in detail. In actual applications, optionally, each historical operation behavior may correspond to one, two or more historical items, and each historical item has its own historical item information that can express some of its attributes, such as historical category information, historical price information, historical title information, historical category information, historical brand information, historical inventory unit information, etc.

[0044] The item pool can be a pre-generated warehouse for storing various candidate items, and each candidate item has its own candidate item information that can represent some of its attributes, such as candidate category information, candidate price information, candidate title information, candidate category information, candidate brand information, candidate inventory unit information, etc.

[0045] It should be noted that, firstly, both “historical items” and “candidate items” are items, and here they are only distinguished by naming as items in different scenarios; similarly, both “historical item information” and “candidate item information” are item information of items, and here they are also distinguished by naming as item information of items in different scenarios. Secondly, among various item information, the stock keeping unit identification (skuID) is the most important item information, which has a one-to-one correspondence with the web page address (Uniform Resource Locator, URL) of the H5 web page used to display the item detail data of the corresponding item, that is, the URL corresponding to it can be determined according to the skuID. Thirdly, the historical item information and the candidate item information can be stored in the same or different storage spaces of the target server, and the target server can be a server associated with the target client, so the corresponding historical item information and / or candidate item information can be obtained from the storage space. In addition, optionally, since the target user may have a large number of historical operating behaviors on the target client, this means that the amount of historical item information may also be very large. In order to obtain historical item information that is very consistent with the target user's recent interests, the historical time period can be determined based on the current time and a pre-set historical time interval to determine the historical item information of the historical items corresponding to the historical operating behaviors of the target client within the historical time period, and the historical item information can present the item information of the user's recent preference items.

[0046] S120 . Determine a target item from among the candidate items based on the similarity between the historical item information and the information of each candidate item, and determine the inventory unit information of the target item.

[0047] Among them, since the historical item information can be the item information of the historical items that the target user has visited or is likely to visit, such as the item information of the historical items visited through historical click behavior, and the item information of the historical items searched through historical search behavior, these are the item information of the historical items that the target user is more interested in; the candidate item information can be the item information of the candidate items that may be visited by the user in the future. Of course, the probability of different candidate items being visited by the target user is different. Therefore, for each historical item information, the similarity between the historical item information and each candidate item information can be determined respectively, and the target item is determined from each candidate item according to each similarity. Then, the target item can be a candidate item with candidate item information that is relatively similar to the historical item information. Such a target item is a candidate item that is likely to be visited by the target user. The number of the target items can be one, two or more, which is not specifically limited here. Further, the skuID of the URL of the H5 webpage used to display the item details data of the target item is determined from the various candidate item information of the target item.

[0048] S130. Send the inventory unit information to the target client, so that the target client determines the web address of the hypertext markup language web page for displaying the item detail data of the target item according to the inventory unit information, and loads the item detail data according to the web address based on the web view component embedded in the target client.

[0049] Among them, there are many ways to implement the target server sending the skuID to the target client, such as sending the skuID to the hardware memory of the target terminal integrated with the target client, and providing a target interface for reading the skuID, so that the target client can read the skuID from the hardware memory through the target interface; for example, storing the skuID in the hardware memory of the target server integrated with the target server, and providing a target interface for reading the skuID, so that the target client can read the skuID from the hardware memory through the target interface, or the target client sends an interface request through the target interface, so that the target server that receives the interface request can push the skuID to the target client. Compared with the former, the latter method of actively pulling the skuID from the target server ensures the real-time nature of the skuID.

[0050] After receiving the skuID sent by the target server, the target client can determine the URL corresponding to it, and then load the item details data corresponding to the URL based on the Webview component embedded in itself, that is, control the Webview component to find the corresponding item details data from the network according to the URL, and the item details data can be data used to introduce the target item in detail, and then pre-load (i.e. cache) the item details data into the hardware memory of the target terminal. In this way, if the item details data to be loaded by the subsequent target client are exactly these cached item details data, then the item details data to be loaded can be directly obtained from the cached item details data, and the obtained item details data can be loaded. At this time, there is no need to obtain these item details data to be loaded from the network, thereby improving the data loading speed of the Webview component, and the increase in data loading speed means the increase in the rendering speed of the H5 web page, thereby ensuring the experience of the target user when browsing the H5 web page, and thus solving the problem of item conversion rate loss caused by the need for the target user to wait for a long time for the rendering of the H5 web page. In addition, as can be seen from the foregoing, these cached item detail data are mostly item detail data of target items that the target user is interested in, that is, they are likely to be item detail data that need to be loaded in the future, thereby effectively ensuring a high cache hit rate and resource utilization.

