Data call processing method and its device, equipment, medium, and product

Abstract

The present application discloses a data call processing method and its device, equipment, medium, and product, the method comprising: receiving target data collected by multiple independent sites of an e-commerce platform via a first protocol layer, the target data being activity data corresponding to consumer users of the e-commerce platform; a data optimization module working at a second protocol layer pre-processing the target data according to the data format required by a third protocol layer and persisting it to a database; a data encapsulation module working at a third protocol layer performing data mining on the target data in the database to generate a portrait tag and its corresponding attribute data adapted to the consumer user of the e-commerce platform based on the target data; responding to a call instruction of a preset application service working at a fourth protocol layer, executing a call output of the portrait tag and its corresponding attribute data for the preset application service. The present application can realize centralized and unified data mining of massive user activity data of an e-commerce platform.

Description

Technical Field

[0001] The present application relates to the field of e-commerce information technology, and in particular to a data call processing method and its corresponding device, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] Cross-border e-commerce platforms are different from ordinary e-commerce platforms. They provide an independent site for each online store, and multiple independent sites are operated in a distributed manner. Therefore, cross-border e-commerce platforms generally need to use cluster servers to achieve centralized and unified management and control of the application services required by a large number of independent sites.

[0003] As we all know, the various user activity data generated by e-commerce platforms on a daily basis rely on the activities of consumer users of the e-commerce platforms in various online stores, including but not limited to user behavior data, data related to the operating performance of the user's terminal device, user registration data, user transaction data, user interaction data, etc., all belong to the category of user activity data. Such activity data types are diverse and scattered, but the massive amount of activity data itself contains rich data value. Effective data mining of these activity data and obtaining valuable processing information can contribute to the improvement and enhancement of service quality of e-commerce platforms.

[0004] In the prior art, the use of user activity data on cross-border e-commerce platforms is usually based on a specific business need. Different business needs use their own inherent business logic to utilize certain categories of activity data generated on the platform, which ultimately leads to an increase in the entropy of the data on the e-commerce platform, resulting in the generation of various types of junk data, reducing the operating efficiency and service experience of the e-commerce platform.

[0005] The applicant has been engaged in the research and development of technologies related to cross-border e-commerce platforms for a long time, and therefore attempts to make effective explorations into the problems existing in the existing technologies. Summary of the invention

[0006] The primary purpose of the present application is to solve at least one of the above problems and to provide a data call processing method and its corresponding device, computer equipment, computer-readable storage medium, and computer program product.

[0007] In order to meet the various objectives of this application, this application adopts the following technical solutions:

[0008] A data call processing method provided to meet one of the purposes of this application includes the following steps:

[0009] Receive target data collected by multiple independent sites of the e-commerce platform, where the target data is activity data corresponding to consumer users of the e-commerce platform, and is distributed by the cluster server of the e-commerce platform to the data collection plug-in running in each of the independent sites for collection, and the data collection plug-in works at the first protocol layer;

[0010] The data optimization module working in the second protocol layer performs data preprocessing on the target data according to the data format required by the third protocol layer and persists the data to the database for calling by the data encapsulation module of the third protocol layer;

[0011] The data encapsulation module working in the third protocol layer performs data mining on the target data in the database to generate a portrait label of a consumer user adapted to the e-commerce platform and its corresponding attribute data according to the target data;

[0012] In response to a call instruction from a preset application service operating at the fourth protocol layer, the preset application service executes a call output for the portrait tag and its corresponding attribute data, so that the preset application service determines the portrait tag based on the activity data of consumer users generated by independent sites in the e-commerce platform, thereby realizing the preset downstream task.

[0013] In an extended embodiment, before the step of receiving target data collected by multiple independent sites of the e-commerce platform, the following steps are included:

[0014] Respond to a data collection plug-in configuration request from any independent site in the e-commerce platform, distribute the corresponding data collection plug-in to the independent site, and configure it to run in the background of the independent site;

[0015] Register the independent site configured with the data collection plug-in into the configuration information library;

[0016] Obtain the target data access request sent by the independent site configured with the data acquisition plug-in, query whether the independent site exists in the configuration information library, and if the independent site exists, send a target data transmission token to the independent site, so as to receive the target data according to the data transmission token carried by the target data when the independent site transmits its target data.

[0017] In a further embodiment, a data encapsulation module working at the third protocol layer performs data mining on the target data in the database to generate a portrait label of a consumer user adapted to the e-commerce platform and its corresponding attribute data according to the target data, including the following steps:

[0018] The data modeling submodule in the data encapsulation module performs data abstraction on the target data, processes the target data into a standard format specification, and makes each target data include a plurality of attribute data;

[0019] The data portrait submodule in the data encapsulation module performs data aggregation on the target data to obtain portrait tags corresponding to the target data to form a portrait tag set;

[0020] A mapping relationship data set is constructed according to the correspondence between the portrait tags in the portrait tag set and the attribute data in the target data for invocation by a preset application service.

[0021] In an extended embodiment, in response to a call instruction of a preset application service operating at the fourth protocol layer, after executing the step of calling and outputting the portrait tag and its corresponding attribute data for the preset application service, the following steps are included:

[0022] The preset application service calls the portrait label and its attribute data to form a training sample, and trains the neural network classification model until it converges, wherein the portrait label is used as a supervisory label for the neural network classification model to classify and predict the attribute data;

[0023] The preset application service enables the neural network classification model that has been trained to a convergence state to make it suitable for reasoning about the processed data input therein to determine the corresponding portrait label;

[0024] The preset application service receives the activity data of consumer users submitted by any independent site, calls the data optimization module to perform data preprocessing and then performs word segmentation to obtain a word segmentation set, and constructs encoding information corresponding to the word segmentation set;

[0025] The preset application service calls the neural network classification model to perform representation learning and classification on the coded information, obtains the portrait label corresponding to the coded information and returns it to the independent site that submitted the activity data.

[0026] In an alternative embodiment, after the step of obtaining the portrait tag corresponding to the coded information and returning it to the independent site that submitted the activity data, the following steps are included:

[0027] The independent site searches for corresponding commodity items according to the portrait tags returned by the preset application service, and obtains a commodity list consisting of these commodity items;

[0028] The product list is output to a graphical user interface display of a consumer user who generates activity data based on determining the portrait tag.

[0029] In another alternative embodiment, after the step of obtaining the portrait tag corresponding to the coded information and returning it to the independent site that submitted the activity data, the following steps are included:

[0030] The independent site determines whether the portrait tag returned by the preset application service belongs to a preset violation tag. If it does belong to a violation tag, the access event of the consumer user who generates the activity data based on the portrait tag is blocked.

