A data display method, device, equipment and storage medium thereof

Abstract

The embodiment of the application belongs to the technical field of financial science and technology, is applied to the scene of classified and highlighted display of financial policy data according to different reviewers, and relates to a data display method, device, equipment and storage medium thereof, comprising training content optimization models for different categories of users through federated learning and FedAvg aggregation algorithm, so that when the same text data is displayed, different display focus data is extracted according to the identity information of different categories of users, and different categories of users can perform data personalized display on the corresponding browser page according to different focus data. It is convenient to highlight the display of focus data combined with the identity information of data viewers, and convenient for viewing and analysis. The output results corresponding to all the successfully trained content optimization sub-models are respectively encrypted by adopting differential privacy and homomorphic encryption technology, so that the privacy and security of data browsing are guaranteed when the browsing behavior data is sent to the server.

Description

Technical Field

[0001] The present application relates to the field of financial technology, and is applied to scenarios where financial insurance policy data is classified and highlighted according to different reviewers, and in particular to a data display method, device, equipment and storage medium thereof. Background Art

[0002] With the advancement of informatization in the financial industry and the advent of the big data era, financial data from financial companies is increasing. For example, in insurance or banking, after reaching an agreement with a financial client, it is often necessary to present the client with an electronic or printed copy of the contract.

[0003] Taking insurance institutions and insurance customers as an example, due to the different identities of the reviewers, the legal advisors in insurance institutions and insurance customers often focus on different data when reviewing and checking the same insurance contract. For example, legal advisors often focus on whether there are loopholes in the expression of insurance terms, or which insurance terms have ambiguous expressions in terms of rights and obligations. The starting point of their review is to avoid certain legal or insurance risks, while the data focus of insurance customers is generally whether the customer's own information is entered correctly. The starting point of their review is to check whether the entered personal information is correct. If the same display method is used for the same insurance contract regardless of the identity of the reviewer, there is a problem of not being able to highlight key data. If key data is set blindly, it will also cause the business to become complicated. Therefore, the existing technology still lacks a data display method, and there is a problem of not being able to display the same insurance text data in a personalized way to viewers with different identities. Summary of the Invention

[0004] The purpose of the embodiments of the present application is to propose a data display method, device, equipment and storage medium thereof to solve the problem that the existing technology still lacks a data display method and there is an inability to display the same insurance text data in a personalized manner to viewers of different identities.

[0005] In order to solve the above technical problems, the present application provides a data display method, which adopts the following technical solutions:

[0006] A data display method comprises the following steps:

[0007] Acquire all data in the document to be displayed and the objects to be displayed, wherein the objects to be displayed include the first user and the second user;

[0008] respectively setting the first user and the second user as target users;

[0009] Use browser scripts to collect the target user's browsing history data content of the corresponding browser page and the pause time when browsing each line of historical data content;

[0010] Construct the target user's corresponding browsing behavior representation data based on the target user's browsing history data content on the corresponding browser page and the pause time when browsing each line of historical data content;

[0011] Based on the browsing behavior representation data of target users, train the content optimization model corresponding to the target users;

[0012] Sending the full amount of data to the content optimization models corresponding to the first user and the second user respectively;

[0013] Obtaining first optimization data output from a content optimization model corresponding to the first user;

[0014] Obtaining second optimization data output from the content optimization model corresponding to the second user;

[0015] Displaying the first optimization data on a first browser page for viewing by the first user;

[0016] The second optimization data is displayed on a second browser page for viewing by the second user.

[0017] In order to solve the above technical problems, the embodiment of the present application also provides a data display device, which adopts the following technical solutions:

[0018] A data display device, comprising:

[0019] A document data acquisition module, configured to acquire the full amount of data in the document to be displayed and the objects to be displayed, wherein the objects to be displayed include the first user and the second user;

[0020] a target user setting module, configured to set the first user and the second user as target users respectively;

[0021] The browsing behavior data acquisition module is used to collect the historical data content of the target user's browsing of the corresponding browser page and the pause time when browsing each line of historical data content through the browser script;

[0022] A browsing behavior representation data construction module is used to construct the browsing behavior representation data corresponding to the target user based on the historical data content browsed by the target user on the corresponding browser page and the pause time when browsing each line of the historical data content;

[0023] A content optimization model training module is used to train a content optimization model corresponding to a target user based on the browsing behavior representation data corresponding to the target user;

[0024] a document data sending module, configured to send the full amount of data to the content optimization models corresponding to the first user and the second user respectively;

[0025] A first optimization data acquisition module, configured to acquire first optimization data output from the content optimization model corresponding to the first user;

[0026] A second optimization data acquisition module, configured to acquire second optimization data output from the content optimization model corresponding to the second user;

[0027] A first display module, configured to display the first optimization data on a first browser page for viewing by the first user;

[0028] The second display module is used to display the second optimization data on a second browser page for viewing by the second user.

[0029] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0030] A computer device includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the data display method described above when executing the computer-readable instructions.

[0031] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0032] A computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the data display method described above.

