Data feature classification method and device, electronic equipment and storage medium

Abstract

The application provides a data feature classification method and device, electronic equipment and a storage medium, belonging to the technical field of artificial intelligence, through obtaining second initial label data of a second terminal, obtaining a data category label set according to first initial label data and the second initial label data, obtaining third initial label data and first user transaction features according to the data category label set, receiving fourth initial label data and second user transaction features fed back by the second terminal according to the data category label set and the second user transaction data, obtaining a label quantile point according to the first user transaction features, the third initial label data, the second user transaction features and the fourth initial label data, performing feature classification on the first user transaction data according to the label quantile point to obtain a first feature classification result, and receiving a second feature classification result obtained by the second terminal according to the label quantile point, so that the data quality of the user transaction data can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and particularly relates to a data feature classification method and device, an electronic device and a storage medium. BACKGROUND

[0002] Generally, each user transaction data exists in a unique storage location. In the related art, when selecting user transaction data, multiple data parties read user transaction data from respective storage locations and store the user transaction data in a unified shared location to realize data sharing. Since each data party can modify the shared user transaction data, if a data party randomly modifies the shared user transaction data, the shared user transaction data will be disordered and the quality of the user transaction data will be low. SUMMARY

[0003] The main purpose of the embodiments of the present application is to provide a data feature classification method and device, an electronic device and a storage medium, which aims to improve the data quality of user transaction data.

[0004] To achieve the above purpose, a first aspect of the embodiments of the present application provides a data feature classification method applied to a first terminal, wherein the first terminal includes first initial label data, and the method comprises the following steps:

[0005] obtaining second initial label data of a second terminal, wherein the second initial label data is obtained by sorting second user transaction data by the second terminal;

[0006] obtaining a data category label set according to the first initial label data and the second initial label data, and broadcasting the data category label set to the second terminal, wherein the first initial label data is obtained by sorting first user transaction data stored in advance by the first terminal, and the data category label set includes multiple data category labels;

[0007] screening the first initial label data according to the data category label set to obtain third initial label data, and performing query processing on the first user transaction data according to the third initial label data to obtain first user transaction features of the third initial label data, wherein the third initial label data is used to represent the sorting order of the first user transaction features in the first user transaction data;

[0008] receive fourth initial label data and second user transaction features of the fourth initial label data fed back by the second terminal according to the data category label set and the second user transaction data, wherein the fourth initial label data is obtained by the second terminal filtering the second initial label data according to the data category label set, the second user transaction features are obtained by the second terminal querying and processing the second user transaction data according to the fourth initial label data, and the fourth initial label data is used to represent the sorting order of the second user transaction features in the second user transaction data;

[0009] obtain label quantile points of each data category label according to the first user transaction features, the third initial label data, the second user transaction features and the fourth initial label data;

[0010] perform feature classification on the first user transaction data according to the label quantile points to obtain a first feature classification result, wherein the first feature classification result is used to represent the result of classifying the first user transaction features to the data category label;

[0011] broadcast the label quantile points to the second terminal, and receive a second feature classification result obtained by the second terminal performing feature classification according to the label quantile points, wherein the second feature classification result is used to represent the result of classifying the second user transaction features to the data category label.

[0012] In some embodiments, the obtaining of the data category label set according to the first initial label data and the second initial label data comprises:

[0013] comparing the first initial label data and the second initial label data;

[0014] if the first initial label data is greater than the second initial label data, taking the first initial label data as intermediate label data, and if the first initial label data is less than or equal to the second initial label data, taking the second initial label data as the intermediate label data;

[0015] performing label processing on a preset label according to the intermediate label data to obtain the data category label set.

[0016] In some embodiments, the performing of label processing on the preset label according to the intermediate label data to obtain the data category label set comprises:

[0017] obtaining a label quantity of the preset label;

[0018] According to the intermediate label data and the label quantity, a user transaction feature quantity of each preset label is calculated;

[0019] According to the user transaction feature quantity and the label quantity, a label processing is performed on a preset label to obtain a data category label set.

[0020] In some embodiments, the filtering of the first initial label data according to the data category label set to obtain third initial label data comprises:

[0021] If the first initial label data is the same as the data category label of the data category label set, the first initial label data is taken as the third initial label data.

[0022] In some embodiments, the filtering of the first initial label data according to the data category label set to obtain third initial label data comprises:

[0023] If the first initial label data is different from the data category label of the data category label set, the third initial label data is filtered from the first initial label data according to the data category label; wherein the third initial label data is less than the data category label.

[0024] In some embodiments, the obtaining of the label quantile point of each data category label according to the first user transaction feature, the third initial label data, the second user transaction feature and the fourth initial label data comprises:

[0025] According to the third initial label data and the fourth initial label data, target label data is determined;

[0026] According to the target label data, target user transaction features are filtered from the first user transaction feature and the second user transaction feature, and the target user transaction features are taken as the label quantile point.

[0027] In some embodiments, the determination of the target label data according to the third initial label data and the fourth initial label data comprises:

[0028] The third initial label data and the data category label are compared to obtain a first label;

[0029] The fourth initial label data and the data category label are compared to obtain a second label;

[0030] If the first label is less than the second label, the third initial label data is taken as the target label data, and if the first label is greater than or equal to the second label, the fourth initial label data is taken as the target label data.