[0031] A data call processing device provided to meet one of the purposes of the present application includes: a data collection unit, a data optimization unit, a data profiling unit, and a data call unit, wherein the data collection unit is used to receive target data collected by multiple independent sites of an e-commerce platform, and the target data is activity data corresponding to consumer users of the e-commerce platform, and is distributed by a cluster server of the e-commerce platform to each of the independent sites. The data collection plug-in running in the independent sites is responsible for collection, and the data collection plug-in works at a first protocol layer; the data optimization unit is used to perform data preprocessing and processing on the target data according to the data format required by the third protocol layer by a data optimization module working at the second protocol layer The data is persisted to a database for calling by a data encapsulation module of the third protocol layer; the data portrait unit is used for the data encapsulation module working at the third protocol layer to perform data mining on the target data in the database, so as to generate a portrait label and its corresponding attribute data of the consumer user adapted to the e-commerce platform based on the target data; the data calling unit is used to respond to the calling instruction of the preset application service working at the fourth protocol layer, and execute the calling output of the portrait label and its corresponding attribute data for the preset application service, so that the preset application service determines its portrait label according to the activity data of the consumer user generated by the independent site in the e-commerce platform, and realizes the preset downstream task.

[0032] In an extended embodiment, the data call processing device of the present application also includes: a request response unit, which is used to respond to a data acquisition plug-in configuration request of any independent site in the e-commerce platform, distribute the corresponding data acquisition plug-in to the independent site, and configure it to run in the background of the independent site; a site registration unit, which is used to register the independent site configured with the data acquisition plug-in into the configuration information library; an authentication processing unit, which is used to obtain a target data access request sent by the independent site configured with the data acquisition plug-in, and query whether the independent site exists in the configuration information library. If the independent site exists, a target data transmission token is sent to the independent site, so that when the independent site transmits its target data, the target data can be received according to the data transmission token carried by the target data.

[0033] In a further embodiment, the data profiling unit comprises: a modeling subunit, configured to perform data abstraction on the target data by the data modeling submodule in the data encapsulation module, and process the target data into a standard format specification so that each target data includes multiple attribute data; an aggregation subunit, configured to perform data aggregation on the target data by the data profiling submodule in the data encapsulation module to obtain the portrait labels corresponding to the target data to form a portrait label set; a mapping subunit, configured to construct a mapping relationship data set according to the correspondence between the portrait labels in the portrait label set and the attribute data in the target data for the preset application service to call.

[0034] In an extended embodiment, the data call processing device of the present application further includes: a model training unit, configured to call the portrait label and its attribute data to form a training sample by the preset application service, train the neural network classification model, and train it to a convergence state, wherein the portrait label is used as a supervisory label for the neural network classification model to classify and predict the attribute data; a model activation unit, configured to activate the neural network classification model that has been trained to a convergence state by the preset application service so that it is suitable for reasoning the processed data input therein to determine the corresponding portrait label; a word segmentation encoding unit, configured to receive the activity data of consumer users submitted by any independent site by the preset application service, call the data optimization module to perform data preprocessing and then perform word segmentation to obtain a word segmentation set, and construct the encoding information corresponding to the word segmentation set; a label acquisition unit, configured to call the neural network classification model by the preset application service to represent and learn and classify the encoded information, obtain the portrait label corresponding to the encoded information and return it to the independent site that submitted the activity data.

[0035] In one embodiment of a flexible expansion, the data call processing device of the present application also includes: a list generation unit, configured to search for corresponding product items according to the portrait tag returned by the preset application service by the independent site, and obtain a product list consisting of these product items; a list push unit, configured to output the product list to a graphical user interface display of a consumer user who generates activity data based on the portrait tag.

[0036] In another embodiment of the flexible expansion, the data call processing device of the present application also includes: a security control unit, which is configured to determine by the independent site whether the portrait tag returned by the preset application service belongs to a preset violation tag, and if it does belong to a violation tag, block the access event of the consumer user who generates the activity data based on the portrait tag.

[0037] A computer device provided to meet one of the purposes of the present application includes a central processing unit and a memory, wherein the central processing unit is used to call and run a computer program stored in the memory to execute the steps of the data call processing method described in the present application.

[0038] A computer-readable storage medium is provided to meet another purpose of the present application, which stores a computer program implemented according to the data call processing method in the form of computer-readable instructions. When the computer program is called and executed by a computer, the steps included in the method are executed.

[0039] A computer program product provided to meet another purpose of the present application includes a computer program / instruction, which, when executed by a processor, implements the steps of the method described in any embodiment of the present application.

[0040] Compared with the prior art, the advantages of this application are as follows:

[0041] This application uses a data collection module, a data optimization module, and a data encapsulation module at different protocol layers to complete the full business chain data processing process of the activity data of consumer users generated by independent sites of the cross-border e-commerce platform, realize data mining, and allow the preset application service to call the portrait tags and corresponding attribute data generated by data mining through another protocol layer. It decouples each link in the data processing and application process through different protocol layers, and realizes centralized data control through the cluster server of the e-commerce platform, so that the user activity data generated by the e-commerce platform can serve various application services and realize corresponding downstream tasks. It provides an effective technical framework for the standardized control of user activity data of cross-border e-commerce platforms, avoids the e-commerce platform from generating too much data entropy in the daily operation process, enables the same data to be reused by multiple preset services, avoids the generation of too much data garbage, and ensures that the cluster servers and independent sites of the e-commerce platform have high operating efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0043] Figure 1 A schematic diagram of the network architecture of the application environment of the data call processing method of this application;

[0044] Figure 2 A flowchart of a typical embodiment of the data call processing method of the present application;

[0045] Figure 3 A flowchart of the process of configuring a data acquisition plug-in in an embodiment of the present application;

[0046] Figure 4 This is a flowchart of the working process of the data encapsulation module in the embodiment of the present application;

[0047] Figure 5 This is a flowchart of the preset application service working process in the embodiment of the present application;

[0048] Figure 6 This is a principle block diagram of the data call processing device of the present application;

[0049] Figure 7 A schematic diagram of the structure of a computer device used in this application. DETAILED DESCRIPTION

[0050] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be interpreted as limiting the present application.

[0051] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present application refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0052] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art to which this application belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with those in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as here.