[0033] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0034] The data display method described in the embodiment of the present application obtains the full data in the document to be displayed; sends the full data to the content optimization models corresponding to the first user and the second user respectively; obtains the first optimization data output from the content optimization model corresponding to the first user; obtains the second optimization data output from the content optimization model corresponding to the second user; displays the first optimization data on a first browser page for the first user to view; and displays the second optimization data on a second browser page for the second user to view. By adopting the federated learning algorithm and the FedAvg aggregation algorithm, content optimization models are trained for different categories of users respectively, so that when displaying the same text data, different display focus data is extracted according to the identity information of different categories of users. For different categories of users, data can be personalized displayed on the corresponding browser page according to different focus data. It is convenient to highlight the focus data in combination with the identity information of the data viewer, which is convenient for viewing and analysis. By adopting differential privacy and homomorphic encryption technology to encrypt the output results corresponding to all successfully trained content optimization sub-models, the privacy and security of data browsing are guaranteed when sending browsing behavior data to the server. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the solutions in this application, a brief introduction will be given below to the drawings required for use in the description of the embodiments of this application. Obviously, the drawings described below are some embodiments of this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0036] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0037] Figure 2 is a flow chart of an embodiment of a data display method according to the present application;

[0038] Figure 3 yes Figure 2 A flowchart of a specific embodiment of step 205 is shown;

[0039] Figure 4 yes Figure 3 A flowchart of a specific embodiment of step 301 is shown;

[0040] Figure 5 yes Figure 3 A flowchart of a specific embodiment of step 302 is shown;

[0041] Figure 6 yes Figure 3 A flowchart of a specific embodiment of step 303 is shown;

[0042] Figure 7 yes Figure 3 A flowchart of a specific embodiment of step 307 is shown;

[0043] Figure 8 is a schematic structural diagram of an embodiment of a data display device according to the present application;

[0044] Figure 9 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art to which this application belongs. The terms used in the specification of the application are for the purpose of describing specific embodiments only and are not intended to limit this application. The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0046] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0047] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0048] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. Network 104 is a medium for providing communication links between terminal devices 101, 102, 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0049] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0050] Terminal devices 101, 102, and 103 can be various electronic devices with display screens and support web browsing, including but not limited to smartphones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV), laptop computers, desktop computers, etc.

[0051] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0052] It should be noted that the data display method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the data display device is generally set in the server / terminal device.

[0053] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0054] Continue to refer Figure 2 , shows a flow chart of an embodiment of a data display method according to the present application. The data display method comprises the following steps:

[0055] Step 201: Acquire all data in a document to be displayed and objects to be displayed, wherein the objects to be displayed include a first user and a second user.

[0056] In this embodiment, the full amount of data in the document to be displayed may be the data content in the same insurance contract, or the transaction information content in the transaction voucher of both parties, or a set of tax report data to be reviewed.

[0057] Taking the same insurance contract as an example, due to their different identities, the legal advisor and the insurance client will focus on different data in the same insurance contract. Similarly, taking the transaction information content in the transaction voucher as an example, due to the different responsibilities and obligations of the two parties, the focus data in the transaction voucher that the two parties focus on will also be different. Similarly, for a tax report data to be reviewed, the focus data of the tax report provider and the tax report reviewer will also be different.

[0058] Step 202: Set the first user and the second user as target users respectively.

[0059] Step 203: collecting, through a browser script, the historical data content browsed by the target user on the corresponding browser page and the pause time when browsing each line of the historical data content.

[0060] In this embodiment, the historical data content of the target user browsing the corresponding browser page and the pause time when browsing each line of historical data content are collected through the browser script. In addition, it can also include collecting the historical data content of the target user browsing the corresponding browser page and the annotation data when browsing each line of historical data content. The pause time and annotation data here are both for analyzing the user's browsing behavior.

[0061] Continuing with the insurance contract example, let's assume the first user is the insurance client's legal advisor, and the second user is the insurance client. The insurance client is the party signing the insurance contract. Because the insurance client and their legal advisor have different identities, when reviewing the same insurance contract, the legal advisor will often focus on the contract terms, while the insurance client will focus on personal information, the insured amount, and other details.

[0062] Assuming that the first user, that is, the legal advisor is the target user, the browser script is used to collect the historical data content of the legal advisor's browsing of the corresponding browser page and the pause time when browsing each line of the historical data content; based on the historical data content of the legal advisor's browsing of the corresponding browser page and the pause time when browsing each line of the historical data content, the browsing behavior representation data corresponding to the legal advisor is constructed; based on the browsing behavior representation data corresponding to the legal advisor, the content optimization model corresponding to the legal advisor is trained.

[0063] Assuming that the second user, that is, the insurance client, is the target user at this time, the historical data content of the insurance client's browsing of the corresponding browser page and the pause time when browsing each line of historical data content are collected through the browser script; based on the historical data content of the insurance client's browsing of the corresponding browser page and the pause time when browsing each line of historical data content, the browsing behavior representation data corresponding to the insurance client are constructed; based on the browsing behavior representation data corresponding to the insurance client, the content optimization model corresponding to the insurance client is trained.