[0051] It should be noted that the above technical solution can be executed once at a preset time interval, such as once a week; it can also be executed once when the item detail data corresponding to the cached item detail data in the network changes. It is worth noting that when the item detail data changes, the corresponding URL may also change. This is because the URL is an address dynamically spliced ​​out by the network when receiving a data request. Therefore, it can also be executed once when the URL corresponding to the cached item detail data changes.

[0052] The technical solution of the embodiment of the present invention can determine the target item whose item details data is likely to be displayed by the target client from among the candidate items by acquiring the historical item information of the historical items corresponding to the historical operation behavior of the target client and the similarity between the candidate item information of each candidate item in the item pool, and determine the inventory unit information of the target item, and there is a one-to-one correspondence between the inventory unit information and the web page address of the hypertext markup language web page used to display the item details data of the target item; then, the inventory unit information is sent to the target client, so that the target client determines the corresponding web page address according to the inventory unit information, and pre-loads (i.e., caches) the item details data of the target item according to the web page address based on the web page view component embedded in itself, so that when the target client has a loading demand for the cached item details data, it does not need to obtain data from the network, and can directly load the data, thereby improving the data loading speed of the web page view component, and because the cached item details data is the item details data that is likely to be loaded, a high cache hit rate and resource utilization rate are guaranteed. The above technical solution achieves the effect of improving the data loading speed of the web view component and ensures a high cache hit rate and resource utilization.

[0053] Embodiment 2

[0054] Figure 2 It is a flow chart of a data loading method provided in the second embodiment of the present invention. This embodiment is optimized based on the above-mentioned technical solutions. In this embodiment, optionally, according to the similarity between the historical item information and the candidate item information, the target item is determined from the candidate items, including: pre-processing the historical item information and the candidate item information based on the deep learning framework to obtain the historical item feature set and the candidate item feature set that match the deep learning framework, the historical item feature set includes the historical item features corresponding to each historical item information, and the candidate item feature set includes the candidate item features corresponding to each candidate item information; for each historical item feature, the similarity between the historical item feature and the candidate item feature is determined based on the deep learning framework, and the target item similar to the historical item with the historical item feature is determined from the candidate items according to the similarities. The explanations of the terms that are the same or corresponding to the above-mentioned embodiments are not repeated here.

[0055] See also Figure 2 The method of this embodiment may specifically include the following steps:

[0056] S210: Acquire historical item information of historical items corresponding to historical operation behaviors of the target client and candidate item information of each candidate item in the item pool.

[0057] S220, preprocessing each historical item information and each candidate item information based on a deep learning framework to obtain a historical item feature set and a candidate item feature set that match the deep learning framework, wherein the historical item feature set includes historical item features corresponding to each historical item information, and the candidate item feature set includes candidate item features corresponding to each candidate item information.

[0058] Among them, the information types of the historical item information corresponding to each historical operation behavior may be different. For example, when the historical operation behavior is a historical click behavior, since the historical click behavior can correspond to a specific historical item, the historical item information at this time can be the item information after at least one combination of the historical category information, historical price information, historical title information, historical category information, historical brand information and historical inventory unit information of the historical item; when the historical operation behavior is a historical search behavior, since the number of historical items searched by the App according to the search term corresponding to the historical search behavior can be one, two or more, the historical item information at this time can be the item information that can show the commonality of the at least one historical item, such as the item information after the combination of category information and brand information; etc. On this basis, in order to better analyze the historical item information of each historical item, the historical item information can be pre-processed based on the deep learning framework to obtain a historical item feature set that can match the deep learning framework, that is, to generate a historical item feature set that can be processed by the deep learning framework, and each historical item feature in the historical item feature set has a historical item information corresponding to it one by one. Similarly, the candidate item information can also be preprocessed based on the deep learning framework, thereby obtaining a candidate item feature set including candidate item features corresponding to each candidate item information. The above deep learning framework can be TenserFlow, Caffe, etc., and the item features obtained by preprocessing based on these deep learning frameworks can be features that match their data formats.

[0059] S230. For each historical item feature, determine the similarity between the historical item feature and each candidate item feature based on the deep learning framework, and determine a target item similar to the historical item having the historical item feature from each candidate item according to each similarity.

[0060] Among them, since the historical item features and the candidate item features are both item features that have been processed by the deep learning framework and can be recognized by it, such as item features with the same data format, the similarity between the two can be directly compared based on the deep learning framework, so as to determine the target item similar to the historical item with the historical item features from each candidate item, that is, the candidate item features that are relatively similar to the historical item features can be used as the target item. In practical applications, the similarity determination process and comparison process can be optionally implemented by a variety of algorithms such as clustering algorithms and classification algorithms in the deep learning framework, which are not specifically limited here.

[0061] S240. Send the inventory unit information to the target client, so that the target client determines the web address of the hypertext markup language web page for displaying the item detail data of the target item according to the inventory unit information, and loads the item detail data according to the web address based on the web view component embedded in the target client.

