Customer portrait construction system and method based on ETC data

Through joint modeling of encrypted data in the financial and ETC data subsystem, the problem of incomplete corporate portraits in the transportation industry is solved, and the comprehensiveness and accuracy of customer portraits are improved without obtaining plain text ETC data.

CN114529336BActive Publication Date: 2024-11-12BANK OF CHINA FINANCIAL TECH CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202210100759.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-27
Publication Date
2024-11-12
Estimated Expiration
2042-01-27

AI Technical Summary

Technical Problem

When financial institutions provide financial services to transportation industry enterprises, they can only use limited information to perform customer portraits, resulting in incomplete portraits of transportation industry enterprises, and ETC data cannot be directly provided to financial institutions in plain text, resulting in incomplete and inaccurate portraits.

Method used

Design a customer profile building system based on ETC data, including financial subsystem and ETC data subsystem. Through customer profile calculation module, data application module, financial data component and ETC data component, use pre-trained financial data model and ETC data component. The data model performs joint modeling of encrypted data to obtain customer portraits, ensuring that the data does not leave the local area to achieve comprehensiveness and accuracy of the portrait.

Benefits of technology

Without obtaining plain text ETC data, the comprehensiveness and accuracy of customer portraits are improved and the accuracy and completeness of user portraits are ensured.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114529336B_ABST
    Figure CN114529336B_ABST
Patent Text Reader

Abstract

The present invention relates to a customer portrait construction system and method based on ETC data, the system includes a financial subsystem and an ETC data subsystem, the financial subsystem includes a customer portrait calculation module, a first data application module, a financial data component, and a first portrait module, the ETC data subsystem includes a second data application module, an ETC data component, and a second portrait module, the first portrait module and the second portrait module can obtain encrypted financial data, encrypted ETC data and jointly model to obtain customer portraits. Compared with the prior art, the ETC data subsystem and the financial subsystem of the present invention jointly draw customer portraits on the premise that the original data does not leave the local area, thereby improving the comprehensiveness and accuracy of the portraits.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of customer profile construction, and in particular to a customer profile construction system and method based on ETC data. Background Art

[0002] The national ETC data records the highway traffic status of each vehicle with the corresponding license plate, including the time, where the vehicle entered the highway, where the vehicle left the highway, how much highway toll was paid, the type of vehicle passed, the loading status, etc.; the ETC electronic invoice data records the employer organization information of a vehicle's trip, which can reflect the operating behavior of the transportation company.

[0003] When financial institutions provide financial services to companies in the transportation industry, they can currently only use limited information to profile their customers, and often cannot fully measure companies in the transportation industry. However, when financial institutions introduce ETC data, ETC data cannot be directly provided to financial institutions in plain text to draw corporate portraits, which will cause certain incompleteness in the corporate portraits of the transportation industry. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a customer portrait construction system and method based on ETC data.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A customer portrait construction system based on ETC data includes a financial subsystem and an ETC data subsystem, wherein the financial subsystem includes a customer portrait calculation module, a first data application module, a financial data component, and a first portrait module, and the ETC data subsystem includes a second data application module, an ETC data component, and a second portrait module.

[0007] The customer portrait calculation module is used to initiate a customer portrait request; the first data application module is used to receive the customer portrait request and send data requests to the financial data component and the second data application module respectively; the financial data component is used to send the encrypted financial data corresponding to the data request to the first portrait module;

[0008] The second data application module is used to send the data request to the ETC data component, and the ETC data component is used to send the encrypted ETC data corresponding to the data request to the second portrait module;

[0009] The first portrait module and the second portrait module are used to obtain customer portraits based on the joint modeling of encrypted financial data and encrypted ETC data.

[0010] Preferably, the first portrait module is equipped with a pre-trained financial data model and a joint model, and the second portrait module is equipped with a pre-trained ETC data model and a joint model.

[0011] The first portrait module sends the input encrypted financial data to the financial data model for processing, obtains primary financial portrait data, and sends it to the joint model of the second portrait module.

[0012] The second portrait module sends the input encrypted ETC data to the ETC data model for processing, obtains primary ETC portrait data, and sends it to the joint model of the first portrait module.

[0013] The joint model of the first portrait module and the second portrait module obtains the user portrait based on the acquired primary financial portrait data and primary ETC portrait data.

[0014] Preferably, the financial data model, ETC data model, and joint model include but are not limited to any one of the following models: YOLO series, Faster RCNN, MaskRCNN, SSD, CenterNet, and DyHead.