[0064] Step 204 : construct browsing behavior representation data corresponding to the target user based on the historical data content browsed by the target user on the corresponding browser page and the pause time when browsing each line of the historical data content.

[0065] Step 205 : training a content optimization model corresponding to the target user based on the browsing behavior representation data corresponding to the target user.

[0066] Continue to refer Figure 3 , Figure 3 yes Figure 2 The flowchart of a specific embodiment of step 205 shown includes:

[0067] Step 301: Obtain browsing behavior characterization data corresponding to a target user, and input the browsing behavior characterization data into a browsing behavior analysis model built into a local client to analyze the browsing behavior of the target user and obtain a browsing behavior analysis result corresponding to the target user;

[0068] Continuing with the above insurance contract as an example, assuming that the first user is the insurance client's legal advisor, the second user is the insurance client, and the target client is the legal advisor or the insurance client, it is obvious that the corresponding browsing behavior analysis results will also be different. That is, for the same insurance contract, when viewing, the legal advisor and the insurance client will have different pause times when browsing each line of historical data.

[0069] Continue to refer Figure 4 , Figure 4 yes Figure 3 The flowchart of a specific embodiment of step 301 shown includes:

[0070] Step 401: Obtain browsing behavior representation data corresponding to the target user;

[0071] Step 402: Obtain, through the browsing behavior analysis model, the historical data content browsed by the target user on the corresponding browser page and the pause time when browsing each line of the historical data content;

[0072] Step 403: constructing a correlation map between each row of historical data content and its corresponding pause time based on the pause time when the target user browses each row of historical data content;

[0073] Step 404: Output the association comparison map as the browsing behavior analysis result through the browsing behavior analysis model.

[0074] Step 302: identifying the focus data browsed by the target user based on the browsing behavior analysis results of the target user;

[0075] Similarly, using the insurance contract as an example, when the target customer is a legal advisor or an insurance customer, the focus data they browse differs. Therefore, the corresponding association maps for legal advisors and insurance customers are different.

[0076] Continue to refer Figure 5 , Figure 5 yes Figure 3 The flowchart of a specific embodiment of step 302 shown includes:

[0077] Step 501: Identify the pause time of the target user when browsing each row of historical data content through the association comparison map;

[0078] Step 502: Compare the pause time when the target user browses each row of historical data content with a preset pause time threshold, and obtain a magnitude relationship between the pause time when the target user browses each row of historical data content and the preset pause time threshold;

[0079] Step 503: If the target user's pause time when browsing the current row of historical data content is greater than the preset pause time threshold, the current row of historical data content is the focus data;

[0080] Step 504: If the target user's pause time when browsing the current row of historical data content is not greater than the preset pause time threshold, the current row of historical data content is non-focus data.

[0081] Step 303: training a content optimization sub-model corresponding to the target user on the local client based on the focus data browsed by the target user;

[0082] Continue to refer Figure 6 , Figure 6 yes Figure 3 The flowchart of a specific embodiment of step 303 shown includes:

[0083] Step 601: Obtain all focus data browsed by the target user;

[0084] Step 602: Obtain location information of different focus data in the corresponding browser page through DOM parsing;

[0085] Step 603: constructing training data pairs based on all focus data and location information of different focus data in corresponding browser pages;

[0086] Step 604: Input the training data pair into a preset content optimization sub-model. Using the content optimization sub-model and a preset word frequency algorithm, analyze each browsing keyword contained in all focus data and the location information of each browsing keyword in the browser page, and output each browsing keyword and the location information of each browsing keyword in the browser page as an output result of the content optimization sub-model. The preset content optimization sub-model is a deep learning model based on a CNN neural network architecture.

[0087] In this embodiment, the preset word frequency algorithm refers to treating all the focus data as a data text domain, identifying each word contained in all the focus data, calculating the word frequency of each word, and screening out each browsing keyword according to the word frequency size corresponding to each word. Specifically, the word frequency algorithm can be a TF-IDF algorithm.

[0088] Step 605: If the output result of the content optimization sub-model is a null value, the training of the content optimization sub-model fails;

[0089] Step 606: If the output result of the content optimization sub-model is not a null value, the content optimization sub-model is successfully trained.

[0090] Since a deep learning model based on the CNN neural network architecture is used to train the content optimization sub-model corresponding to the target user, in an ideal state, the model training is successful and the target user has a trained content optimization sub-model. However, there are also cases where the model training fails under non-ideal conditions. Since the federated learning algorithm is used for comprehensive model training, it is not required that every user has a successfully trained content optimization sub-model on the client. It only requires that multiple users can train successful content optimization sub-models, and then integrate the models on the server to obtain the target content optimization model.

[0091] Step 304: Identify the identity information of the target user according to a preset identity information table, wherein the identity information table includes the identity information of the first user and the identity information of the second user;

[0092] Step 305: Using the target user's identity information as a search field, select N users with the same identity information as the target user from all users to construct a similar user set, where N is a positive integer.