[0062] The technical solution of the embodiment of the present invention pre-processes each historical item information and each candidate item information through a deep learning framework, and obtains a historical item feature set and a candidate item feature set that match the deep learning framework, that is, obtains historical item features that can be processed by the deep learning framework and correspond to each historical item information, and candidate item features that correspond to each candidate item information; thus, for each historical item feature, the similarity between the historical item feature and each candidate item feature can be determined based on the deep learning framework, and a target item similar to the historical item having the historical item feature is determined from each candidate item according to each similarity, thereby achieving the effect of accurately screening the target item.

[0063] An optional technical solution, according to each similarity, determines the target item similar to the historical item with the historical item characteristics from each candidate item, which may include: obtaining a preset neighbor parameter, sorting the candidate item features corresponding to each similarity based on the numerical value of each similarity, selecting the target item features whose number is the neighbor parameter from the sorting result; and taking the candidate items with the target item characteristics from each candidate item as the target item similar to the historical item with the historical item characteristics. Among them, the K-Nearest Neighbor (KNN) clustering algorithm in the deep learning framework has the advantages of being simple and easy to use, short model training time, good prediction effect, and insensitivity to outliers. Therefore, the KNN clustering algorithm can be used here to cluster the target items that are relatively similar to the historical items from each candidate item. Specifically, obtain the specific value of the preset neighbor parameter, i.e., K, and each similarity. On this basis, first sort the candidate item features corresponding to each similarity, such as sorting the corresponding candidate item features from large to small or from small to large according to the numerical value of each similarity, and then select the target item features whose number is the neighbor parameter from the sorting result, such as taking the candidate item features of the neighbor parameter with larger similarity as the target item features, and taking the candidate items corresponding to these target item features as the target items, thereby achieving the effect of fast and accurate screening of the target items. Exemplarily, taking KNN in TenserFlow as an example, the historical item features and each candidate item feature can be points in a two-dimensional plane map. The specific position of the point corresponding to a candidate item feature in the two-dimensional plane map is related to the similarity between the candidate item feature and the historical item feature. Draw a circle with the point corresponding to the historical item feature as the center and the neighbor parameter as the radius. The candidate items corresponding to the points in the circle except the center are the target items.

[0064] Embodiment 3

[0065] Figure 3 It is a flow chart of a data loading method provided in the third embodiment of the present invention. This embodiment can be applied to the case of accelerating the data loading speed of the Webview component, and is particularly applicable to the case of improving the cache hit rate and resource utilization while accelerating the data loading speed of the Webview component. The method can be executed by the data loading device provided in the embodiment of the present invention, and the device can be implemented by software and / or hardware, and the device can be integrated on the target terminal.

[0066] See also Figure 3 The method of the embodiment of the present invention specifically includes the following steps:

[0067] S310, obtaining inventory unit information of a target item corresponding to a target client, wherein the target item is an item determined by a target server from among candidate items based on similarity between historical item information of historical items corresponding to historical operation behaviors of the target client and candidate item information of each candidate item in an item pool.

[0068] Among them, the meanings of the terms such as target item, target server, target client, historical operation behavior, historical item, historical item information, item pool, candidate item, candidate item information, similarity, etc. have been described in detail in the above embodiments, and will not be repeated here. On this basis, the specific way in which the target client obtains the skuID of the target item corresponding to itself depends on the way in which the target server sends the skuID to the target client. For example, if the target server sends the skuID to the hardware memory of the target terminal integrated with the target client and provides a target interface for reading the skuID, then the target client can read the skuID by obtaining a target interface for reading the skuID of the target item corresponding to itself and reading the skuID from the target terminal based on the target interface; if the target server sends the skuID by storing the skuID in the hardware memory of the target server integrated with itself and providing a target interface for reading the skuID, then the target client can read the skuID by obtaining a target interface for reading the skuID of the target item corresponding to itself and reading the skuID from the target server based on the target interface; etc., which are not specifically limited here.

[0069] S320: Determine a webpage address of a hypertext markup language webpage for displaying item detail data of the target item according to the inventory unit information, and load the item detail data according to the webpage address based on a webpage view component embedded in the target client.

[0070] Among them, since the skuID and the URL of the H5 webpage for displaying the item details data of the target item with the skuID have a one-to-one correspondence, the URL corresponding to it can be determined according to the skuID. Then, the item details data corresponding to the URL is loaded based on the Webview component embedded in itself, that is, the Webview component is controlled to find the corresponding item details data from the network according to the URL, and then the item details data is pre-loaded (i.e., cached) into the hardware memory of the target terminal. In this way, if the item details data to be loaded by the subsequent target client are exactly these cached item details data, then the item details data to be loaded can be directly obtained from the cached item details data, and the obtained item details data can be loaded. At this time, it is no longer necessary to obtain these item details data to be loaded from the network, thereby improving the data loading speed of the Webview component, and the improvement of the data loading speed means the improvement of the rendering speed of the H5 webpage, thereby ensuring the experience of the target user when browsing the H5 webpage, and thus solving the problem of item conversion rate loss caused by the need for the target user to wait for a long time for the rendering of the H5 webpage. In addition, as can be seen from the foregoing, these cached item detail data are mostly item detail data of target items that the target user is interested in, that is, they are likely to be item detail data that need to be loaded in the future, thereby ensuring a high cache hit rate and resource utilization.