[0053] It will be understood by those skilled in the art that the "client", "terminal" and "terminal device" used herein include both devices with wireless signal receivers, which are devices with only wireless signal receivers without transmission capabilities, and devices with receiving and transmitting hardware, which are devices with receiving and transmitting hardware capable of two-way communication on a two-way communication link. Such devices may include: cellular or other communication devices such as personal computers, tablet computers, which have single-line displays or multi-line displays or cellular or other communication devices without multi-line displays; PCS (Personal Communications Service, personal communication system), which can combine voice, data processing, fax and / or data communication capabilities; PDA (Personal Digital Assistant, personal digital assistant), which may include a radio frequency receiver, pager, Internet / intranet access, web browser, notepad, calendar and / or GPS (Global Positioning System, global positioning system) receiver; conventional laptop and / or palmtop computers or other devices, which have and / or include a conventional laptop and / or palmtop computer or other device with and / or including a radio frequency receiver. The "client", "terminal" and "terminal device" used herein may be portable, transportable, installed in a vehicle (air, sea and / or land), or suitable for and / or configured to run locally, and / or in a distributed form, at any other location on the earth and / or in space. The "client", "terminal" and "terminal device" used herein may also be a communication terminal, an Internet terminal, a music / video playing terminal, for example, a PDA, a MID (Mobile Internet Device) and / or a mobile phone with a music / video playing function, or a smart TV, a set-top box and other devices.

[0054] The hardware referred to by the names such as "server", "client", and "service node" in this application is essentially an electronic device with the equivalent capabilities of a personal computer. It is a hardware device with the necessary components revealed by the von Neumann principle, such as a central processing unit (including an arithmetic unit and a controller), a memory, an input device, and an output device. The computer program is stored in its memory, and the central processing unit calls the program stored in the external memory into the internal memory for execution, executes the instructions in the program, and interacts with the input and output devices to complete specific functions.

[0055] It should be pointed out that the concept of "server" referred to in this application can also be extended to the case of server clusters. According to the network deployment principle understood by those skilled in the art, the servers should be logically divided. In physical space, these servers can be independent of each other but can be called through interfaces, or integrated into a physical computer or a set of computer clusters. Those skilled in the art should understand this flexibility, and should not use it to restrict the implementation of the network deployment method of this application.

[0056] Unless expressly specified, one or more technical features of the present application can be deployed on a server for implementation and accessed by a client through a remote call to obtain an online service interface provided by the server, or can be directly deployed and run on a client for access.

[0057] The neural network models referenced or may be referenced in this application, unless expressly specified, can be deployed on a remote server and remotely called on the client, or can be deployed and directly called on a client with sufficient device capabilities. In some embodiments, when it runs on the client, its corresponding intelligence can be obtained through transfer learning to reduce the requirements for the client's hardware operating resources and avoid excessive occupation of the client's hardware operating resources.

[0058] Unless explicitly specified, the various data involved in this application can be stored remotely on a server or on a local terminal device, as long as it is suitable for being called by the technical solution of this application.

[0059] Those skilled in the art should be aware that, although the various methods of the present application are described based on the same concept and thus present commonality to each other, unless otherwise specified, these methods can be independently executed. Similarly, for each embodiment disclosed in the present application, they are all proposed based on the same inventive concept, therefore, concepts with the same expression, and concepts that are appropriately changed for convenience despite different expressions, should be understood as equivalent.

[0060] Unless the mutually exclusive relationship between the embodiments to be disclosed in this application is explicitly stated, the relevant technical features involved in each embodiment can be cross-combined to flexibly construct a new embodiment, as long as such combination does not deviate from the creative spirit of this application and can meet the needs of the prior art or solve certain deficiencies in the prior art. Those skilled in the art should be aware of this flexibility.

[0061] A data call processing method of the present application can be programmed as a computer program product and deployed in an e-commerce platform cluster server for execution.

[0062] See also Figure 1The network deployment architecture of the cross-border e-commerce platform shown in the figure includes a cluster server and a large number of independent sites running in a distributed manner. The cluster server provides registration and configuration services to each independent site, so that the two can communicate with each other, and the cluster server obtains the user traffic data generated by each independent site and makes corresponding controls on the registration, configuration, operation and other functions of each independent site. Each independent site is used to run an online store, so that consumer users of the e-commerce platform can access the online store and generate corresponding activity data. The activity data, including but not limited to user behavior data, data related to the operating performance of the terminal device where the user is located, user registration data, user transaction data, user interaction data, etc., can usually be collected and provided by independent sites. The activity data collected by the independent site can be submitted to the cluster server of the e-commerce platform as target data for further data mining processing.

[0063] See also Figure 2 The data call processing method of the present application, in its typical embodiment, comprises the following steps:

[0064] Step S1100: Receive target data collected by multiple independent sites of the e-commerce platform. The target data is activity data generated by consumers of the e-commerce platform. The data collection plug-in running in each of the independent sites is responsible for collection by the cluster server of the e-commerce platform. The data collection plug-in works at the first protocol layer:

[0065] In order to facilitate the cross-border e-commerce platform to obtain target data from the massive independent sites pre-registered on its platform, the cluster server distributes data collection plug-ins to the independent sites in advance. After receiving the data collection plug-in, the independent site implements background installation to start the corresponding process to process the activity data generated by consumer users active in the online store of the independent site, and submit the activity data as target data to the cluster server.

[0066] The data collection plug-in complies with the pre-specified protocol and works at the first protocol layer. It is specifically used to perform preliminary processing on the activity data triggered and submitted by consumer users in independent sites, convert it into corresponding target data according to the protocol specifications that can be flexibly preset by technical personnel in this field based on the disclosure herein, and then submit the target data to the cluster server.

[0067] The cluster server continuously receives, at the first protocol layer, target data submitted by all independent sites within the e-commerce platform that are configured with the data collection plug-in. These target data are mainly user activity data generated within the independent sites, but in some alternative embodiments, the data source can be further expanded. For example, it can be other various predicted types of target data crawled from other external sites outside the e-commerce platform to enrich the target data source and the potential value of the data.

[0068] Generally speaking, the activity data generated by consumer users usually includes the attribute field name of the data and the attribute data corresponding to the attribute field name. In real applications, since a certain data with the same attribute may have different attribute field names in different development scenarios, the massive amount of activity data is relatively messy, and the format of the activity data generated by different pages may also be inconsistent. Therefore, in a workaround, the data collection plug-in can first be responsible for the basic format specification of the activity data, such as organizing it into data pairs of attribute field names and their attribute data, so as to form the target data of the target preset specification, which is convenient for the second protocol layer to call.