[0093] Specifically, assuming that the target user is a legal advisor, execute step 305, i.e., screen out N legal advisors from all users; assuming that the target user is an insurance client, execute step 305, i.e., screen out N insurance clients from all users.

[0094] Step 306: Select different users from the similar user set as target users in sequence, and repeat steps 301 to 303 to train content optimization sub-models corresponding to all users in the similar user set on the local client.

[0095] Specifically, assuming that the similar user set is a legal advisor set, execute step 306, that is, train content optimization sub-models corresponding to different legal advisors in the local clients of different legal advisors; similarly, assuming that the similar user set is an insurance customer set, execute step 306, that is, train content optimization sub-models corresponding to different insurance customers in the local clients of different insurance customers.

[0096] Step 307: Obtain all successfully trained content optimization sub-models, and use a federated learning algorithm to comprehensively optimize all successfully trained content optimization sub-models to obtain a target content optimization model.

[0097] Continue to refer Figure 7 , Figure 7 yes Figure 3 The flowchart of a specific embodiment of step 307 shown includes:

[0098] Step 701: Obtain the output results corresponding to all successfully trained content optimization sub-models;

[0099] Step 702: Send the output results corresponding to all successfully trained content optimization sub-models from different clients to a unified target server;

[0100] In this embodiment, before executing the step of sending the output results corresponding to all the successfully trained content optimization sub-models from different clients to a unified target server, the method also includes: using differential privacy and homomorphic encryption technology to encrypt the output results corresponding to all the successfully trained content optimization sub-models respectively, and obtain the encrypted result corresponding to each output result, wherein, when encrypting the output results corresponding to different content optimization sub-models, the differential privacy algorithm used is consistent, and the homomorphic encryption algorithm used is also consistent.

[0101] In this embodiment, the step of sending the output results corresponding to all successfully trained content optimization sub-models from different clients to a unified target server specifically includes: using the encrypted result corresponding to each output result as transmission data; and sending the transmission data corresponding to different clients to a unified target server.

[0102] By employing differential privacy and homomorphic encryption to encrypt the output results of all successfully trained content optimization sub-models, we ensure the privacy and security of browsing behavior data when it is sent to the server. Differential privacy ensures that data is not leaked while being used for research or analysis, while homomorphic encryption ensures that data comparisons and calculations can still be performed after encryption.

[0103] Step 703: On the target server, a preset aggregation strategy in the federated learning algorithm is used to aggregate the output results corresponding to all successfully trained content optimization sub-models to obtain a target output result, wherein the preset aggregation strategy is the FedAvg aggregation algorithm.

[0104] The FedAvg aggregation algorithm randomly selects the output results of several clients, calculates the output results of the several clients to obtain a global output result, and replaces the sub-models corresponding to all clients with the current global model.

[0105] Step 704: parse the target output result to obtain each browsing keyword contained in the target output result and location information of each browsing keyword in the browser page;

[0106] Step 705: Use each browsing keyword included in the target output result and the location information of each browsing keyword in the browser page as model configuration parameter information to construct a target content optimization model.

[0107] By adopting the federated learning algorithm to comprehensively optimize all the successfully trained content optimization sub-models, the target content optimization model is obtained, that is, the comprehensive content optimization model corresponding to the same type of users is obtained. Regardless of whether the content optimization sub-models are trained for the same type of users screened in the early stage, the model configuration parameter information of the target content optimization model is uniformly issued, and the content optimization models of all the same type of users on the client are updated, ensuring that the same type of users have the content optimization models with the same configuration. In addition, the common browsing behaviors of the same type of users are combined to make the model integration more scientific.

[0108] Step 308: Obtain the model configuration parameter information of the target content optimization model, send the model configuration parameter information to the clients of all users with the same identity information as the target user through the service console, and set the target content optimization model on the clients of all users with the same identity information as the target user according to the model configuration parameter information.

[0109] Step 206: Send the full amount of data to the content optimization models corresponding to the first user and the second user respectively.

[0110] Step 207: Obtain first optimization data output from the content optimization model corresponding to the first user.

[0111] In this embodiment, the content optimization model corresponding to the first user can perform prompt color, scaling, bolding and tilting processing on the focus data corresponding to the first user in the content optimization model. The main purpose is to optimize the focus data so that the first user can easily identify the focus data, which makes it easier for the first user to view and analyze the focus data. The focus data after prompt color, scaling, bolding and tilting processing is the first optimized data.

[0112] Step 208: Acquire second optimization data output from the content optimization model corresponding to the second user.

[0113] Similarly, the content optimization model corresponding to the second user can perform prompt color, scaling, bolding and tilting processing on the focus data corresponding to the second user in the content optimization model. The main purpose is to optimize the focus data so that the second user can easily identify the focus data, which makes it easier for the second user to view and analyze the focus data. The focus data after prompt color, scaling, bolding and tilting processing is the second optimized data.

[0114] Step 209: Display the first optimization data on a first browser page for viewing by the first user.

[0115] Step 210: Display the second optimization data on a second browser page for viewing by the second user.