[0031] To achieve the above object, a second aspect of the embodiment of the present application provides a data feature classification device, applied to a first terminal, the first terminal comprising first initial label data, the device comprising:

[0032] an obtaining module, configured to obtain second initial label data of a second terminal, wherein the second initial label data is obtained by the second terminal from sorting second user transaction data;

[0033] a first calculating module, configured to obtain a data category label set according to the first initial label data and the second initial label data, and broadcast the data category label set to the second terminal, wherein the first initial label data is obtained by the first terminal from sorting first user transaction data stored in advance, and the data category label set comprises a plurality of data category labels;

[0034] a screening module, configured to screen the first initial label data according to the data category label set, to obtain third initial label data, and to perform query processing on the first user transaction data according to the third initial label data, to obtain first user transaction features of the third initial label data, wherein the third initial label data is used to represent the sorting order of the first user transaction features in the first user transaction data;

[0035] a receiving module, configured to receive fourth initial label data and second user transaction features of the fourth initial label data, which are fed back by the second terminal according to the data category label set and the second user transaction data, wherein the fourth initial label data is obtained by the second terminal from screening the second initial label data according to the data category label set, the second user transaction features are obtained by the second terminal from query processing on the second user transaction data according to the fourth initial label data, and the fourth initial label data is used to represent the sorting order of the second user transaction features in the second user transaction data;

[0036] a second calculating module, configured to obtain a label quantile point of each data category label according to the first user transaction features, the third initial label data, the second user transaction features and the fourth initial label data;

[0037] The first feature classification module is configured to perform feature classification on the first user transaction data according to the label split point, and obtain a first feature classification result; wherein the first feature classification result is used to represent a result of classifying the first user transaction feature into the data category label.

[0038] The second feature classification module is configured to broadcast the label split point to the second terminal, and receive a second feature classification result obtained by the second terminal according to the label split point; wherein the second feature classification result is used to represent a result of classifying the second user transaction feature into the data category label.

[0039] To achieve the above object, a third aspect of embodiments of the present application provides an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the method of the first aspect when executing the computer program.

[0040] To achieve the above object, a fourth aspect of embodiments of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method of the first aspect.

[0041] The data feature classification method, data feature classification device, electronic device, and computer-readable storage medium proposed in this application, by acquiring second initial tag data from a second terminal, and by acquiring second initial tag data instead of second user transaction data, can ensure that the second user transaction data does not leave the local storage of the second terminal, prevent the first terminal from modifying the second user transaction data of the second terminal, and ensure the data security of the second user transaction data. A data category tag set is obtained based on the first and second initial tag data, and the data category tag set is broadcast to the second terminal. The first initial tag data is filtered according to the data category tag set to obtain third initial tag data. The first user transaction data is then queried and processed according to the third initial tag data to obtain the first user transaction features of the third initial tag data. Feedback from the second terminal based on the data category tag set and the second user transaction data is received. The system uses the fourth initial label data and the second user transaction features of the fourth initial label data. Based on the first user transaction features, the third initial label data, the second user transaction features, and the fourth initial label data, it obtains the label quantiles for each data category label. Based on the label quantiles, it performs feature classification on the first user transaction data to obtain the first feature classification result. The label quantiles are broadcast to the second terminal, and the system receives the second feature classification result obtained by the second terminal based on the label quantiles. By performing feature classification on the first data transaction data and the second user transaction data based on the label quantiles, it can ensure that the first user transaction data does not leave the local area of ​​the first terminal, and the second user transaction data does not leave the local area of ​​the second terminal. By classifying the user transaction data by feature, the user transaction features are classified into the corresponding feature categories, which can ensure the orderliness of the user transaction data and improve the data quality of the user transaction data. Attached Figure Description

[0042] Figure 1 This is a flowchart of the data feature classification method provided in the embodiments of this application;

[0043] Figure 2 yes Figure 1 The flowchart of step S120 in the middle;

[0044] Figure 3 yes Figure 2 The flowchart of step S230 in the middle;

[0045] Figure 4 yes Figure 1 The flowchart of step S130 in the process;

[0046] Figure 5 yes Figure 1 The flowchart of step S150 in the middle;

[0047] Figure 6 yes Figure 5 The flowchart of step S510 in the process;

[0048] Figure 7 is a structural schematic diagram of a data feature classification device provided by an embodiment of the present application;

[0049] Figure 8 is a hardware structure schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0051] It should be noted that although the functional modules are divided in the device schematic diagram, and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a manner different from the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification and claims and the above-described drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application, and are not intended to limit the present application.

[0053] First, the several terms involved in the present application are analyzed:

[0054] Artificial intelligence (AI): is a new technical science that studies, develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence; artificial intelligence is a branch of computer science, and artificial intelligence attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a similar way to human intelligence. The research in this field includes robots, language recognition, image recognition, natural language processing and expert systems, etc. Artificial intelligence can simulate the information process of human consciousness and thinking. Artificial intelligence is also the theory, method, technology and application system of using digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, to perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0055] Generally, each user transaction data exists in a unique storage location. In the related art, when selecting user transaction data, multiple data parties read the user transaction data from respective storage locations and store the user transaction data to a unified shared location to realize data sharing. Since each data party can modify the shared user transaction data, if a data party randomly modifies the shared user transaction data, the shared user transaction data will be disordered and the quality of the user transaction data will be low.

[0056] Based on this, the embodiment of the present application provides a data feature classification method, a data feature classification device, an electronic device and a computer readable storage medium, aiming to improve the data quality of user transaction data.

[0057] The data feature classification method and device, electronic device and storage medium provided by the embodiment of the present application are specifically explained by the following embodiment. First, the data feature classification method in the embodiment of the present application is described.

[0058] The embodiment of the present application can acquire and process related data based on artificial intelligence technology. Among them, artificial intelligence (Artificial Intelligence, AI) is to use digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, perceive environment, acquire knowledge and use knowledge to obtain the best results.

[0059] The basic technology of artificial intelligence generally includes technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. The software technology of artificial intelligence mainly includes computer vision technology, robot technology, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning, etc.

[0060] The data feature classification method provided by the embodiment of the present application relates to the field of artificial intelligence technology. The data feature classification method provided by the embodiment of the present application can be applied in a terminal, can also be applied in a server end, and can also be software running in a terminal or a server end. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc.; the server end can be configured as an independent physical server, can also be configured as a server cluster or a distributed system composed of multiple physical servers, can also be configured as a cloud server providing cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN and basic cloud computing services such as big data and artificial intelligence platform; the software can be an application that realizes the data feature classification method, etc., but is not limited to the above forms.