[0069] Step S1200: The data optimization module working at the second protocol layer performs data preprocessing on the target data according to the data format required by the third protocol layer and persists the data to the database for calling by the data encapsulation module of the third protocol layer:

[0070] A data optimization module runs in the cluster server. The data optimization module works at the second protocol layer of the pre-standardized protocol described in this application. It can serve the data format specifications required by the third layer protocol to perform data preprocessing on the target data, including structured abstraction and unstructured abstraction of the target data, and then clean or calculate the target data after data abstraction, and finally persist the target data to a preset database so that the data encapsulation module working at the third layer protocol can directly call data from the database.

[0071] The target data is structured and abstracted. For example, similar items of target data with different attribute field names having the same attributes may be merged, thereby compressing the total amount of target data and abstracting more effective target data.

[0072] The target data is unstructuredly abstracted. For example, different attribute data with the same attribute field name are concatenated as strings. Similarly, the total amount of target data can be compressed to abstract more concentrated target data.

[0073] The target data after abstraction is cleaned mainly to delete some data in the target data that do not meet the preset targets required by the preset third-layer protocol, for example, to delete the data pairs corresponding to some attribute fields that have no data mining value.

[0074] Data calculation is performed on the abstracted target data, mainly to statistically analyze the attribute data corresponding to the numerical attribute fields in the target data to form new comprehensive attribute fields and their corresponding statistical values, and reconstruct them into data pairs between new attribute field names and attribute data.

[0075] When persisting the final target data, the data optimization module can simultaneously maintain connections with multiple databases of different types and tasks, and then store the corresponding target data persistently in the corresponding database according to the preset classification criteria. The database includes but is not limited to HDFS, MangoDB, Hbase, Redis, SQL Server, etc., which can be flexibly set by those skilled in the art, as long as the access function of the corresponding database is implemented in the data optimization module.

[0076] Step S1300: The data encapsulation module working in the third protocol layer performs data mining on the target data in the database to generate a portrait label of a consumer user adapted to the e-commerce platform and its corresponding attribute data according to the target data:

[0077] The cluster server also runs a data encapsulation module in its background. The data encapsulation module works at the third protocol layer of the preset protocol referred to in this application, and is used to perform data mining tasks on the target data in the database.

[0078] The data encapsulation module can be flexibly implemented by technicians in this field by using any well-known data mining algorithm, so that when it is running, it is suitable to use its corresponding data mining algorithm to obtain the original target data from the database, and then use these target data to generate corresponding portrait labels, and further use these portrait labels to classify these target data, and construct a knowledge graph between the portrait labels and the attribute data.

[0079] When the data encapsulation module performs data mining on the target data, it first adapts to a preset downstream task and determines the range of target data it needs to call. Then, it can perform data modeling on the target data within the range based on preset rules, mathematical models, preset algorithms, etc., to obtain standardized target data suitable for constructing portrait labels.

[0080] On the basis of the normalized target data, the preset data portrait algorithm is further used to construct portrait labels for the target data. The process of constructing portrait labels is also the process of classifying and aggregating the attribute fields in the normalized target data. Through classification, the target data is divided into multiple classification labels, namely portrait labels, and then these portrait labels are described by text. These portrait labels constitute a portrait label set, which constitutes a corresponding label system.

[0081] In order to facilitate the subsequent calling of the corresponding original target data stored in the database according to the portrait tags in the tag system, the portrait tags can also be used to further label the original target data persisted by the second protocol layer in the database, so as to establish a mapping relationship between the portrait tags and the attribute data in the original target data, so that each target data in the database is labeled with the corresponding portrait tags, thereby facilitating its calling through the portrait tags.

[0082] Step S1400, responding to the call instruction of the preset application service working at the fourth protocol layer, executing the call output of the portrait tag and its corresponding attribute data for the preset application service, so that the preset application service determines the portrait tag according to the activity data of the consumer user generated by the independent site in the e-commerce platform, and realizes the preset downstream task:

[0083] In a cross-border e-commerce platform, its cluster server can respond to calling instructions triggered by application services running in itself or its independent site, or even a third party by running a background service process working at the fourth protocol layer of the preset protocol referred to in this application, and call the portrait tag and its corresponding attribute data for it.

[0084] The application service is pre-registered with the cluster server and has a legal identity. Therefore, for the cluster server, the application service that triggers the call instruction is generally a preset application service, which is intended to implement corresponding preset downstream tasks, such as implementing the classification model training task of goods in the e-commerce platform, implementing the commodity search task, and implementing the security operation identification task. What kind of downstream task the application service specifically performs can be flexibly implemented by those skilled in the art in response to specific business scenarios, and does not affect the embodiment of the creativity of this application.

[0085] However, as an inherent mechanism for the present application to respond to the call instruction, after the cluster server receives the call instruction, it can respond to the call instruction and call the portrait tag and its corresponding attribute data from the database according to the data range specified by the call instruction, and return it to the corresponding application service for use. The data range can be determined by carrying information in the call instruction. The information includes specifying the portrait tag, specifying the time range corresponding to the target data, specifying other field information corresponding to the target data, etc. It can be flexibly determined by technical personnel in this field according to the needs of the actual downstream tasks, and similarly does not affect the creative embodiment of the present application.

[0086] When the preset application service obtains the portrait tag and its target data output by the cluster server of the present application, it can realize its own expected business function according to the business logic preset by the application service itself, so that it can subsequently quickly determine the portrait tag corresponding to the activity data based on the activity data of consumer users generated by independent sites in the e-commerce platform, so as to further search for similar content based on the portrait tag. For example, based on a portrait tag used to describe the classification of goods, other product information belonging to the portrait tag can be found, so as to achieve the purpose of product retrieval and realize the preset product search task.

[0087] Through the disclosure of this typical embodiment, it is not difficult to see that the present application uses a data collection module, a data optimization module, and a data encapsulation module at different protocol layers to complete the full business chain data processing process of the activity data of consumer users generated by independent sites of the cross-border e-commerce platform, realize data mining, and allow the preset application service to call the portrait tags and corresponding attribute data generated by data mining through another protocol layer. By decoupling each link in the data processing and application process through different protocol layers, and realizing centralized data control through the cluster server of the e-commerce platform, the user activity data generated by the e-commerce platform can serve various application services and realize corresponding downstream tasks, which provides an effective technical framework for the standardized control of user activity data of the cross-border e-commerce platform, avoids the e-commerce platform from generating too much data entropy increase in the daily operation process, enables the same data to be reused by multiple preset services, avoids the generation of too much data garbage, and ensures that the cluster servers and independent sites of the e-commerce platform have high operating efficiency.