[0116] By employing the Federated Learning algorithm and the FedAvg aggregation algorithm, content optimization models are trained for different user categories. This allows us to extract different focus data for display when displaying the same text data, based on the user's identity information. This allows for personalized data display on the corresponding browser page for each user category. This facilitates highlighting of focus data based on the user's identity information, making it easier to view and analyze.

[0117] This application obtains the full amount of data in the document to be displayed; sends the full amount of data to the content optimization models corresponding to the first user and the second user respectively; obtains the first optimization data output from the content optimization model corresponding to the first user; obtains the second optimization data output from the content optimization model corresponding to the second user; displays the first optimization data on the first browser page for the first user to view; and displays the second optimization data on the second browser page for the second user to view. By adopting the federated learning algorithm and the FedAvg aggregation algorithm, the content optimization models are trained for different categories of users respectively, so that when displaying the same text data, different display focus data are extracted according to the identity information of different categories of users. For different categories of users, data can be personalized displayed on the corresponding browser page according to different focus data. It is convenient to highlight the focus data in combination with the identity information of the data viewer, which is convenient for viewing and analysis. By adopting differential privacy and homomorphic encryption technology to encrypt the output results corresponding to all the successfully trained content optimization sub-models, the privacy and security of data browsing are guaranteed when sending browsing behavior data to the server.

[0118] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal results.

[0119] Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0120] In the embodiment of the present application, by adopting the federated learning algorithm and the FedAvg aggregation algorithm, content optimization models are trained for different categories of users respectively, so that when displaying the same text data, different display focus data are extracted according to the identity information of different categories of users, and for different categories of users, data can be personalized displayed on the corresponding browser page according to different focus data. It is convenient to highlight the focus data in combination with the identity information of the data viewer, which is convenient for viewing and analysis. By adopting differential privacy and homomorphic encryption technology to encrypt the output results corresponding to all the successfully trained content optimization sub-models, the privacy and security of data browsing are guaranteed when sending browsing behavior data to the server.

[0121] Further references Figure 8 , as a response to the above Figure 2 In order to realize the method shown in FIG, the present application provides an embodiment of a data display device. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0122] like Figure 8 As shown, the data display device 800 of this embodiment includes: a document data acquisition module 801, a target user setting module 802, a browsing behavior data acquisition module 803, a browsing behavior representation data construction module 804, a content optimization model training module 805, a document data sending module 806, a first optimization data acquisition module 807, a second optimization data acquisition module 808, a first display module 809, and a second display module 810. Among them:

[0123] The document data acquisition module 801 is used to acquire the full amount of data in the document to be displayed and the objects to be displayed, wherein the objects to be displayed include the first user and the second user;

[0124] A target user setting module 802 is configured to set the first user and the second user as target users respectively;

[0125] Browsing behavior data acquisition module 803, used to collect the target user's browsing history data content of the corresponding browser page and the pause time when browsing each line of the historical data content through the browser script;

[0126] Browsing behavior representation data construction module 804, configured to construct browsing behavior representation data corresponding to the target user based on the historical data content of the target user browsing the corresponding browser page and the pause time when browsing each line of the historical data content;

[0127] The content optimization model training module 805 is used to train the content optimization model corresponding to the target user based on the browsing behavior representation data corresponding to the target user;

[0128] A document data sending module 806 is configured to send the full amount of data to the content optimization models corresponding to the first user and the second user respectively;

[0129] A first optimization data acquisition module 807 is configured to acquire first optimization data output from the content optimization model corresponding to the first user;

[0130] A second optimization data acquisition module 808 is configured to acquire second optimization data output from the content optimization model corresponding to the second user;

[0131] A first display module 809 is configured to display the first optimization data on a first browser page for viewing by the first user;

[0132] The second display module 810 is used to display the second optimization data on a second browser page for viewing by the second user.

[0133] In some specific embodiments of the present application, the data display device 800 also includes a content optimization sub-model training module, which is used to obtain all focus data browsed by the target user, and is also used to obtain the position information of different focus data in the corresponding browser page through DOM parsing. It is also used to construct training data pairs based on all focus data and the position information of different focus data in the corresponding browser page, and is also used to input the training data pairs into a preset content optimization sub-model, and analyze the various browsing keywords contained in all focus data and the position information of each browsing keyword in the browser page through the content optimization sub-model and a preset word frequency algorithm, and output the various browsing keywords and the position information of each browsing keyword in the browser page as the output result of the content optimization sub-model, wherein the preset content optimization sub-model is a deep learning model based on the CNN neural network architecture, and is also used to indicate that if the output result of the content optimization sub-model is a null value, then the content optimization sub-model training fails, and if the output result of the content optimization sub-model is not a null value, then the content optimization sub-model training is successful.