[0015] Preferably, the financial data component includes a first data platform and a first encryption and decryption module. The first data platform is used to store the customer's financial data and send the corresponding customer's financial data to the first encryption and decryption module according to the data request of the first data application module. The first encryption and decryption module is used to encrypt the customer's financial data and send it to the first training module.

[0016] Preferably, the first data platform includes a financial internal database and an ETC external database dictionary, the financial internal database is used to store the customer's financial data, and the ETC external database dictionary is used to store the definition and description of the user's ETC data.

[0017] Preferably, the ETC data component includes a second data platform and a second encryption and decryption module. The second data platform is used to store the customer's ETC data and send the corresponding customer's ETC data to the second encryption and decryption module according to the data request of the second data application module; the second encryption and decryption module is used to encrypt the customer's ETC data and send it to the second training module.

[0018] Preferably, the second data platform includes an ETC internal database and a financial external database dictionary, the ETC internal database is used to store the customer's ETC data, and the financial external database dictionary is used to store the definition and description of the user's financial data.

[0019] A method for constructing a customer profile based on ETC data, based on the above-mentioned system for constructing a customer profile based on ETC data, is characterized by comprising the following steps:

[0020] S1: The customer portrait calculation module initiates a customer portrait request and sends it to the first data application module;

[0021] S2: The first data application module receives the customer portrait request and sends a data request to the financial data component and the second data application module respectively;

[0022] S3: The financial data component sends the encrypted financial data corresponding to the data request to the first portrait module;

[0023] S4: The second data application module sends the data request to the ETC data component, and the ETC data component sends the encrypted ETC data corresponding to the data request to the second portrait module;

[0024] S5: The first portrait module and the second portrait module jointly model the encrypted financial data and the encrypted ETC data to obtain the customer portrait.

[0025] Preferably, the first portrait module is equipped with a pre-trained financial data model and a joint model, and the second portrait module is equipped with a pre-trained ETC data model and a joint model.

[0026] Preferably, the step S5 specifically includes:

[0027] S51: the first portrait module sends the input encrypted financial data to the financial data model for processing, obtains primary financial portrait data, and sends it to the joint model of the second portrait module;

[0028] S52: the second portrait module sends the input encrypted ETC data to the ETC data model for processing, obtains primary ETC portrait data, and sends it to the joint model of the first portrait module;

[0029] S53: The joint model of the first portrait module and the second portrait module obtains the user portrait based on the acquired primary financial portrait data and primary ETC portrait data.

[0030] Compared with the prior art, the present invention has the following advantages:

[0031] 1) The present invention can make the ETC data subsystem and the financial subsystem jointly draw customer portraits without obtaining the plain text ETC and without the original data being exported locally, thereby improving the comprehensiveness and accuracy of the portraits;

[0032] 2) The present invention utilizes the first portrait module and the second portrait module to perform joint modeling to obtain customer portraits, and can process encrypted user data to improve the accuracy and completeness of user portrait acquisition. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the system structure of the present invention;

[0034] Figure 2 The figure is a flow chart of the method of the present invention.

[0035] Among them, 1. Financial subsystem, 101. Customer portrait calculation module, 102. First data application module, 103. First data platform, 1031. Financial internal database, 1032. ETC external database dictionary, 104. First encryption and decryption module, 105. First portrait module, 2. ETC data subsystem, 201. Second data application module, 202. Second data platform, 2021. ETC internal database, 2022. Financial external database dictionary, 203. Second encryption and decryption module, 204. Second portrait module. DETAILED DESCRIPTION

[0036] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. Note that the following embodiments are merely illustrative in nature, and the present invention is not intended to limit its applicable objects or uses, and the present invention is not limited to the following embodiments.

[0037] Example

[0038] A customer profile building system based on ETC data, such as Figure 1 As shown, it includes a financial subsystem 1 and an ETC data subsystem 2. The financial subsystem 1 includes a customer portrait calculation module 101, a first data application module 102, a financial data component, and a first portrait module 105. The ETC data subsystem 2 includes a second data application module 201, an ETC data component, and a second portrait module 204.

[0039] The customer portrait calculation module 101 is used to initiate a customer portrait request; the first data application module 102 is used to receive the customer portrait request and send data requests to the financial data component and the second data application module 201 respectively; the financial data component is used to send the encrypted financial data corresponding to the data request to the first portrait module 105;

[0040] The second data application module 201 is used to send the data request to the ETC data component, and the ETC data component is used to send the encrypted ETC data corresponding to the data request to the second portrait module 204;

[0041] The first portrait module 105 and the second portrait module 204 are used to obtain customer portraits based on the joint modeling of encrypted financial data and encrypted ETC data.