[0061] The application is operable in a variety of generic or specific computer system environments or configurations. For example: personal computer, server computer, handheld or portable device, tablet device, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like. The application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like, which perform particular tasks or implement particular abstract data types. The application can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules can be located in local and remote computer storage media including memory storage devices.

[0062] It should be noted that in various specific embodiments of the present application, when it is necessary to perform relevant processing according to user information, user behavior data, user history data, and user location information, and other data related to the identity or characteristics of the user, the user's permission or consent will be obtained first, and the collection, use and processing of such data will comply with relevant laws, regulations and standards. In addition, when the embodiments of the present application need to obtain sensitive personal information of the user, the separate permission or separate consent of the user will be obtained through a pop-up window or by jumping to a confirmation page, and after obtaining the separate permission or separate consent of the user, the necessary user-related data for enabling the embodiments of the present application to function normally will be obtained.

[0063] Figure 1 is an optional flowchart of the data feature classification method provided by the embodiments of the present application, applied to a first terminal, the first terminal including first initial label data, Figure 1 The method in can include but is not limited to steps S110 to S170.

[0064] Step S110, obtaining second initial label data of a second terminal; wherein the second initial label data is obtained by the second terminal from sorting second user transaction data;

[0065] Step S120, obtaining a data category label set according to the first initial label data and the second initial label data, and broadcasting the data category label set to the second terminal; wherein the first initial label data is obtained by the first terminal from sorting pre-stored first user transaction data, and the data category label set includes a plurality of data category labels.

[0066] Step S130, filtering the first initial label data according to the data category label set to obtain third initial label data, and querying the first user transaction data according to the third initial label data to obtain first user transaction features of the third initial label data; wherein the third initial label data is used to represent the sorting order of the first user transaction features in the first user transaction data;

[0067] Step S140, receiving fourth initial label data and second user transaction features of the fourth initial label data fed back by the second terminal according to the data category label set and the second user transaction data; wherein the fourth initial label data is obtained by the second terminal filtering the second initial label data according to the data category label set, and the second user transaction features are obtained by the second terminal querying the second user transaction data according to the fourth initial label data, and the fourth initial label data is used to represent the sorting order of the second user transaction features in the second user transaction data;

[0068] Step S150, obtaining label quantile points of each data category label according to the first user transaction features, the third initial label data, the second user transaction features and the fourth initial label data;

[0069] Step S160, classifying the first user transaction data according to the label quantile points to obtain a first feature classification result; wherein the first feature classification result is used to represent the result of classifying the first user transaction features to the data category label;

[0070] Step S170, broadcasting the label quantile points to the second terminal, and receiving a second feature classification result obtained by the second terminal classifying according to the label quantile points; wherein the second feature classification result is used to represent the result of classifying the second user transaction features to the data category label.

[0071] The steps S110 to S170 shown in the embodiments of the present application can ensure that the second user transaction data is not outside the second terminal, avoid the first terminal modifying the second user transaction data of the second terminal, ensure the data security of the second user transaction data, obtain the data category label set according to the first initial label data and the second initial label data, broadcast the data category label set to the second terminal, filter the first initial label data according to the data category label set to obtain third initial label data, perform query processing on the first user transaction data according to the third initial label data to obtain the first user transaction feature of the third initial label data, receive the fourth initial label data and the second user transaction feature of the fourth initial label data fed back by the second terminal according to the data category label set and the second user transaction data, obtain the label quantile point of each data category label according to the first user transaction feature, the third initial label data, the second user transaction feature and the fourth initial label data, perform feature classification on the first user transaction data according to the label quantile point to obtain the first feature classification result, broadcast the label quantile point to the second terminal, receive the second feature classification result obtained by the second terminal according to the label quantile point, and perform feature classification on the first data transaction data and the second user transaction data based on the label quantile point, which can ensure that the first user transaction data is not outside the first terminal and the second user transaction data is not outside the second terminal, perform feature classification on the user transaction data, classify the user transaction feature into the corresponding feature category, ensure the order of the user transaction data, and improve the data quality of the user transaction data.

[0072] In step S110 of some embodiments, one data party server is selected as the first terminal from a plurality of data party servers, and other data party servers are selected as the second terminal. The first terminal includes first user transaction data, and the second terminal includes second user transaction data. The first terminal executes a privacy sorting protocol on the first user transaction data to obtain first initial label data, and the second terminal executes the privacy sorting protocol on the second user transaction data to obtain second initial label data. The first user transaction data and the second user transaction data constitute a set of user transaction data. The user transaction data is transaction records of a plurality of users or the same user, including transaction objects of the user, transaction amounts for the transaction objects, transaction times, deposit amounts, withdrawal amounts, and other dimension information. Through the privacy sorting protocol, the first user transaction data is not outside the first terminal, and the second user transaction data is not outside the second terminal, so that the sorting result of the first user transaction data and the second user transaction data can be obtained, and the security of the clear data of each data party server is ensured.

[0073] The first terminal obtains second initial label data of the second terminal, wherein the second initial label data is obtained by performing a privacy sorting protocol on second user transaction data by the second terminal. A dimension information such as a transaction amount is selected as a data feature, and the second user transaction data of a plurality of users is sorted based on the data feature by performing a privacy sorting protocol to obtain the second initial label data, and the second initial label data is used to represent the sorting order of the data feature in the first user transaction data and the second user transaction data in the second user transaction data.