[0088] See also Figure 3 In an extended embodiment, before the step S1100, the step of receiving target data collected by multiple independent sites of the e-commerce platform, the following steps are included:

[0089] Step S2100: respond to a data collection plug-in configuration request from any independent site in the e-commerce platform, distribute the corresponding data collection plug-in to the independent site, and configure it to run in the background of the independent site:

[0090] In this embodiment, the data collection plug-in working at the first protocol layer is distributed on demand by the cluster server of the present application. Therefore, when any independent site needs to install the data collection plug-in, the independent site sends a corresponding configuration request to the cluster server. The cluster service distributes the data collection plug-in to the independent site in response to the request. After the independent site downloads the data collection plug-in, it installs and configures it in the background of the online store to run it, so that it can start monitoring the user activity data generated by the online store according to the preset programming business logic, so as to start collecting the target data.

[0091] Step S2200: register the independent site configured with the data collection plug-in into the configuration information library:

[0092] When the independent site completes the installation of the data acquisition plug-in in its background, it returns a notification message to the cluster server. Based on this notification message, the cluster server can store the registration information of the independent site into its preset configuration information library, indicating that the independent site is configured with the data acquisition plug-in and has the legal identity to submit target data to the cluster server.

[0093] Step S2300: Obtain the target data access request sent by the independent site configured with the data acquisition plug-in, query the configuration information library to see whether the independent site exists, and if the independent site exists, send a target data transmission token to the independent site, so that when the independent site transmits its target data, the target data is received according to the data transmission token carried by the target data:

[0094] Before the data collection plug-in running on the independent site starts to send the target data it has collected to the cluster server, in order to establish a stable data transmission link with the cluster server, the data collection plug-in first sends a target data access request to the cluster server. The cluster server responds to the access request and queries the configuration information library to see whether the independent site that triggered the access request has been pre-registered in the configuration information library. If the registration information of the independent site exists in the configuration information library, a latest data transmission token can be generated for it, and the data transmission token can be sent to the independent site that initiated the access request for use. The data transmission token can use the public key in the asymmetric key. The independent site uses the public key to encrypt its target data to form encrypted target data. After the cluster server receives the encrypted target data, it uses the private key in the asymmetric key to decrypt and obtain the corresponding original text. It is not difficult to understand that the data transmission token serves as a role for the cluster server to verify the legality and validity of the target data submitted by the independent site. Subsequently, the independent site can verify whether the target data is suitable for its reception and further processing based on the data transmission token recommended by the target data, thereby avoiding the cluster server from being subjected to saturation attacks and enabling it to concentrate on processing legitimate target data.

[0095] This embodiment adopts unique pre-security measures for the distribution of data acquisition plug-ins and the transmission of target data, thereby improving the security of data transmission between cluster servers and independent sites in the e-commerce platform.

[0096] See also Figure 4 In a further embodiment, the step S1300, in which the data encapsulation module working at the third protocol layer performs data mining on the target data in the database, so as to generate a portrait label of a consumer user adapted to the e-commerce platform and its corresponding attribute data according to the target data, comprises the following steps:

[0097] Step S1310: The data modeling submodule in the data encapsulation module performs data abstraction on the target data, and processes the target data into a standard format specification so that each target data includes a plurality of attribute data:

[0098] In this embodiment, the data encapsulation module is implemented to include multiple sub-modules with relatively independent functions, wherein the first sub-module is a data modeling sub-module, which is used to perform data abstraction on the target data called from the database of the second protocol layer, so that the target data is processed into a standard format specification, so that the target data includes multiple attribute data according to the specification. This standard format specification can be flexibly set according to the needs of technical personnel in this field for data mining. It should be understood that generally a target data will contain multiple attribute fields for describing an event, and there will be multiple attribute data correspondingly.

[0099] As mentioned above, the specific method and algorithm of data abstraction can also be flexibly set by technicians in this field as needed, without affecting the creative embodiment of the present application.

[0100] Step S1320: The data portrait submodule in the data encapsulation module aggregates the target data to obtain portrait tags corresponding to the target data to form a portrait tag set:

[0101] Another submodule in the data encapsulation module is the data portrait submodule, which is mainly implemented using a data aggregation algorithm such as K-Means, and is used to perform data aggregation on each target data according to the attribute field name in the target data, thereby aggregating multiple target data sets, and determining the corresponding portrait label for each target data set. These portrait labels constitute a portrait label set, which is a label system. Other methods of data aggregation can also be applied to this embodiment, such as mean shift clustering algorithm, density-based clustering algorithm, maximum expectation (EM) clustering algorithm using Gaussian mixture model (GMM), agglomerative hierarchical clustering algorithm, graph group detection clustering algorithm, etc., and those skilled in the art can flexibly select them as needed.

[0102] Step S1330: construct a mapping relationship data set according to the correspondence between the portrait tags in the portrait tag set and the attribute data in the target data, so as to be called by the preset application service:

[0103] After determining the portrait tag set, the correspondence between each portrait tag and the attribute data in the target data has been determined. Based on this, the database can be further modified to further highlight the correspondence between the portrait tag and the attribute data in the target data. In a more convenient processing method, the corresponding portrait tag can be directly marked at the location of the attribute data of the target data in the database to establish a mapping relationship between the two, thereby realizing the construction of the mapping relationship data set in the database, which can then be directly called by the preset application service.

[0104] This embodiment first performs data modeling on the target data formed by the second protocol layer through the data modeling submodule to obtain normalized target data, and then further clusters the target data through the data profiling submodule to construct a label system, thereby realizing high-level semantic abstraction of the target data, thereby achieving effective mining of the data value of massive target data, making it suitable for opening to a variety of preset application service calls, and laying a solid foundation for the efficient use of user activity data on the e-commerce platform.

[0105] See also Figure 5In an extended embodiment, the step S1400, in response to a call instruction of a preset application service operating at the fourth protocol layer, and after executing the step of calling and outputting the portrait tag and its corresponding attribute data for the preset application service, includes the following steps:

[0106] Step S1500: The preset application service calls the portrait label and its attribute data to form a training sample, and trains the neural network classification model until it converges, wherein the portrait label is used as a supervisory label for the neural network classification model to classify and predict the attribute data:

[0107] In this embodiment, a preset application service is implemented to train a neural network classification model using the target data of this application, and then put it into use. Accordingly, the preset application service working in the fourth protocol layer calls the database through an interface, and the mapping relationship data set between the image label and the attribute data obtained after the third protocol layer completes the data mining is used as the training set required for the classification model training, and each mapping relationship data constitutes its training sample. In each training sample, the attribute data constitutes the input data of the classification model, and its corresponding image label constitutes its classification supervision label.