[0071] The technical solution of the embodiment of the present invention can determine the webpage address of the hypertext markup language webpage for displaying the item details data of the target item by obtaining the inventory unit information of the target item corresponding to the target client. The target item can be determined by the target server from the candidate items according to the historical item information of the historical item corresponding to the historical operation behavior of the target client and the similarity between the candidate item information of each candidate item in the item pool, and the item details data of the target item is likely to be displayed by the target client; then, based on the webpage view component embedded in itself, the item details data of the target item is pre-loaded (i.e., cached) according to the webpage address, so that when the target client has a loading demand for the cached item details data, it does not need to obtain data from the network, and can directly load the data, thereby improving the data loading speed of the webpage view component, and because the cached item details data is the item details data that is likely to be loaded, a high cache hit rate and resource utilization rate are guaranteed. The above technical solution achieves the effect of improving the data loading speed of the webpage view component and guarantees a high cache hit rate and resource utilization rate.

[0072] An optional technical solution, based on a web view component embedded in a target client, loads item detail data according to a web address, which may include: based on the web view component embedded in the target client, loading the item detail data into a hardware memory of a target terminal integrated with the target client according to the web address; when a current operation behavior acting on the target client is monitored, obtaining an address to be loaded corresponding to the current operation behavior, and determining whether there is data to be loaded corresponding to the address to be loaded in the item detail data loaded into the hardware memory; if so, loading the data to be loaded from the hardware memory into the target client, and rendering the item detail data loaded into the target client to display a hypertext markup language web page, wherein the hypertext markup language web page includes a web page for displaying the item detail data loaded into the target client.

[0073] Among them, the hardware memory can be a memory set on the target terminal by hardware, such as a built-in read-only memory (ROM) in the target terminal, an external secure digital memory card (SD card), etc. The current operation behavior can be an operation behavior performed by the target user on a certain item among multiple items displayed on a certain H5 webpage of the target client at the current moment. After monitoring such a current operation behavior, the address to be loaded corresponding to the current operation behavior can be obtained, and the address to be loaded can be the URL of the H5 webpage used to display the item detail data of the certain item. Further, if there is data to be loaded corresponding to the address to be loaded in the item detail data loaded into the hardware memory, the data to be loaded can be the item detail data of the certain item, then the data to be loaded can be directly loaded from the hardware memory to the target client, that is, the data to be loaded is directly obtained from the hardware memory, and the obtained data to be loaded is loaded into the target client. Of course, if the item details data loaded into the hardware memory does not contain the data to be loaded corresponding to the address to be loaded, the data to be loaded can only be obtained from the network according to the URL request, and the obtained data to be loaded is loaded into the target client. Furthermore, the item details data loaded on the target client is rendered to display the H5 webpage for displaying the item details data loaded on the target client, thereby achieving the effect of completing the H5 webpage display through the Webview component.

[0074] An optional technical solution, after loading the item details data according to the web page address based on the web page view component embedded in the target client, the loaded item details data can be the item details data of the target item that the target user is more interested in and is likely to be willing to obtain, then in order to improve the item conversion rate and further improve the cache hit rate, the above data loading method also includes: obtaining the information to be pushed from each inventory unit information, and pushing the information to be pushed to the push position set on the target client, wherein the information to be pushed can be one, two or more skuIDs among each skuID. In this way, the target user can browse the information to be pushed on the push position of a certain H5 web page displayed by the target client, and is likely to click on the link of the target item corresponding to the information to be pushed. When the target user clicks on the link, the target client can obtain the item details data of the target item from the loaded item details data, and load the acquired item details data again, thereby further improving the cache hit rate through active push, and obtaining an offline intelligent push method based on the target client.

[0075] In order to better understand the specific implementation process of the above steps, the data loading method of each embodiment is exemplified below with reference to specific examples. For example, the loading process of product details data of a product on an e-commerce platform is taken as an example. Figure 4 , where the target server executes S1-S4, and the target client executes S5-S6. Specifically,

[0076] S1. Collect historical product information corresponding to the historical operation behavior of the target user to form a historical product information set. The data formats of each historical product information in the historical product information set may be different.

[0077] S2. Preprocess the historical product information set collected by S1 based on a deep learning framework such as TenserFlow, Caffe, etc. to obtain a historical product feature set that can match the data format of the deep learning framework; at the same time, preprocess the candidate product information of each candidate product in the product pool to obtain a candidate product feature set.