[0074] The method for privacy sorting is that there are a plurality of data server S1, S2, L, S n Each data server includes user transaction data, and the plurality of data servers select the same dimension information as a data feature to locally sort the user transaction data of each data server to obtain ordered user transaction data X1, X2, L, X n The ordered user transaction data X i is sorted by the data server S i Each data server S i generates a secret key key i , and distributes the secret key key i to S j and j≠i+1, that is, the secret key key i is distributed to each data server except the next data server S i of the current data server S i+1 , and the next data server of the last data server S n is the first data server S1; the data server S i encrypts the data feature x j in the user transaction data X i according to the secret key key i and j≠i-1 to obtain n-1 encrypted ciphertexts and sends the encrypted ciphertexts to the data server S j and j≠i-1; the data server S j sorts the received n-1 encrypted ciphertexts according to a preset sorting algorithm to obtain The data server S i obtains from the other n-1 data servers The first position data in and the second position data ind of the data feature x i in the user transaction data X i when locally sortingi obtaining the ranking sequence number of the data feature in all user transaction data, i.e.

[0075]

[0076] It should be noted that if the sort sorting algorithm is ascending sorting, i.e. from small to large, the result of the privacy sorting is from small to large, and if the sort sorting algorithm is descending sorting, i.e. from large to small, the result of the privacy sorting is from large to small. It can be understood that the sorting order of the local sorting should be consistent with the sorting order of the sort sorting algorithm.

[0077] Specifically, the first terminal performs the privacy sorting protocol to sort the first user transaction data in ascending order to obtain first initial label data, the second terminal performs the privacy sorting protocol to sort the second user transaction data in ascending order to obtain second initial label data, and the first terminal obtains the maximum value in the second initial label data from the second terminal. It should be noted that if the first user transaction feature in the first user transaction data is the same, the first user transaction feature is sorted in parallel, i.e. multiple same first user transaction features share the same ranking sequence number, and if the second user transaction feature in the second user transaction data is the same, the second user transaction feature is sorted in parallel, i.e. multiple same second user transaction features share the same ranking sequence number.

[0078] Please refer to Figure 2 In some embodiments, step S120 can include but is not limited to steps S210 to S230:

[0079] Step S210, comparing the first initial label data and the second initial label data;

[0080] Step S220, if the first initial label data is greater than the second initial label data, the first initial label data is taken as the intermediate label data, and if the first initial label data is less than or equal to the second initial label data, the second initial label data is taken as the intermediate label data;

[0081] Step S230, performing label processing on the preset label according to the intermediate label data to obtain a data category label set.

[0082] In step S210 of some embodiments, if the maximum value of the first initial label data is represented as M1 and the maximum value of the second initial label data is represented as M2, the first terminal compares the maximum value M1 of the first initial label data and the maximum value M2 of the second initial label data.

[0083] In step S220 of some embodiments, if the maximum value M1 of the first initial label data is greater than the maximum value M2 of the second initial label data, i.e., M1>M2, the maximum value M1 of the first initial label data is taken as the intermediate label data, and if the maximum value M1 of the first initial label data is less than or equal to the maximum value M2 of the second initial label data, i.e., M1≤M2, the maximum value M2 of the second initial label data is taken as the intermediate label data, and the intermediate label data is the maximum value in the first initial label data and the second initial label data.

[0084] In step S230 of some embodiments, the label quantity of the preset label is obtained, the feature quantity calculation is performed according to the intermediate label data and the label quantity, the user transaction feature quantity of each preset label is obtained, the preset label is processed according to the user transaction feature quantity and the label quantity, the data category label set is obtained, and the data category label set is broadcast to the second terminal.

[0085] The steps S210 to S230 described above determine the data category label set according to the first initial label data and the second initial label data, rather than directly determining the data category label set according to the first user transaction data and the second user transaction data, so that the first user transaction data cannot be known by the second terminal, and the second user transaction data cannot be known by the first terminal, which can ensure the security of the first user transaction data and the second user transaction data, and realize the feature classification of the user transaction data according to the data category label set without relying on a third-party trusted agency.

[0086] Please refer to Figure 3 In some embodiments, step S230 can include but is not limited to steps S310 to S330:

[0087] In step S310, the label quantity of the preset label is obtained.

[0088] In step S320, the feature quantity calculation is performed according to the intermediate label data and the label quantity, and the user transaction feature quantity of each preset label is obtained.

[0089] In step S330, the preset label is processed according to the user transaction feature quantity and the label quantity, and the data category label set is obtained.

[0090] In step S310 of some embodiments, the label quantity num of the preset label is obtained, and the initial value of the preset label can be obtained according to the label quantity, for example, the initial value of the i-th preset label is i. It can be understood that the initial value of the preset label can also be a random value or a null value.

[0091] In step S320 of some embodiments, the intermediate label data M is the maximum value of the first initial label data and the second initial label data, the intermediate label data M is divided by the label quantity num of the preset label to obtain the user transaction feature quantity t of each preset label.

[0092] In step S330 of some embodiments, the initial value of the preset label is updated according to the user transaction feature quantity t and the label quantity num, the updated preset label is taken as a data category label, and all data category labels constitute a data category label set. Specifically, an array I = {1, 2, …, num} is obtained according to the label quantity, t is multiplied by each first value in each array I to obtain a plurality of second values, i.e., the second values can be t, 2t, L, numt, the initial value of the preset label is updated with the second values, and the updated preset label is taken as a data category label. The data category label set is represented as L = (l0, L, l num ), wherein l i = ixt.

[0093] The steps S310 to S330 described above update the preset label by the user transaction feature quantity and the label quantity to obtain the data category label, the calculation method is simple, the data category label set can be quickly obtained, and the efficiency of classifying the user transaction data by features is improved.

[0094] Please refer to Figure 4 In some embodiments, step S130 can include but is not limited to including step S410 or step S420:

[0095] Step S410, if the first initial label data is the same as the data category label of the data category label set, the first initial label data is taken as the third initial label data.

[0096] Step S420, if the first initial label data is different from the data category label of the data category label set, the third initial label data is filtered from the first initial label data according to the data category label; wherein the third initial label data is less than the data category label.