[0108] The neural network classification model includes a text feature extraction sub-model and a classifier. The text feature extraction sub-model can be used as a well-known basic model such as TextCNN, LSTM, Transformer, Bert, Albert, Electra, etc., and is used to represent and learn the encoding information of the input data provided by the training sample to obtain the corresponding text feature information. The classifier uses a fully connected layer to map the text feature information to a classification space preset by the classifier. The classification space contains multiple classification labels corresponding to each image label in the image label set. After being mapped to the classification space, the classification probability of each classification label corresponding to the input data is obtained, and the one with the largest classification probability is determined as the classification label corresponding to the input data. The portrait label corresponding to the training sample is used to correct the classification label obtained after the input data provided by the training sample is classified, and the model loss corresponding to the classification label is calculated to obtain the corresponding loss value. Then, when the loss value does not reach the preset threshold, the entire neural network classification model is gradient updated, and the weight parameters of the model are corrected through back propagation to make it close to the convergence state; when the loss value reaches the preset threshold, the classification model is deemed to have been trained to the convergence state, and the training task of the classification model can be terminated.

[0109] Step S1600: The preset application service activates the neural network classification model that has been trained to a convergence state so that it is suitable for reasoning the processed data input therein to determine the corresponding portrait label:

[0110] After the neural network classification model is trained to a convergence state, the preset application service can configure it in the cluster server of the e-commerce platform or the designated independent site and enable it. Based on this, the classification model learns the ability to classify the coded information of the attribute data input therein, and can serve the downstream tasks corresponding to the application service. In one embodiment, the preset application service is deployed in an independent site, so the neural network classification model can be trained and enabled in the independent site, and is dedicated to serving the activity data generated within the scope of the independent site, and determines the portrait labels corresponding to the activity data after processing the coded information based on the activity data. In another embodiment, the preset application service is a service run by the cluster server, so it trains the neural network classification model at the cluster server and enables it in the background of the cluster server, and serves any independent site of the entire e-commerce platform, and directly determines the corresponding portrait labels for the activity data of any independent site.

[0111] Step S1700: The preset application service receives the activity data of consumer users submitted by any independent site, calls the data optimization module to perform data preprocessing and then performs word segmentation to obtain a word segmentation set, and constructs the encoding information corresponding to the word segmentation set:

[0112] The preset application service is independently responsible for the activity data of consumer users submitted by any independent site. Similarly, it complies with the specific specifications of the second protocol layer of the protocol specification referred to in this application, calls the data optimization model to perform data preprocessing on the activity data to obtain target data in a corresponding format, and then directly performs word segmentation on the preprocessed target data to obtain a corresponding word segmentation set, and then converts the word segmentation set into corresponding encoding information suitable for processing by the neural network classification model according to a preset word segmentation dictionary.

[0113] Step S1800: The preset application service calls the neural network classification model to perform representation learning and classification on the coded information, obtains the portrait label corresponding to the coded information and returns it to the independent site that submitted the activity data:

[0114] As mentioned above, the neural network classification model will use its text feature extraction sub-model to perform representation learning on the encoded information of the previous step, extract the deep semantic information corresponding to the word set, that is, the text feature information, and then use the classifier to perform classification mapping on the text feature information to obtain its corresponding classification label, that is, determine the portrait label corresponding to the classification label, and this portrait label can be returned to the independent site that submitted the activity data for further processing. In other words, after implementing the preset application service, the preset application service has learned the ability to directly make portrait label judgments based on the input data with the help of the database referred to in this application. Therefore, when subsequently serving each independent site, it can directly classify the activity data of consumer users generated by the independent site without the participation of other protocol layers of this application, and quickly determine the portrait label corresponding to the activity data for use by downstream tasks.

[0115] In this embodiment, the data in the database library processed by this application is reused through the preset application service of the fourth protocol layer. After the preset application service learns the ability to directly determine the portrait label according to the underlying input data with the help of the neural network classification model, it is put into use. As a result, the preset application service can serve downstream tasks without the need to perform multi-layer protocol specifications on the subsequently generated activity data through the prior steps of this application. The corresponding portrait label can be quickly determined, which improves the data recognition capability and makes the implementation of downstream tasks such as product search and information security identification more efficient.

[0116] In an alternative embodiment, in step S1800, after the step of obtaining the portrait tag corresponding to the coded information and returning it to the independent site that submitted the activity data, the following steps are included:

[0117] Step S1911: The independent site searches for corresponding commodity items according to the image tags returned by the preset application service, and obtains a commodity list consisting of these commodity items:

[0118] The independent site is configured with a product information database for storing the products online, in which the product information corresponding to each product unit is pre-stored, and the product information is pre-labeled with an image tag in the label system based on the product information. Accordingly, the related product unit can be searched according to an image tag.

[0119] When the preset application service returns the portrait tag corresponding to the activity data of the consumer user generated by an independent site, for example, the consumer user clicks to visit a product in an online store, thereby submitting an activity data to the independent site. The activity data contains the description information corresponding to the product, including various product attributes, detail text, product title, etc. of the product. These description information are regarded as the composition of the activity data and submitted to the preset application service to determine the corresponding portrait tag. The independent site can then retrieve all product items carrying the portrait tag from its product information database according to the determined portrait tag, and then construct the retrieved product items into a product list.

[0120] Step S1912: Output the product list to a graphical user interface of a consumer user who generates activity data based on the portrait tag for display:

[0121] After the independent site obtains the product list, it can push it to the terminal device of the consumer user who generated the activity data. The terminal device will parse the product list and display it in its graphical user interface to complete information interaction with the consumer user and achieve the purpose of product recommendation, similar product matching, etc.

[0122] This embodiment demonstrates the practical value of the technical solution of the present application by showing a specific downstream task for performing product search implemented by cooperating with the preset application service and the business logic of the independent site. It is not difficult to understand that with the help of the technical framework of the present application, the portrait tags mined by the technical solution of the present application can be efficiently reused through the preset application service. For the e-commerce platform, it can greatly improve service efficiency and improve and enhance the service experience.