[0078] S3. For each historical commodity feature in the historical commodity feature set obtained by the preprocessing in S2, a KNN clustering algorithm is used to cluster the historical commodity features and the candidate commodity features in each candidate commodity feature set, and a target commodity that is relatively similar to the historical commodity is obtained from each candidate commodity.

[0079] S4. Arrange the target commodities extracted in S3 in chronological order, that is, select target commodities that are similar to the historical commodities corresponding to the recent historical operation behaviors from the target commodities extracted in S3, and determine the skuId of the latest target commodities. These target commodities can be used as recommended commodities.

[0080] S5. When the target user launches the App, the App actively initiates a request to call the target interface, and obtains a skuId list consisting of various skuIds based on the target interface; then, the corresponding URL is determined according to the skuId list, and the Webview component is initialized in the background so that the Webview component loads each URL into the local hardware memory in sequence.

[0081] S6. Recommend the target products corresponding to the skuId list obtained in S5 to the user, thereby forming an end-based offline intelligent recommendation system.

[0082] Embodiment 4

[0083] Figure 5 This is a structural block diagram of a data loading device provided in Embodiment 4 of the present invention. The device is used to execute the data loading method provided in Embodiment 1 or Embodiment 2. The device and the data loading method in the above embodiments belong to the same inventive concept. For details not described in detail in the embodiment of the data loading device, please refer to the embodiment of the above data loading method. Figure 5 The device is configured on the target server, and may specifically include: an item information acquisition module 410, an inventory unit information determination module 420, and a first item detail data loading module 430.

[0084] The item information acquisition module 410 is used to acquire the historical item information of the historical items corresponding to the historical operation behavior of the target client, and the candidate item information of each candidate item in the item pool;

[0085] The SKU information determination module 420 is used to determine the target item from the candidate items according to the similarity between the historical item information and the candidate item information, and determine the SKU information of the target item;

[0086] The first item details data loading module 430 is used to send the inventory unit information to the target client, so that the target client determines the web address of the hypertext markup language web page used to display the item details data of the target item based on the inventory unit information, and loads the item details data according to the web address based on the web view component embedded in the target client.

[0087] Optionally, the inventory unit information determination module 420 may specifically include:

[0088] The item feature set obtaining submodule is used to pre-process each historical item information and each candidate item information based on the deep learning framework to obtain a historical item feature set and a candidate item feature set that match the deep learning framework, wherein the historical item feature set includes historical item features corresponding to each historical item information, and the candidate item feature set includes candidate item features corresponding to each candidate item information;

[0089] The target item determination submodule is used to determine the similarity between the historical item feature and the candidate item features for each historical item feature based on the deep learning framework, and determine the target item similar to the historical item having the historical item feature from the candidate items according to the similarities.

[0090] On this basis, optionally, the target item determination submodule may specifically include:

[0091] The target item feature selection unit is used to obtain the preset neighbor parameters, sort the candidate item features corresponding to each similarity based on the numerical value of each similarity, and select the target item features whose number is the neighbor parameter from the sorting results;

[0092] The target item determination unit is used to select a candidate item having the target item feature among the candidate items as a target item similar to the historical item having the historical item feature.

[0093] Optionally, the item information acquisition module 410 may specifically include:

[0094] The historical item information acquisition unit is used to acquire the current time and a preset historical time interval, determine the historical time period according to the current time and the historical time interval, and determine the historical item information of the historical item corresponding to the historical operation behavior of the target client within the historical time period.

[0095] Optionally, historical operation behavior includes at least one of historical click behavior, historical cart addition behavior, historical search behavior, historical collection behavior, historical attention behavior and historical interest classification behavior, and / or historical item information includes at least one of historical category information, historical price information, historical title information, historical category information and historical brand information.

[0096] The data loading device provided in the fourth embodiment of the present invention cooperates with the item information acquisition module and the inventory unit information determination module. According to the historical item information of the historical items corresponding to the historical operation behavior of the target client and the similarity between the candidate item information of each candidate item in the item pool, the target item whose item details data is likely to be displayed by the target client can be determined from each candidate item, and the inventory unit information of the target item can be determined. There is a one-to-one correspondence between the inventory unit information and the web page address of the hypertext markup language web page for displaying the item details data of the target item. Then, the inventory unit information is sent to the target client through the first item details data loading module, so that the target client determines the corresponding web page address according to the inventory unit information, and pre-loads (i.e., caches) the item details data of the target item according to the web page address based on the web page view component embedded in itself. In this way, when the target client has a loading demand for the cached item details data, it does not need to obtain data from the network, and can directly load the data, thereby improving the data loading speed of the web page view component, and because the cached item details data is the item details data that is likely to be loaded, a high cache hit rate and resource utilization rate are guaranteed. The above device achieves the effect of improving the data loading speed of the web page view component and effectively ensures a high cache hit rate and resource utilization rate.