[0097] In step S410 of some embodiments, if the same label data as the data category label in the data category label set can be found from the first initial label data, the label data is taken as the third initial label data.

[0098] In step S420 of some embodiments, if the same label data as the data category label in the data category label set is not found in the first initial label data, third initial label data is filtered from the first initial label data according to the data category label. Specifically, label data smaller than the data category label is found from the first initial label data, and label data with the smallest absolute value of the difference from the data category label is taken as the third initial label data. The first user transaction data is processed according to the third initial label data to obtain the first user transaction feature of the third initial label data, where the third initial label data and the first user transaction feature in the first user transaction data correspond one-to-one, and the first user transaction feature needs to match the data dimension information selected by the first terminal when performing the privacy sorting protocol. For example, if the data dimension information selected by the first terminal when performing the privacy sorting protocol is transaction amount, the first user transaction feature is the specific value of the transaction amount, and if the data dimension information is transaction frequency, the first user transaction feature is the specific value of the transaction frequency.

[0099] The above steps S410 to S420 compare the first initial label data with the data category label data to find label data matching the data category label from the first initial label data, and classify the data features of the first user transaction data based on the label data, which can improve the accuracy of feature classification.

[0100] In step S140 of some embodiments, the second terminal receives the data category label set broadcast by the first terminal, filters the second initial label data according to the data category label set to obtain fourth initial label data, and processes the second user transaction data according to the fourth initial label data to obtain the second user transaction feature of the fourth initial label data, where the fourth initial label data and the second user transaction feature correspond one-to-one, and the second user transaction feature needs to match the data dimension information selected by the second terminal when performing the privacy sorting protocol. For example, if the data dimension information selected by the second terminal when performing the privacy sorting protocol is transaction amount, the second user transaction feature is the specific value of the transaction amount, and if the data dimension information is transaction frequency, the second user transaction feature is the specific value of the transaction frequency. It should be noted that the data dimension information of the first user transaction feature and the second user transaction feature is the same, i.e., the first user transaction feature is transaction amount, and the second user transaction feature is also transaction amount.

[0101] Specifically, the second terminal receives the data category label set broadcasted by the first terminal, compares the data category label in the data category label set with the second initial label data, if the same label data as the data category label can be found in the second initial label data, the label data is taken as the fourth initial label data, if the same label data as the data category label cannot be found in the second initial label data, the fourth initial label data is screened from the second initial label data according to the data category label, and the fourth initial label data is smaller than the data category label. Specifically, if the same label data as the data category label cannot be found in the second initial label data, the label data smaller than the data category label is screened from the second initial label data, and the label data with the minimum absolute value of the difference value from the data category label is taken as the fourth initial label data.

[0102] The second initial label data is the sorting order of the second user transaction features in the second user transaction data, and each of the second user transaction features can be queried according to the second initial label data. Since the fourth initial label data is a subset of the second initial label data, the second user transaction data can be queried and processed according to the fourth initial label data to obtain the second user transaction features matched with the fourth initial label data from the second user transaction data.

[0103] Please refer to Figure 5 In some embodiments, the step S150 can include but is not limited to steps S510 to S520:

[0104] Step S510, determining the target label data according to the third initial label data and the fourth initial label data;

[0105] Step S520, screening the target user transaction feature from the first user transaction feature and the second user transaction feature according to the target label data, and taking the target user transaction feature as the label quantile point.

[0106] In step S510 of some embodiments, since the first terminal and the second terminal can obtain third initial label data and fourth initial label data respectively according to the same data category label, in order to screen out more optimal target label data from the third initial label data and the fourth initial label data, the third initial label data and the data category label are compared to obtain a first label, the fourth initial label data and the data category label are compared to obtain a second label, if the first label is less than the second label, the third initial label data is taken as the target label data, if the first label is greater than or equal to the second label, the fourth initial label data is taken as the target label data. If the data category label used by the first terminal to determine the third initial label data is different from the data category label used by the second terminal to determine the fourth initial label data, the third initial label data and the fourth initial label data are taken as the target label data.

[0107] In step S520 of some embodiments, the first terminal performs query processing according to the target label data, screens out target user transaction features from the first user transaction features and the second user transaction features, and takes the target user transaction features as label quantile points, wherein the label quantile points are used to classify the first user transaction data into corresponding data category labels.

[0108] The above steps S510 to S520 are used to ensure the accuracy of data feature classification, further screen the third initial label data and the fourth initial label data to determine the target label data from the third initial label data and the fourth initial label data, and determine the label quantile points according to the target label data, and perform data feature classification based on the label quantile points.

[0109] Please refer to Figure 6 In some embodiments, step S510 can include but is not limited to steps S610 to S630:

[0110] In step S610, the third initial label data and the data category label are compared to obtain a first label.

[0111] In step S620, the fourth initial label data and the data category label are compared to obtain a second label.

[0112] In step S630, if the first label is less than the second label, the third initial label data is taken as the target label data, if the first label is greater than or equal to the second label, the fourth initial label data is taken as the target label data.

[0113] In step S610 of some embodiments, if the data category label used by the first terminal to determine the third initial label data is the same as the data category label used by the second terminal to determine the fourth initial label data, in order to filter out better target label data from the third initial label data and the fourth initial label data, the third initial label data and the data category label used to determine the third initial label data are compared to obtain a first label, wherein the first label is the absolute value of the difference between the third initial label data and the data category label.

[0114] In step S620 of some embodiments, the fourth initial label data and the data category label used to determine the fourth initial label data are compared to obtain a second label, wherein the second label is the absolute value of the difference between the fourth initial label data and the data category label.

[0115] In step S630 of some embodiments, if the first label is less than the second label, the third initial label data is taken as the target label data, and if the first label is greater than or equal to the second label, the fourth initial label data is taken as the target label data.