[0123] In another embodiment of the alternative expansion, in step S1800, after the step of obtaining the portrait tag corresponding to the coded information and returning it to the independent site that submitted the activity data, the following steps are included:

[0124] Step S1920: The independent site determines whether the portrait tag returned by the preset application service is a preset violation tag. If it is a violation tag, the access event of the consumer user who generates the activity data based on the portrait tag is blocked:

[0125] It is also crucial for cross-border e-commerce platforms to have security control over the activity data generated by their platforms. To serve this downstream task, the portrait tags determined by the preset application service are returned to the corresponding independent sites for further violation detection. Therefore, the portrait tags can be identification tags corresponding to information security, such as "pornography-related", "gambling-related", etc. After the preset application service identifies the portrait tags of the activity data submitted by the independent site, it returns them to the independent site. The independent site then identifies the portrait tags and detects whether they belong to the preset violation tags. If so, it can block the current access event of the consumer user who generated the activity data, thereby ensuring the information security of the platform.

[0126] This embodiment demonstrates the practical value of the technical solution of the present application by showing a specific downstream task for performing information security identification implemented by cooperating with the preset application service and the business logic of the independent site. Similar to the previous embodiment, with the help of the technical framework of the present application, the portrait tags mined by the technical solution of the present application can be efficiently reused through the preset application service. For e-commerce platforms, the intelligence level of information security detection can be provided to ensure information security.

[0127] See also Figure 6 , a data call processing device provided to meet one of the purposes of the present application is a functional embodiment of the data call processing method of the present application, the device includes: a data acquisition unit 1100, a data optimization unit 1200, a data profiling unit 1300, and a data call unit 1400, wherein the data acquisition unit 1100 is used to receive target data collected by multiple independent sites of the e-commerce platform, the target data is the activity data corresponding to the consumer users of the e-commerce platform, and is distributed by the cluster server of the e-commerce platform to the data acquisition plug-in running in each of the independent sites for collection, and the data acquisition plug-in works at the first protocol layer; the data optimization unit 1200 is used for the data optimization module working at the second protocol layer to collect the target data according to the requirements of the third protocol layer The target data is pre-processed in a data format and persisted to a database for calling by a data encapsulation module of the third protocol layer; the data portrait unit 1300 is used for the data encapsulation module working in the third protocol layer to perform data mining on the target data in the database, so as to generate a portrait label and its corresponding attribute data of the consumer user adapted to the e-commerce platform according to the target data; the data calling unit 1400 is used to respond to a calling instruction of a preset application service working in the fourth protocol layer, execute a calling output of the portrait label and its corresponding attribute data for the preset application service, so that the preset application service determines its portrait label according to the activity data of the consumer user generated by an independent site in the e-commerce platform, and realizes the preset downstream task.

[0128] In an extended embodiment, the data call processing device of the present application also includes: a request response unit, which is used to respond to a data acquisition plug-in configuration request of any independent site in the e-commerce platform, distribute the corresponding data acquisition plug-in to the independent site, and configure it to run in the background of the independent site; a site registration unit, which is used to register the independent site configured with the data acquisition plug-in into the configuration information library; an authentication processing unit, which is used to obtain a target data access request sent by the independent site configured with the data acquisition plug-in, and query whether the independent site exists in the configuration information library. If the independent site exists, a target data transmission token is sent to the independent site, so that when the independent site transmits its target data, the target data can be received according to the data transmission token carried by the target data.

[0129] In a further embodiment, the data portrait unit 1300 includes: a modeling subunit, configured to perform data abstraction on the target data by the data modeling submodule in the data encapsulation module, and process the target data into a standard format specification so that each target data includes multiple attribute data; an aggregation subunit, configured to perform data aggregation on the target data by the data portrait submodule in the data encapsulation module to obtain portrait tags corresponding to the target data to form a portrait tag set; a mapping subunit, configured to construct a mapping relationship data set according to the correspondence between the portrait tags in the portrait tag set and the attribute data in the target data for the preset application service to call.

[0130] In an extended embodiment, the data call processing device of the present application further includes: a model training unit, configured to call the portrait label and its attribute data to form a training sample by the preset application service, train the neural network classification model, and train it to a convergence state, wherein the portrait label is used as a supervisory label for the neural network classification model to classify and predict the attribute data; a model activation unit, configured to activate the neural network classification model that has been trained to a convergence state by the preset application service so that it is suitable for reasoning the processed data input therein to determine the corresponding portrait label; a word segmentation encoding unit, configured to receive the activity data of consumer users submitted by any independent site by the preset application service, call the data optimization module to perform data preprocessing and then perform word segmentation to obtain a word segmentation set, and construct the encoding information corresponding to the word segmentation set; a label acquisition unit, configured to call the neural network classification model by the preset application service to represent and learn and classify the encoded information, obtain the portrait label corresponding to the encoded information and return it to the independent site that submitted the activity data.

[0131] In one embodiment of a flexible expansion, the data call processing device of the present application also includes: a list generation unit, configured to search for corresponding product items according to the portrait tag returned by the preset application service by the independent site, and obtain a product list consisting of these product items; a list push unit, configured to output the product list to a graphical user interface display of a consumer user who generates activity data based on the portrait tag.

[0132] In another embodiment of the flexible expansion, the data call processing device of the present application also includes: a security control unit, which is configured to determine by the independent site whether the portrait tag returned by the preset application service belongs to a preset violation tag, and if it does belong to a violation tag, block the access event of the consumer user who generates the activity data based on the portrait tag.

[0133] In order to solve the above technical problems, the present application also provides a computer device. Figure 7 As shown, a schematic diagram of the internal structure of a computer device. The computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected via a system bus. Among them, the computer-readable storage medium of the computer device stores an operating system, a database, and computer-readable instructions. The database may store a control information sequence. When the computer-readable instructions are executed by the processor, the processor can implement a data call processing method. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The memory of the computer device may store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor can execute the data call processing method of the present application. The network interface of the computer device is used to connect and communicate with a terminal. Those skilled in the art can understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0134] In this embodiment, the processor is used to execute Figure 6 The memory stores the program codes and various data required to execute the above-mentioned units or sub-units. The network interface is used to transmit data between user terminals or servers. The memory in this embodiment stores the program codes and data required to execute all units / sub-units in the data call processing device of this application, and the server can call the program codes and data of the server to execute the functions of all sub-units.

[0135] The present application also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the data call processing method of any embodiment of the present application.

[0136] The present application also provides a computer program product, including a computer program / instruction, which implements the steps of the method described in any embodiment of the present application when executed by one or more processors.