[0097] The data loading device provided in the embodiment of the present invention can execute the data loading method provided in the first embodiment or the second embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0098] It is worth noting that in the embodiment of the above-mentioned data loading device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0099] Embodiment 5

[0100] Figure 6 This is a structural block diagram of a data loading device provided in Embodiment 5 of the present invention, and the device is used to execute the data loading method provided in Embodiment 3 above. The device and the data loading method in the above embodiments belong to the same inventive concept, and the details not described in detail in the embodiment of the data loading device can be referred to the embodiment of the above data loading method. Figure 6 The device is configured on the target terminal and may specifically include: a stock quantity unit information acquisition module 510 and a second item detail data loading module 520.

[0101] The SKU information acquisition module 510 is used to acquire the SKU information of the target item corresponding to the target client, wherein the target item is an item determined by the target server from each candidate item according to the similarity between the historical item information of the historical item corresponding to the historical operation behavior of the target client and the candidate item information of each candidate item in the item pool;

[0102] The second item detail data loading module 520 is used to determine the webpage address of the hypertext markup language webpage for displaying the item detail data of the target item according to the inventory unit information, and load the item detail data according to the webpage address based on the webpage view component embedded in the target client.

[0103] Optionally, the inventory unit information acquisition module 510 may specifically include:

[0104] The inventory unit information acquisition unit is used to acquire a target interface for reading the inventory unit information of a target item corresponding to a target client, and read the inventory unit information from a target server integrated with a target server and / or a target terminal integrated with a target client based on the target interface; wherein the inventory unit information stored on the target terminal is information sent to the target terminal by the target server.

[0105] Optionally, the second item details data loading module 520 may specifically include:

[0106] The item details data first loading unit is used to load the item details data into the hardware memory of the target terminal integrated with the target client according to the web page address based on the web page view component embedded in the target client;

[0107] The item details data judging unit is used to obtain the to-be-loaded address corresponding to the current operation behavior when monitoring the current operation behavior acting on the target client, and to judge whether there is to-be-loaded data corresponding to the to-be-loaded address in the item details data loaded into the hardware memory;

[0108] The item detail data reloading unit is used to load the data to be loaded from the hardware memory to the target client, and render the item detail data loaded into the target client to display a hypertext markup language web page, wherein the hypertext markup language web page includes a web page for displaying the item detail data loaded into the target client.

[0109] Optionally, based on the above device, the device may further include:

[0110] The module for pushing information to be pushed is used to obtain the information to be pushed from the information of each inventory unit, and push the information to be pushed to the push position set on the target client.

[0111] The data loading device provided in the fifth embodiment of the present invention can determine the webpage address of the hypertext markup language webpage for displaying the item details data of the target item according to the obtained inventory unit information of the target item corresponding to the target client through the cooperation between the inventory unit information of the target item corresponding to the target client, and the target item can be determined by the target server from each candidate item according to the similarity between the historical item information of the historical item corresponding to the historical operation behavior of the target client and the candidate item information of each candidate item in the item pool, and the item details data of the target item is likely to be displayed by the target client; then, based on the webpage view component embedded in itself, the item details data of the target item is pre-loaded (i.e., cached) according to the webpage address, so that when the target client has a loading demand for the cached item details data, it does not need to obtain data from the network, and can directly load the data, thereby improving the data loading speed of the webpage view component, and because the cached item details data is the item details data that is likely to be loaded, a high cache hit rate and resource utilization rate are guaranteed. The above device achieves the effect of improving the data loading speed of the webpage view component and guarantees a high cache hit rate and resource utilization rate.

[0112] The data loading device provided in the embodiment of the present invention can execute the data loading method provided in the third embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0113] It is worth noting that in the embodiment of the above-mentioned data loading device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0114] Embodiment 6

[0115] Figure 7 A schematic diagram of the structure of a target server provided in Embodiment 6 of the present invention is shown in FIG. Figure 7 As shown, the target server includes a memory 610, a processor 620, an input device 630, and an output device 640. The number of processors 620 in the target server can be one or more. Figure 7 A processor 620 is used as an example; the memory 610, the processor 620, the input device 630 and the output device 640 in the target server can be connected via a bus or other means. Figure 7 The connection via bus 650 is taken as an example.

[0116] The memory 610, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the data loading method in the embodiment of the present invention (for example, the item information acquisition module 410, the inventory unit information determination module 420 and the first item details data loading module 430 in the data loading device). The processor 620 executes various functional applications and data processing of the target server by running the software programs, instructions and modules stored in the memory 610, that is, implements the above-mentioned data loading method.