[0042] In this embodiment, the first portrait module 105 is equipped with a pre-trained financial data model and a joint model, and the second portrait module 204 is equipped with a pre-trained ETC data model and a joint model. The first portrait module 105 sends the input encrypted financial data to the financial data model for processing, obtains primary financial portrait data, and sends it to the joint model of the second portrait module 204. The second portrait module 204 sends the input encrypted ETC data to the ETC data model for processing, obtains primary ETC portrait data, and sends it to the joint model of the first portrait module 105. The joint model of the first portrait module 105 and the second portrait module 204 obtains the user portrait based on the obtained primary financial portrait data and primary ETC portrait data.

[0043] Specifically, the financial data model, ETC data model, and joint model include but are not limited to any one of the following models: YOLO series, Faster RCNN, MaskRCNN, SSD, CenterNet, and DyHead.

[0044] When training the financial data model, ETC data model, and joint model, the labeled customers and their corresponding encrypted financial data and encrypted ETC data are used to iteratively train the financial data model, ETC data model, and joint model and optimize its parameters, so that the user portrait prediction accuracy of the model is higher than the set threshold.

[0045] In addition, in this embodiment, the financial data component includes a first data platform 103 and a first encryption and decryption module 104. The first data platform 103 is used to store the customer's financial data, and send the corresponding customer's financial data to the first encryption and decryption module 104 according to the data request of the first data application module 102. The first encryption and decryption module 104 is used to encrypt the customer's financial data and send it to the first training module. The first data platform 103 includes a financial internal database 1031 and an ETC external database dictionary 1032. The financial internal database 1031 is used to store the customer's financial data, and the ETC external database dictionary 1032 is used to store the definition and description of the user's ETC data.

[0046] The ETC data component includes a second data platform 202 and a second encryption and decryption module 203. The second data platform 202 is used to store the customer's ETC data and send the corresponding customer's ETC data to the second encryption and decryption module 203 according to the data request of the second data application module 201; the second encryption and decryption module 203 is used to encrypt the customer's ETC data and send it to the second training module. The second data platform 202 includes an ETC internal database 2021 and a financial external database dictionary 2022. The ETC internal database 2021 is used to store the customer's ETC data, and the financial external database dictionary 2022 is used to store the definition and description of the user's financial data.

[0047] Corresponding to the above system, the present invention also provides a method for constructing a customer profile based on ETC data, such as Figure 2 As shown, the following steps are included:

[0048] S1: The customer profile calculation module 101 initiates a customer profile request and sends it to the first data application module 102;

[0049] S2: The first data application module 102 receives the customer portrait request and sends a data request to the financial data component and the second data application module 201 respectively;

[0050] S3: The financial data component sends the encrypted financial data corresponding to the data request to the first portrait module 105;

[0051] S4: The second data application module 201 sends the data request to the ETC data component, and the ETC data component sends the encrypted ETC data corresponding to the data request to the second portrait module 204;

[0052] S5: The first portrait module 105 and the second portrait module 204 jointly model the encrypted financial data and the encrypted ETC data to obtain the customer portrait.

[0053] Wherein, step S5 specifically includes:

[0054] S51: the first portrait module 105 sends the input encrypted financial data to the financial data model for processing, obtains primary financial portrait data, and sends it to the joint model of the second portrait module 204;

[0055] S52: The second portrait module 204 sends the input encrypted ETC data to the ETC data model for processing, obtains primary ETC portrait data, and sends it to the joint model of the first portrait module 105;

[0056] S53: The joint model of the first portrait module 105 and the second portrait module 204 obtains the user portrait based on the acquired primary financial portrait data and primary ETC portrait data.

[0057] The above embodiments are merely examples and do not limit the scope of the present invention. These embodiments can also be implemented in various other ways, and various omissions, substitutions, and changes can be made without departing from the technical concept of the present invention.