[0134] In some specific embodiments of the present application, the data display device 800 also includes a content optimization sub-model integration module, which is used to obtain all successfully trained content optimization sub-models and use a federated learning algorithm to comprehensively optimize all successfully trained content optimization sub-models to obtain a target content optimization model. Specifically, the output results corresponding to all successfully trained content optimization sub-models are obtained, and the output results corresponding to all successfully trained content optimization sub-models are sent from different clients to a unified target server. At the target server, the aggregation strategy preset in the federated learning algorithm is used to aggregate the output results corresponding to all successfully trained content optimization sub-models to obtain a target output result, wherein the preset aggregation strategy is the FedAvg aggregation algorithm. By parsing the target output result, the browsing keywords contained in the target output result and the location information of each browsing keyword in the browser page are obtained, and the browsing keywords contained in the target output result and the location information of each browsing keyword in the browser page are used as model configuration parameter information to construct a target content optimization model.

[0135] This application obtains the full amount of data in the document to be displayed; sends the full amount of data to the content optimization models corresponding to the first user and the second user respectively; obtains the first optimization data output from the content optimization model corresponding to the first user; obtains the second optimization data output from the content optimization model corresponding to the second user; displays the first optimization data on the first browser page for the first user to view; and displays the second optimization data on the second browser page for the second user to view. By adopting the federated learning algorithm and the FedAvg aggregation algorithm, the content optimization models are trained for different categories of users respectively, so that when displaying the same text data, different display focus data are extracted according to the identity information of different categories of users. For different categories of users, data can be personalized displayed on the corresponding browser page according to different focus data. It is convenient to highlight the focus data in combination with the identity information of the data viewer, which is convenient for viewing and analysis. By adopting differential privacy and homomorphic encryption technology to encrypt the output results corresponding to all the successfully trained content optimization sub-models, the privacy and security of data browsing are guaranteed when sending browsing behavior data to the server.

[0136] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware via computer-readable instructions. The computer-readable instructions can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes in the above-described method embodiments. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0137] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0138] To solve the above technical problems, the present application also provides a computer device. Figure 9 , Figure 9 This is a basic structural block diagram of the computer device in this embodiment.

[0139] The computer device 9 includes a memory 9a, a processor 9b, and a network interface 9c that are interconnected through a system bus. It should be noted that the figure only shows a computer device 9 with components 9a-9c, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0140] The computer device may be a desktop computer, notebook computer, PDA, cloud server, etc. The computer device may interact with the user via a keyboard, mouse, remote control, touchpad, or voice control device.

[0141] The memory 9a includes at least one type of readable storage medium, including flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic storage, a magnetic disk, an optical disk, etc. In some embodiments, the memory 9a can be an internal storage unit of the computer device 9, such as the hard disk or memory of the computer device 9. In other embodiments, the memory 9a can also be an external storage device of the computer device 9, such as a plug-in hard disk equipped on the computer device 9, a smart media card (SMC), a secure digital (SD) card, a flash memory card, etc. Of course, the memory 9a can also include both the internal storage unit of the computer device 9 and its external storage device. In this embodiment, the memory 9a is generally used to store the operating system and various application software installed on the computer device 9, such as computer-readable instructions for a data display method. In addition, the memory 9a can also be used to temporarily store various types of data that have been output or are to be output.

[0142] In some embodiments, the processor 9b may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 9b is generally used to control the overall operation of the computer device 9. In this embodiment, the processor 9b is used to execute computer-readable instructions stored in the memory 9a or process data, such as computer-readable instructions for executing the data display method.

[0143] The network interface 9c may include a wireless network interface or a wired network interface. The network interface 9c is generally used to establish a communication connection between the computer device 9 and other electronic devices.

[0144] The computer device proposed in this embodiment belongs to the field of financial technology and is used in scenarios where financial insurance policy data is classified and highlighted according to different reviewers. This application obtains the full amount of data in the document to be displayed; sends the full amount of data to the content optimization models corresponding to the first user and the second user respectively; obtains the first optimization data output from the content optimization model corresponding to the first user; obtains the second optimization data output from the content optimization model corresponding to the second user; displays the first optimization data on the first browser page for the first user to view; displays the second optimization data on the second browser page for the second user to view. By adopting the federated learning algorithm and the FedAvg aggregation algorithm, content optimization models are trained for different categories of users respectively, so that when displaying the same text data, different display focus data are extracted according to the identity information of different categories of users. For different categories of users, personalized data display can be performed on the corresponding browser page according to different focus data. It is convenient to highlight the focus data in combination with the identity information of the data viewer, which is convenient for viewing and analysis. By adopting differential privacy and homomorphic encryption technology, the output results corresponding to all successfully trained content optimization sub-models are encrypted separately, ensuring the privacy and security of data browsing when sending browsing behavior data to the server.

[0145] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by a processor to enable the processor to perform the steps of the data display method as described above.

[0146] The computer-readable storage medium proposed in this embodiment belongs to the field of financial technology and is applied to scenarios where financial insurance policy data is classified and highlighted according to different reviewers. This application obtains the full amount of data in the document to be displayed; sends the full amount of data to the content optimization models corresponding to the first user and the second user respectively; obtains the first optimization data output from the content optimization model corresponding to the first user; obtains the second optimization data output from the content optimization model corresponding to the second user; displays the first optimization data on the first browser page for the first user to view; displays the second optimization data on the second browser page for the second user to view. By adopting the federated learning algorithm and the FedAvg aggregation algorithm, content optimization models are trained for different categories of users respectively, so that when displaying the same text data, different display focus data are extracted according to the identity information of different categories of users. For different categories of users, personalized data display can be performed on the corresponding browser page according to different focus data. It is convenient to highlight the focus data in combination with the identity information of the data viewer, which is convenient for viewing and analysis. By adopting differential privacy and homomorphic encryption technology, the output results corresponding to all successfully trained content optimization sub-models are encrypted separately, ensuring the privacy and security of data browsing when sending browsing behavior data to the server.