[0137] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments of the present application can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a computer-readable storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0138] To sum up, the present application is suitable for use in cross-border e-commerce platforms to realize standardized data mining and utilization of user activity data generated by each independent site within the platform, and ultimately serve a variety of preset application services to achieve corresponding downstream tasks, avoid the generation of information spam, and ensure that the overall e-commerce platform maintains a high operating efficiency.

[0139] Those skilled in the art will appreciate that the various operations, methods, steps, measures, and schemes in the processes discussed in this application may be alternated, altered, combined, or deleted. Further, other steps, measures, and schemes in the various operations, methods, and processes discussed in this application may also be alternated, altered, rearranged, decomposed, combined, or deleted. Further, the steps, measures, and schemes in the prior art that are similar to those disclosed in this application may also be alternated, altered, rearranged, decomposed, combined, or deleted.

[0140] The above is only a partial implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A data call processing method, characterized in that: The steps include: Receive target data collected by multiple independent sites of the e-commerce platform, where the target data is activity data corresponding to consumer users of the e-commerce platform, and is distributed by the cluster server of the e-commerce platform to the data collection plug-in running in each of the independent sites for collection, and the data collection plug-in works at the first protocol layer; The data optimization module working in the second protocol layer performs data preprocessing on the target data according to the data format required by the third protocol layer and persists the data to the database for calling by the data encapsulation module of the third protocol layer; The data encapsulation module working at the third protocol layer performs data mining on the target data in the database to generate portrait labels of consumer users adapted to the e-commerce platform and their corresponding attribute data according to the target data, including: the data modeling submodule in the data encapsulation module performs data abstraction on the target data, processes the target data into a standard format specification, and makes each target data include multiple attribute data; the data portrait submodule in the data encapsulation module performs data aggregation on the target data to obtain portrait labels corresponding to the target data to form a portrait label set; a mapping relationship data set is constructed according to the correspondence between the portrait labels in the portrait label set and the attribute data in the target data for the preset application service to call; In response to a call instruction from a preset application service operating at the fourth protocol layer, the preset application service executes a call output for the portrait tag and its corresponding attribute data, so that the preset application service determines the portrait tag based on the activity data of consumer users generated by independent sites in the e-commerce platform, thereby realizing the preset downstream task.

2. The data call processing method according to claim 1, characterized in that: Before the step of receiving target data collected by multiple independent sites of the e-commerce platform, the following steps are included: Respond to a data collection plug-in configuration request from any independent site in the e-commerce platform, distribute the corresponding data collection plug-in to the independent site, and configure it to run in the background of the independent site; Register the independent site configured with the data collection plug-in into the configuration information library; Obtain the target data access request sent by the independent site configured with the data acquisition plug-in, query whether the independent site exists in the configuration information library, and if the independent site exists, send a target data transmission token to the independent site, so as to receive the target data according to the data transmission token carried by the target data when the independent site transmits its target data.

3. The data call processing method according to claim 1, characterized in that: In response to a call instruction of a preset application service operating at the fourth protocol layer, after executing the step of calling and outputting the portrait tag and its corresponding attribute data for the preset application service, the method includes the following steps: The preset application service calls the portrait label and its attribute data to form a training sample, and trains the neural network classification model until it converges, wherein the portrait label is used as a supervisory label for the neural network classification model to classify and predict the attribute data; The preset application service enables the neural network classification model that has been trained to a convergence state to make it suitable for reasoning about the processed data input therein to determine the corresponding portrait label; The preset application service receives the activity data of consumer users submitted by any independent site, calls the data optimization module to perform data preprocessing and then performs word segmentation to obtain a word segmentation set, and constructs encoding information corresponding to the word segmentation set; The preset application service calls the neural network classification model to perform representation learning and classification on the coded information, obtains the portrait label corresponding to the coded information and returns it to the independent site that submitted the activity data.

4. The data call processing method according to claim 3, characterized in that: After the step of obtaining the portrait tag corresponding to the coded information and returning it to the independent site that submitted the activity data, the following steps are included: The independent site searches for corresponding commodity items according to the portrait tags returned by the preset application service, and obtains a commodity list consisting of these commodity items; The product list is output to a graphical user interface display of a consumer user who generates activity data based on determining the portrait tag.

5. The data call processing method according to claim 3, characterized in that: After the step of obtaining the portrait tag corresponding to the coded information and returning it to the independent site that submitted the activity data, the following steps are included: The independent site determines whether the portrait tag returned by the preset application service belongs to a preset violation tag. If it does belong to a violation tag, the access event of the consumer user who generates the activity data based on the portrait tag is blocked.

6. A data call processing device, characterized in that: include: A data collection unit is used to receive target data collected by multiple independent sites of the e-commerce platform. The target data is activity data corresponding to consumer users of the e-commerce platform, and is distributed by the cluster server of the e-commerce platform to the data collection plug-in running in each of the independent sites for collection. The data collection plug-in works at the first protocol layer; A data optimization unit, configured to pre-process the target data according to the data format required by the third protocol layer by the data optimization module working at the second protocol layer and persist the data to the database for calling by the data encapsulation module at the third protocol layer; A data profiling unit is used to perform data mining on the target data in the database by a data encapsulation module working in the third protocol layer, so as to generate a portrait label and corresponding attribute data of a consumer user adapted to the e-commerce platform according to the target data, including: performing data abstraction on the target data by a data modeling submodule in the data encapsulation module, processing the target data into a standard format specification, so that each target data includes multiple attribute data; performing data aggregation on the target data by a data profiling submodule in the data encapsulation module, so as to obtain a portrait label corresponding to the target data, and form a portrait label set; constructing a mapping relationship data set according to the correspondence between the portrait label in the portrait label set and the attribute data in the target data, so as to provide a preset application service for calling; A data calling unit is used to respond to a calling instruction of a preset application service working at the fourth protocol layer, and execute a calling output of the portrait tag and its corresponding attribute data for the preset application service, so that the preset application service determines its portrait tag according to the activity data of consumer users generated by independent sites in the e-commerce platform, thereby realizing the preset downstream task.

7. A computer device comprising a central processing unit and a memory, characterized in that: The central processing unit is used to call and run the computer program stored in the memory to execute the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: It stores a computer program implemented according to the method described in any one of claims 1 to 5 in the form of computer-readable instructions, and when the computer program is called and executed by a computer, the steps included in the corresponding method are executed.

9. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method described in any one of claims 1 to 5 are implemented.

Images