[0116] The steps S610 to S630 described above, if the data category label used by the first terminal to determine the third initial label data is the same as the data category label used by the second terminal to determine the fourth initial label data, in order to determine the initial label data optimal for the data category label, the third initial label data and the fourth initial label data are filtered to obtain the target label data, so as to determine the label quantile point according to the target label data, and improve the accuracy of data feature classification.

[0117] In step S160 of some embodiments, the label quantile is a right boundary point of the classification, and is used to classify the first user transaction data into a data category label corresponding to the label quantile. The first terminal classifies the first user transaction data according to the label quantile, and classifies the first user transaction feature into a corresponding data category label to obtain a first feature classification result. If the label quantile of the first data category label is a first label quantile, and the label quantile of the second data category label is a second label quantile, when the first user transaction feature is greater than zero and less than or equal to the first label quantile, the first user transaction feature will be classified into the first data category label; when the first user transaction feature is greater than the first label quantile and less than or equal to the second label quantile, the first user transaction feature will be classified into the second data category label. For example, if the first user transaction feature in the first user transaction data is the transaction amount, the transaction amounts of multiple users are 120, 400 and 731 respectively, and the label quantiles are 200, 500 and 800 respectively, then 120 will be classified into the data category label l1 corresponding to 200, 400 will be classified into the data category label l2 corresponding to 500, and 731 will be classified into the data category label l3 corresponding to 800.

[0118] In step S170 of some embodiments, the label quantile is broadcast to the second terminal, and a second feature classification result obtained by the second terminal classifying a second user transaction feature according to the label quantile is received. The method of classifying data features by the second terminal is the same as that of the first terminal, which will not be described here.

[0119] Embodiments of the present application take the transaction amount as an example of user transaction features for data classification. The first terminal includes a first user transaction amount sequence, and the first user transaction amount sequence includes multiple first user transaction amounts. The second terminal includes a second user transaction amount sequence, and the second user transaction amount sequence includes multiple second user transaction amounts. The first user transaction amount sequence is composed of transaction amounts of multiple users or multiple transaction amounts of the same user, and the second user transaction amount sequence is also composed of transaction amounts of multiple users or multiple transaction amounts of the same user.

[0120] The first terminal executes a privacy sorting protocol to sort the first user's transaction amount sequence in ascending order, obtaining a first sorting number for each first user's transaction amount in the first and second user transaction amount sequences. The second terminal executes a privacy sorting protocol to sort the second user's transaction amount sequence in ascending order, obtaining a second sorting number for each second user's transaction amount in the first and second user transaction amount sequences. The first terminal selects the maximum value from the first sorting numbers of the first user's transaction amounts as the first tag data, and the second terminal selects the maximum value from the second sorting numbers of the second user's transaction amounts as the second tag data, and sends the second tag data to the first terminal. The first terminal selects the maximum value between the first and second tag data as intermediate tag data. The first terminal obtains the number of tags for a preset tag, divides the intermediate tag data by the number of tags to obtain the number of user transaction features for each preset tag, and multiplies the number of user transaction features sequentially by each positive integer from 1 to the number of tags to obtain the data category tag set L = (l1, l2, L, l...). num The data category label set is broadcast to the second terminal, wherein the data category label set includes multiple data category labels. k The number of data category labels is equal to the number of labels. The first terminal performs query processing on the data category labels in the data category label set. If the data category label l k If the first sort number is equal to the first sort number, then the first sort number is used as the third label data, and the first user transaction amount corresponding to the first sort number is used as the first user transaction feature. If no data category label is found in any of the first sort numbers... k Then it will be smaller than the data category label l k And with data category label l k The first sorted index with the smallest absolute difference is used as the third label data. kj1 And the transaction amount of the first user corresponding to the first sorting number is used as the transaction feature v of the first user. kj1 The second terminal performs query processing on the data category labels in the data category label set. If the data category label is l k If the data category label is equal to the second sort number, then the second sort number is used as the fourth label data, and the transaction amount of the second user corresponding to the second sort number is used as the second user transaction feature. If no data category label is found in any of the second sort numbers... k Then it will be smaller than the data category label l k And with data category label l k The second sorted index with the smallest absolute difference is used as the fourth label data. kj2 And the transaction amount of the second user corresponding to the second sorting number is used as the transaction feature v of the second user. kj2and the fourth label data and the second user transaction feature are sent to the first terminal. The first terminal determines a label quantile point according to the third label data, the first user transaction feature, the fourth label data and the second user transaction feature, and the third label data l kj1 and the data category label l k The absolute value of the difference is calculated to obtain a first difference, and the fourth label data l kj2 and the data category label l k The absolute value of the difference is calculated to obtain a second difference, and the minimum value of the first difference and the second difference, that is, min(I kji -I k ) is taken as a label quantile point v k , i is the number of terminals, i is 1 indicating the first terminal, i is 2 indicating the second terminal, the first terminal classifies the first user transaction amount into the corresponding data category label according to the label quantile point to obtain a first feature classification result. The label quantile point (v1, v2, L, v k , K, v num ) is broadcast to the second terminal, so that the second terminal classifies the second user transaction amount into the corresponding data category label according to the label quantile point to obtain a second feature classification result.

[0121] Please refer to Figure 7 The embodiment of the application also provides a data feature classification device applied to a first terminal, the first terminal comprising first initial label data, the data feature classification device can realize the data feature classification method described above, and the device comprises:

[0122] The acquisition module 710 is configured to acquire second initial label data of a second terminal, wherein the second initial label data is obtained by sorting second user transaction data by the second terminal;

[0123] The first calculation module 720 is configured to obtain a data category label set according to the first initial label data and the second initial label data, and broadcast the data category label set to the second terminal, wherein the first initial label data is obtained by sorting first user transaction data pre-stored by the first terminal, and the data category label set comprises a plurality of data category labels;

[0124] The screening module 730 is configured to screen the first initial label data according to the data category label set to obtain third initial label data, and perform query processing on the first user transaction data according to the third initial label data to obtain first user transaction features of the third initial label data, wherein the third initial label data is used to represent the sorting order of the first user transaction features in the first user transaction data;

[0125] The receiving module 740 is configured to receive fourth initial label data and second user transaction features of the fourth initial label data, which are fed back by the second terminal according to the data category label set and the second user transaction data. The fourth initial label data is obtained by the second terminal from the second initial label data according to the data category label set. The second user transaction features are obtained by the second terminal from the second user transaction data according to the fourth initial label data. The fourth initial label data is used to represent the sorting order of the second user transaction features in the second user transaction data.