[0147] Through the above description of the embodiment, those skilled in the art can clearly understand that the above embodiment method can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by means of software plus the necessary general hardware platform.

[0148] Based on this understanding, the technical solution of this application, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, or optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of this application.

[0149] Obviously, the embodiments described above are only some of the embodiments of the present application, rather than all of the embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions described in the aforementioned specific embodiments, or to make equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the present application specification and the accompanying drawings, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present application.

Claims

1. A data display method, characterized in that: The steps include: Acquire all data in the document to be displayed and the objects to be displayed, wherein the objects to be displayed include the first user and the second user; respectively setting the first user and the second user as target users; Use browser scripts to collect the target user's browsing history data content of the corresponding browser page and the pause time when browsing each line of historical data content; Construct the target user's corresponding browsing behavior representation data based on the target user's browsing history data content on the corresponding browser page and the pause time when browsing each line of historical data content; Based on the target user's corresponding browsing behavior representation data, the content optimization model corresponding to the target user is trained, specifically including: Step 301: Obtain browsing behavior characterization data corresponding to a target user, and input the browsing behavior characterization data into a browsing behavior analysis model built into a local client to analyze the browsing behavior of the target user and obtain a browsing behavior analysis result corresponding to the target user; Step 302: identifying the focus data browsed by the target user based on the browsing behavior analysis results of the target user; Step 303: training a content optimization sub-model corresponding to the target user on the local client based on the focus data browsed by the target user; Step 304: Identify the identity information of the target user according to a preset identity information table, wherein the identity information table includes the identity information of the first user and the identity information of the second user; Step 305: Using the target user's identity information as a search field, select N users with the same identity information as the target user from all users to construct a similar user set, where N is a positive integer. Step 306: Select different users from the similar user set as target users in sequence, and repeat steps 301 to 303 to train content optimization sub-models corresponding to all users in the similar user set on the local client. Step 307: Obtain all successfully trained content optimization sub-models, and use a federated learning algorithm to comprehensively optimize all successfully trained content optimization sub-models to obtain a target content optimization model. Step 308: Obtain model configuration parameter information of the target content optimization model, distribute the model configuration parameter information to the clients of all users having the same identity information as the target user through the service console, and set the target content optimization model on the clients of all users having the same identity information as the target user according to the model configuration parameter information; Sending the full amount of data to the content optimization models corresponding to the first user and the second user respectively; Obtaining first optimization data output from a content optimization model corresponding to the first user; Obtaining second optimization data output from the content optimization model corresponding to the second user; Displaying the first optimization data on a first browser page for viewing by the first user; The second optimization data is displayed on a second browser page for viewing by the second user.

2. The data display method according to claim 1, characterized in that: The steps of obtaining browsing behavior characterization data corresponding to the target user, inputting the browsing behavior characterization data into a browsing behavior analysis model built into the local client, performing browsing behavior analysis on the target user, and obtaining browsing behavior analysis results corresponding to the target user specifically include: Obtain browsing behavior representation data corresponding to the target user; Obtaining, through the browsing behavior analysis model, historical data content browsed by the target user on the corresponding browser page and the pause time when browsing each line of historical data content; Constructing a correlation map between each row of historical data content and its corresponding pause time according to the pause time when the target user browses each row of historical data content; Outputting the association control map as the browsing behavior analysis result through the browsing behavior analysis model; The step of identifying the focus data browsed by the target user based on the browsing behavior analysis results corresponding to the target user specifically includes: Identifying the pause time of the target user when browsing each row of historical data content through the association comparison map; Comparing the pause time when the target user browses each line of historical data content with a preset pause time threshold, and obtaining a magnitude relationship between the pause time when the target user browses each line of historical data content and the preset pause time threshold; If the target user's pause time when browsing the current row of historical data content is greater than the preset pause time threshold, the current row of historical data content is the focus data; If the target user's pause time when browsing the current row of historical data content is not greater than the preset pause time threshold, the current row of historical data content is non-focus data.

3. The data display method according to claim 1, characterized in that: The step of training the content optimization sub-model corresponding to the target user on the local client based on the focus data browsed by the target user specifically includes: Obtain all focus data browsed by the target user; Obtain the location information of different focus data in the corresponding browser page through DOM parsing; Construct training data pairs based on all focus data and the location information of different focus data in the corresponding browser pages; Inputting the training data into a preset content optimization sub-model, analyzing each browsing keyword contained in all focus data and the position information of each browsing keyword in the browser page through the content optimization sub-model and a preset word frequency algorithm, and outputting each browsing keyword and the position information of each browsing keyword in the browser page as an output result of the content optimization sub-model, wherein the preset content optimization sub-model is a deep learning model based on a CNN neural network architecture; If the output result of the content optimization sub-model is a null value, the training of the content optimization sub-model fails; If the output result of the content optimization sub-model is not a null value, the content optimization sub-model is trained successfully.