[0117] The memory 610 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the target server, etc. In addition, the memory 610 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 610 may further include a memory remotely arranged relative to the processor 620, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0118] The input device 630 may be used to receive input digital or character information and generate key signal input related to user settings and function control of the device. The output device 640 may include display devices such as display screens.

[0119] Embodiment 7

[0120] Figure 8 A schematic diagram of the structure of a target terminal provided in Embodiment 7 of the present invention is shown in FIG. Figure 8 As shown, the target terminal includes a memory 710, a processor 720, an input device 730, and an output device 740. The number of processors 720 in the target terminal can be one or more. Figure 8 A processor 720 is taken as an example; the memory 710, the processor 720, the input device 730 and the output device 740 in the target terminal can be connected via a bus or other means. Figure 8 The connection via bus 750 is taken as an example.

[0121] The memory 710, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the target detection method in the embodiment of the present invention (for example, the inventory unit information acquisition module 510 and the second item details data loading module 520 in the target detection device). The processor 720 executes various functional applications and data processing of the target terminal by running the software programs, instructions and modules stored in the memory 710, that is, realizing the above-mentioned target detection method.

[0122] The memory 710 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the target terminal, etc. In addition, the memory 710 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 710 may further include a memory remotely arranged relative to the processor 720, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0123] The input device 730 may be used to receive input digital or character information and generate key signal input related to user settings and function control of the device. The output device 740 may include a display device such as a display screen.

[0124] Embodiment 8

[0125] Embodiment 8 of the present invention provides a storage medium containing computer executable instructions, wherein the computer executable instructions are used to execute a data loading method when executed by a computer processor, and may include:

[0126] Acquire historical item information of historical items corresponding to historical operation behaviors of the target client, and candidate item information of each candidate item in the item pool;

[0127] According to the similarity between the historical item information and the information of each candidate item, a target item is determined from each candidate item, and the inventory unit information of the target item is determined;

[0128] The inventory unit information is sent to the target client, so that the target client determines the web address of the hypertext markup language web page for displaying the item detail data of the target item according to the inventory unit information, and loads the item detail data according to the web address based on the web view component embedded in the target client.

[0129] Of course, the computer executable instructions of a storage medium including computer executable instructions provided by an embodiment of the present invention are not limited to the operations of the method described above, but can also execute related operations in the data loading method provided by any embodiment of the present invention.

[0130] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0131] Embodiment 9

[0132] Embodiment 9 of the present invention provides a storage medium containing computer executable instructions, which, when executed by a computer processor, is used to execute a data loading method, which may include:

[0133] Acquire the inventory unit information of the target item corresponding to the target client, wherein the target item is an item determined by the target server from each candidate item according to the similarity between the historical item information of the historical item corresponding to the historical operation behavior of the target client and the candidate item information of each candidate item in the item pool;

[0134] The webpage address of a hypertext markup language webpage for displaying item detail data of the target item is determined according to the inventory unit information, and the item detail data is loaded according to the webpage address based on a webpage view component embedded in the target client.

[0135] Of course, the computer executable instructions of a storage medium including computer executable instructions provided by an embodiment of the present invention are not limited to the operations of the method described above, but can also execute related operations in the data loading method provided by any embodiment of the present invention.

[0136] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0137] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A data loading method, characterized in that: include: Acquire historical item information of historical items corresponding to historical operation behaviors of the target client, and candidate item information of each candidate item in the item pool; Preprocessing each of the historical item information and each of the candidate item information based on a deep learning framework to obtain a historical item feature set and a candidate item feature set that match the deep learning framework, wherein the historical item feature set includes historical item features corresponding to each of the historical item information, and the candidate item feature set includes candidate item features corresponding to each of the candidate item information; For each of the historical item features, respectively determine the similarity between the historical item feature and each of the candidate item features based on the deep learning framework, and determine a target item similar to the historical item having the historical item feature from each of the candidate items according to each of the similarities, wherein the similarity comparison process is implemented by the deep learning framework; Determine the inventory unit information of the target item, and send the inventory unit information to the target client, so that the target client determines the web address of the hypertext markup language web page for displaying the item detail data of the target item according to the inventory unit information, and loads the item detail data according to the web address based on the web view component embedded in the target client.

2. The method according to claim 1, characterized in that: Determining a target item similar to the historical item having the historical item characteristics from the candidate items according to the similarities includes: Obtaining a preset neighbor parameter, sorting the candidate item features corresponding to each of the similarities based on the numerical values ​​of each of the similarities, and selecting target item features whose number is the neighbor parameter from the sorting result, wherein the neighbor parameter is the value of K in the K nearest neighbor clustering algorithm; The candidate items having the target item characteristics among the candidate items are taken as target items similar to the historical items having the historical item characteristics.