[0126] The second calculating module 750 is configured to obtain a label quantile point of each data category label according to the first user transaction features, the third initial label data, the second user transaction features and the fourth initial label data.

[0127] The first feature classification module 760 is configured to perform feature classification on the first user transaction data according to the label quantile point to obtain a first feature classification result. The first feature classification result is used to represent the result of classifying the first user transaction features to the data category label.

[0128] The second feature classification module 770 is configured to broadcast the label quantile point to the second terminal, and receive a second feature classification result obtained by the second terminal according to the label quantile point. The second feature classification result is used to represent the result of classifying the second user transaction features to the data category label.

[0129] The specific implementation of the data feature classification apparatus is basically the same as the specific implementation of the data feature classification method, and will not be repeated here.

[0130] The embodiment of the present application further provides an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, the data feature classification method is realized. The electronic device can be any intelligent terminal including a tablet computer, a vehicle-mounted computer, etc.

[0131] Please refer to Figure 8 , Figure 8 The hardware structure of the electronic device of another embodiment is illustrated. The electronic device includes:

[0132] The processor 810 can be implemented in a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, etc. The processor 810 is configured to execute a related program to implement the technical solutions provided by the embodiments of the present application.

[0133] The memory 820 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), etc. The memory 820 can store an operating system and other application programs. When the technical solutions provided by the embodiments of the present specification are implemented by software or firmware, the related program codes are stored in the memory 820 and are called and executed by the processor 810 to implement the classification method of the data features of the embodiments of the present application;

[0134] The input / output interface 830 is configured to realize information input and output.

[0135] The communication interface 840 is configured to realize the communication interaction between the device and other devices. The communication can be realized by a wired manner (for example, a USB, a network cable, etc.) or a wireless manner (for example, a mobile network, a WI-FI, a Bluetooth, etc.).

[0136] The bus 850 is configured to transmit information between various components (for example, the processor 810, the memory 820, the input / output interface 830, and the communication interface 840) of the device.

[0137] The processor 810, the memory 820, the input / output interface 830, and the communication interface 840 are connected to each other through the bus 850 to realize the communication connection between the device.

[0138] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the above-mentioned classification method of the data features.

[0139] The memory is a non-transitory computer readable storage medium, which can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor. These remote memories can be connected to the processor through 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 a combination thereof.

[0140] The method for classifying data features, the device for classifying data features, the electronic device and the computer readable storage medium provided by the embodiments of the present application can ensure that the second user transaction data is not outside the second terminal, avoid the first terminal modifying the second user transaction data of the second terminal, ensure the data security of the second user transaction data, obtain a data category label set according to the first initial label data and the second initial label data, broadcast the data category label set to the second terminal, filter the first initial label data according to the data category label set to obtain third initial label data, query and process the first user transaction data according to the third initial label data to obtain first user transaction features of the third initial label data, receive fourth initial label data and second user transaction features of the fourth initial label data fed back by the second terminal according to the data category label set and the second user transaction data, obtain a label quantile point of each data category label according to the first user transaction features, the third initial label data, the second user transaction features and the fourth initial label data, classify the first user transaction data according to the label quantile point to obtain a first feature classification result, broadcast the label quantile point to the second terminal, receive a second feature classification result obtained by the second terminal according to the label quantile point, and classify the first user transaction data and the second user transaction data according to the label quantile point respectively, which can ensure that the first user transaction data is not outside the first terminal and the second user transaction data is not outside the second terminal, classify the user transaction features into corresponding feature categories by classifying the user transaction data, ensure the order of the user transaction data, and improve the data quality of the user transaction data.

[0141] The embodiments described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0142] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and can include more or fewer steps than the figures shown, or combine certain steps or different steps.

[0143] The device embodiments described above are only schematic, and the units described as separate components can or can not be physically separate, that is, can be located in one place or distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0144] Those skilled in the art can understand that all or some of the steps in the method disclosed above, the function modules / units in the system and the device can be implemented as software, firmware, hardware or appropriate combination thereof.

[0145] The terms "first", "second", "third", "fourth" and the like in the description of the application and in the claims, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the use of these terms herein is to be construed to cover a general order and / or structure unless otherwise indicated. Furthermore, the terms "comprise", "comprising", "has", "having", "includes", "including", "contain", "containing" or any other similar forms are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises, has, includes, contains items or components does not include items or components not explicitly recited. The terms "a" or "an", as used herein in the detailed description and in the claims, mean "one or more" or "at least one", unless otherwise indicated.

[0146] It should be understood that, in the present application, "at least one" means one or more, and "multiple" means two or more. "And / or" is used to describe the relationship between associated objects, which means that there can be three relationships, for example, "A and / or B" can mean that there are three cases: only A, only B, and A and B at the same time, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects. "At least one of the following" or similar expressions means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can mean a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be singular or plural.

[0147] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above-mentioned units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be omitted or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed objects can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0148] The units described as separate components above can or can not be physically separate, and the components shown as units can or can not be physical units, that is, can be located in one place, or can be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0149] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present alone, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0150] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including multiple instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method of each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program storage media.