4. The data display method according to claim 1, wherein: The step of obtaining all successfully trained content optimization sub-models and comprehensively optimizing all successfully trained content optimization sub-models using a federated learning algorithm to obtain a target content optimization model specifically includes: Obtain the output results corresponding to all successfully trained content optimization sub-models; Sending the output results corresponding to all successfully trained content optimization sub-models from different clients to a unified target server; On the target server, a preset aggregation strategy in the federated learning algorithm is used to aggregate the output results corresponding to all successfully trained content optimization sub-models to obtain a target output result, wherein the preset aggregation strategy is the FedAvg aggregation algorithm; By parsing the target output result, obtaining each browsing keyword contained in the target output result and the position information of each browsing keyword in the browser page; Each browsing keyword included in the target output result and the position information of each browsing keyword in the browser page are used as model configuration parameter information to construct a target content optimization model.

5. The data display method according to claim 4, characterized in that: Before executing the step of sending the output results corresponding to all successfully trained content optimization sub-models from different clients to a unified target server, the method further includes: Differential privacy and homomorphic encryption technologies are used to encrypt the output results corresponding to all successfully trained content optimization sub-models, and an encrypted result corresponding to each output result is obtained. When encrypting the output results corresponding to different content optimization sub-models, the differential privacy algorithm used is consistent and the homomorphic encryption algorithm used is also consistent. The step of sending the output results corresponding to all successfully trained content optimization sub-models from different clients to a unified target server specifically includes: Using the encryption result corresponding to each output result as transmission data; Send the transmission data corresponding to different clients to a unified target server.

6. A data display device, characterized in that: include: A document data acquisition module, configured to acquire the full amount of data in the document to be displayed and the objects to be displayed, wherein the objects to be displayed include the first user and the second user; a target user setting module, configured to set the first user and the second user as target users respectively; The browsing behavior data acquisition module is used to collect the historical data content of the target user's browsing of the corresponding browser page and the pause time when browsing each line of historical data content through the browser script; A browsing behavior representation data construction module is used to construct the browsing behavior representation data corresponding to the target user based on the historical data content browsed by the target user on the corresponding browser page and the pause time when browsing each line of the historical data content; The content optimization model training module is used to train the content optimization model corresponding to the target user based on the browsing behavior representation data corresponding to the target user. The data display device also includes a content optimization sub-model training module and a content optimization sub-model integration module, wherein: A content optimization sub-model training module is used to obtain all focus data browsed by the target user, and is also used to obtain the position information of different focus data in the corresponding browser page through DOM parsing. It is also used to construct training data pairs based on all focus data and the position information of different focus data in the corresponding browser page, and is also used to input the training data pairs into a preset content optimization sub-model, and analyze the various browsing keywords contained in all focus data and the position information of each browsing keyword in the browser page through the content optimization sub-model and a preset word frequency algorithm, and output the various browsing keywords and the position information of each browsing keyword in the browser page as the output result of the content optimization sub-model, wherein the preset content optimization sub-model is a deep learning model based on the CNN neural network architecture, and is also used to determine that if the output result of the content optimization sub-model is a null value, the content optimization sub-model training fails, and if the output result of the content optimization sub-model is not a null value, the content optimization sub-model training is successful; A content optimization sub-model integration module is used to obtain all successfully trained content optimization sub-models, and use a federated learning algorithm to comprehensively optimize all the successfully trained content optimization sub-models to obtain a target content optimization model. Specifically, the output results corresponding to all successfully trained content optimization sub-models are obtained, and the output results corresponding to all successfully trained content optimization sub-models are sent from different clients to a unified target server. On the target server, an aggregation strategy preset in the federated learning algorithm is used to aggregate the output results corresponding to all successfully trained content optimization sub-models to obtain a target output result, wherein the preset aggregation strategy is the FedAvg aggregation algorithm. By parsing the target output result, each browsing keyword contained in the target output result and the position information of each browsing keyword in the browser page are obtained, and each browsing keyword contained in the target output result and the position information of each browsing keyword in the browser page are used as model configuration parameter information to construct a target content optimization model. a document data sending module, configured to send the full amount of data to the content optimization models corresponding to the first user and the second user respectively; A first optimization data acquisition module, configured to acquire first optimization data output from the content optimization model corresponding to the first user; A second optimization data acquisition module, configured to acquire second optimization data output from the content optimization model corresponding to the second user; A first display module, configured to display the first optimization data on a first browser page for viewing by the first user; The second display module is used to display the second optimization data on a second browser page for viewing by the second user.

7. A computer device comprising a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the data presentation method according to any one of claims 1 to 5 when executing the computer-readable instructions.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the data presentation method according to any one of claims 1 to 5.

Images