3. The method according to claim 1, characterized in that The step of obtaining historical item information of historical items corresponding to the historical operation behavior of the target client includes: The current time and a preset historical time interval are obtained, a historical time period is determined according to the current time and the historical time interval, and historical item information of historical items corresponding to historical operation behaviors of the target client within the historical time period is determined.

4. The method according to claim 1, characterized in that: The historical operation behavior includes at least one of historical click behavior, historical add-to-cart behavior, historical search behavior, historical collection behavior, historical attention behavior, and historical interest classification behavior, and / or the historical item information includes at least one of historical category information, historical price information, historical title information, historical category information, and historical brand information.

5. A data loading method, characterized in that: include: Acquire inventory unit information of a target item corresponding to a target client, wherein the target item is an item determined by a target server from among candidate items according to similarity between historical item information of historical items corresponding to historical operation behaviors of the target client and candidate item information of each candidate item in an item pool; Determine, based on the SKU information, a webpage address of a hypertext markup language webpage for displaying item detail data of the target item; Loading the item detail data into a hardware memory of a target terminal integrated with the target client according to the web page address based on a web page view component embedded in the target client; Obtaining an address to be loaded corresponding to a current operation behavior acting on the target client; In the case that the item detail data loaded into the hardware memory includes data to be loaded corresponding to the address to be loaded, the data to be loaded is loaded from the hardware memory to the target client to complete the loading of the item detail data corresponding to the data to be loaded.

6. The method according to claim 5, characterized in that The step of obtaining the inventory unit information of the target item corresponding to the target client includes: Acquire a target interface for reading the inventory unit information of the target item corresponding to the target client, and read the inventory unit information from a target server integrated with the target service end and / or a target terminal integrated with the target client based on the target interface; The inventory unit information stored in the target terminal is information sent to the target terminal by the target server.

7. The method according to claim 5, characterized in that After the loading of the item detail data corresponding to the data to be loaded is completed, the method further includes: The item detail data loaded into the target client is rendered to display the hypertext markup language web page, wherein the hypertext markup language web page includes a web page for displaying the item detail data loaded into the target client.

8. The method according to claim 5, characterized in that After the loading of the item detail data corresponding to the data to be loaded is completed, the method further includes: The information to be pushed is acquired from each of the inventory unit information, and the information to be pushed is pushed to a push position set on the target client.

9. A data loading device, characterized in that: include: An item information acquisition module, used to acquire historical item information of historical items corresponding to historical operation behaviors of a target client, and candidate item information of each candidate item in an item pool; a stock quantity unit information determination module, configured to determine a target item from each of the candidate items according to the similarity between the historical item information and each of the candidate item information, and determine the stock quantity unit information of the target item; A first item details data loading module, configured to send the SKU information to the target client, so that the target client determines the webpage address of a hypertext markup language webpage for displaying item details data of the target item according to the SKU information, and loads the item details data according to the webpage address based on a webpage view component embedded in the target client; Wherein, the inventory unit information determination module includes: An item feature set obtaining submodule is used to pre-process each of the historical item information and each of the candidate item information based on a deep learning framework to obtain a historical item feature set and a candidate item feature set that match the deep learning framework, wherein the historical item feature set includes historical item features corresponding to each of the historical item information, and the candidate item feature set includes candidate item features corresponding to each of the candidate item information; The target item determination submodule is used to determine, for each of the historical item features, the similarity between the historical item features and the candidate item features based on the deep learning framework, and determine, from the candidate items, a target item that is similar to the historical item having the historical item features, wherein the similarity comparison process is implemented through the deep learning framework.

10. A data loading device, characterized in that: include: a stock keeping unit information acquisition module, used to acquire the stock keeping unit information of a target item corresponding to a target client, wherein the target item is an item determined by a target server from among candidate items according to the similarity between historical item information of historical items corresponding to historical operation behaviors of the target client and candidate item information of each candidate item in an item pool; A second item detail data loading module, configured to determine a webpage address of a hypertext markup language webpage for displaying item detail data of the target item according to the inventory unit information, and load the item detail data according to the webpage address based on a webpage view component embedded in the target client; Wherein, the second item details data loading module includes: an item detail data first loading unit, configured to load the item detail data into a hardware memory of a target terminal integrated with the target client according to the web page address based on a web page view component embedded in the target client; An item details data determination unit, used to obtain a to-be-loaded address corresponding to a current operation behavior acting on the target client; The item details data reloading unit is used to load the data to be loaded from the hardware memory to the target client when there is data to be loaded corresponding to the address to be loaded in the item details data loaded into the hardware memory, so as to complete the loading of the item details data corresponding to the data to be loaded.

11. A target server, characterized in that: include: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the data loading method as described in any one of claims 1 to 4.

12. A target terminal, characterized in that: include: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the data loading method as described in any one of claims 5-8.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the data loading method according to any one of claims 1 to 8 is implemented.

Images