[0151] The preferred embodiments of the embodiments of the present application are described above with reference to the accompanying drawings, and the scope of the rights of the embodiments of the present application is not limited thereto. Any modifications, equivalent replacements and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of the rights of the embodiments of the present application.

Claims

1. A data feature classification method, characterized in that, Applied to a first terminal, the first terminal including first initial tag data, the method includes: Obtain the second initial tag data of the second terminal; wherein, the second initial tag data is obtained by the second terminal sorting the second user transaction data; A data category tag set is obtained based on the first initial tag data and the second initial tag data, and the data category tag set is broadcast to the second terminal; wherein, the first initial tag data is obtained by the first terminal sorting the pre-stored first user transaction data, and the data category tag set includes multiple data category tags; The first initial label data is filtered according to the data category label set to obtain the third initial label data. The first user transaction data is then queried and processed according to the third initial label data to obtain the first user transaction feature of the third initial label data. The third initial label data is used to characterize the sorting order of the first user transaction feature in the first user transaction data. The system receives fourth initial label data and second user transaction features fed back by the second terminal based on the data category label set and the second user transaction data; wherein, the fourth initial label data is obtained by the second terminal filtering the second initial label data according to the data category label set, and the second user transaction features are obtained by the second terminal querying the second user transaction data based on the fourth initial label data, and the fourth initial label data is used to characterize the sorting order of the second user transaction features in the second user transaction data; The tag quantile of each data category tag is obtained based on the first user transaction characteristics, the third initial tag data, the second user transaction characteristics, and the fourth initial tag data; The first user transaction data is classified according to the label quantile to obtain a first feature classification result; wherein, the first feature classification result is used to characterize the result of the first user transaction features being classified into the data category label; The label quantile is broadcast to the second terminal, and the second terminal is received to perform feature classification based on the label quantile; wherein the second feature classification result is used to characterize the result of the second user transaction feature being classified into the data category label.

2. The data feature classification method according to claim 1, characterized in that, The step of obtaining the data category label set based on the first initial label data and the second initial label data includes: Compare the first initial label data and the second initial label data; If the first initial label data is greater than the second initial label data, then the first initial label data is used as the intermediate label data; if the first initial label data is less than or equal to the second initial label data, then the second initial label data is used as the intermediate label data. The preset tags are processed based on the intermediate tag data to obtain the data category tag set.

3. The data feature classification method according to claim 2, characterized in that, The step of performing tag processing on preset tags based on the intermediate tag data to obtain the data category tag set includes: Obtain the number of tags for the preset tags; The number of user transaction features for each preset tag is calculated based on the intermediate tag data and the number of tags. The preset tags are processed based on the number of user transaction features and the number of tags to obtain the data category tag set.

4. The data feature classification method according to claim 1, characterized in that, The step of filtering the first initial label data according to the data category label set to obtain the third initial label data includes: If the first initial label data is the same as the data category label of the data category label set, then the first initial label data is used as the third initial label data.

5. The data feature classification method according to claim 1, characterized in that, The step of filtering the first initial label data according to the data category label set to obtain the third initial label data includes: If the first initial label data is different from the data category label of the data category label set, then the third initial label data is selected from the first initial label data according to the data category label; wherein the third initial label data is smaller than the data category label.

6. The data feature classification method according to any one of claims 1 to 5, characterized in that, The step of obtaining the tag quantile for each data category tag based on the first user transaction characteristics, the third initial tag data, the second user transaction characteristics, and the fourth initial tag data includes: The target label data is determined based on the third initial label data and the fourth initial label data; Based on the target tag data, target user transaction features are selected from the first user transaction features and the second user transaction features, and the target user transaction features are used as the tag quantiles.

7. The data feature classification method according to claim 6, characterized in that, The step of determining the target label data based on the third initial label data and the fourth initial label data includes: The third initial label data and the data category label are compared to obtain the first label; The fourth initial label data and the data category label are compared to obtain the second label; If the first tag is less than the second tag, then the third initial tag data is used as the target tag data; if the first tag is greater than or equal to the second tag, then the fourth initial tag data is used as the target tag data.

8. A data feature classification device, characterized in that, Applied to a first terminal, the first terminal including first initial tag data, the device includes: The acquisition module is used to acquire the second initial tag data of the second terminal; wherein the second initial tag data is obtained by the second terminal sorting the second user transaction data; The first calculation module is configured to obtain a data category tag set based on the first initial tag data and the second initial tag data, and broadcast the data category tag set to the second terminal; wherein, the first initial tag data is obtained by the first terminal sorting pre-stored first user transaction data, and the data category tag set includes multiple data category tags; The filtering module is used to filter the first initial label data according to the data category label set to obtain third initial label data, and to query the first user transaction data according to the third initial label data to obtain the first user transaction features of the third initial label data; wherein, the third initial label data is used to characterize the sorting order of the first user transaction features in the first user transaction data. The receiving module is configured to receive fourth initial label data and second user transaction features fed back by the second terminal based on the data category label set and the second user transaction data; wherein, the fourth initial label data is obtained by the second terminal filtering the second initial label data according to the data category label set, and the second user transaction features are obtained by the second terminal querying the second user transaction data based on the fourth initial label data, and the fourth initial label data is used to characterize the sorting order of the second user transaction features in the second user transaction data; The second calculation module is used to obtain the tag quantile of each data category tag based on the first user transaction characteristics, the third initial tag data, the second user transaction characteristics, and the fourth initial tag data; The first feature classification module is used to perform feature classification on the first user transaction data according to the label quantile to obtain a first feature classification result; wherein, the first feature classification result is used to characterize the result of the first user transaction feature being classified into the data category label; The second feature classification module is used to broadcast the tag quantile to the second terminal and receive the second feature classification result obtained by the second terminal based on the tag quantile; wherein the second feature classification result is used to characterize the result of the second user transaction feature being classified into the data category tag.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the data feature classification method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the data feature classification method according to any one of claims 1 to 7.

Images