Claims

1. A customer profile building system based on ETC data, characterized in that: The invention comprises a financial subsystem (1) and an ETC data subsystem (2), wherein the financial subsystem (1) comprises a customer portrait calculation module (101), a first data application module (102), a financial data component, and a first portrait module (105), and the ETC data subsystem (2) comprises a second data application module (201), an ETC data component, and a second portrait module (204). The customer portrait calculation module (101) is used to initiate a customer portrait request; the first data application module (102) is used to receive the customer portrait request and send a data request to the financial data component and the second data application module (201) respectively; the financial data component is used to send the encrypted financial data corresponding to the data request to the first portrait module (105); The second data application module (201) is used to send a data request to the ETC data component, and the ETC data component is used to send the encrypted ETC data corresponding to the data request to the second portrait module (204); The first portrait module (105) and the second portrait module (204) are used to obtain customer portraits by joint modeling based on encrypted financial data and encrypted ETC data; The first portrait module (105) is equipped with a pre-trained financial data model and a joint model, and the second portrait module (204) is equipped with a pre-trained ETC data model and a joint model. The first portrait module (105) sends the input encrypted financial data to the financial data model for processing, obtains primary financial portrait data, and sends it to the joint model of the second portrait module (204). The second portrait module (204) sends the input encrypted ETC data to the ETC data model for processing, obtains primary ETC portrait data, and sends it to the joint model of the first portrait module (105). The joint model of the first portrait module (105) and the second portrait module (204) obtains the user portrait based on the acquired primary financial portrait data and primary ETC portrait data.

2. A customer profile building system based on ETC data according to claim 1, characterized in that: The financial data model, ETC data model, and joint model include any one of the following models: YOLO series, FasterRCNN, MaskRCNN, SSD, CenterNet, and DyHead.

3. A customer profile building system based on ETC data according to claim 1, characterized in that: The financial data component comprises a first data platform (103) and a first encryption and decryption module (104). The first data platform (103) is used to store the financial data of the customer and send the corresponding customer's financial data to the first encryption and decryption module (104) according to the data request of the first data application module (102). The first encryption and decryption module (104) is used to encrypt the customer's financial data and send it to the first training module.

4. A customer profile building system based on ETC data according to claim 3, characterized in that: The first data platform (103) includes a financial internal database (1031) and an ETC external database dictionary (1032), wherein the financial internal database (1031) is used to store the financial data of the customer, and the ETC external database dictionary (1032) is used to store the definition and description of the user's ETC data.

5. A customer profile building system based on ETC data according to claim 1, characterized in that: The ETC data component comprises a second data platform (202) and a second encryption and decryption module (203). The second data platform (202) is used to store the customer's ETC data and send the corresponding customer's ETC data to the second encryption and decryption module (203) according to the data request of the second data application module (201); the second encryption and decryption module (203) is used to encrypt the customer's ETC data and send it to the second training module.

6. A customer profile building system based on ETC data according to claim 5, characterized in that: The second data platform (202) includes an ETC internal database (2021) and a financial external database dictionary (2022), wherein the ETC internal database (2021) is used to store the customer's ETC data, and the financial external database dictionary (2022) is used to store the definition and description of the user's financial data.

7. A method for constructing a customer profile based on ETC data, based on a system for constructing a customer profile based on ETC data as claimed in any one of claims 1 to 6, characterized in that: The following steps are involved: S1: the customer portrait calculation module (101) initiates a customer portrait request and sends it to the first data application module (102); S2: the first data application module (102) receives the customer portrait request and sends a data request to the financial data component and the second data application module (201) respectively; S3: The financial data component sends the encrypted financial data corresponding to the data request to the first portrait module (105); S4: the second data application module (201) sends the data request to the ETC data component, and the ETC data component sends the encrypted ETC data corresponding to the data request to the second portrait module (204); S5: The first portrait module (105) and the second portrait module (204) jointly model the encrypted financial data and the encrypted ETC data to obtain a customer portrait; The first portrait module (105) is equipped with a pre-trained financial data model and a joint model, and the second portrait module (204) is equipped with a pre-trained ETC data model and a joint model; The step S5 specifically includes: S51: the first portrait module (105) sends the input encrypted financial data to the financial data model for processing, obtains primary financial portrait data, and sends it to the joint model of the second portrait module (204); S52: the second portrait module (204) sends the input encrypted ETC data to the ETC data model for processing, obtains primary ETC portrait data, and sends it to the joint model of the first portrait module (105); S53: The joint model of the first portrait module (105) and the second portrait module (204) obtains the user portrait based on the acquired primary financial portrait data and primary ETC portrait data.

Citation Information

Patent Citations

  • Method for generating user portrait, and terminal

    CN110710190A

  • Highway network user portrait information acquisition and analysis method and system

    CN111291236A

  • Joint training method and device for machine learning model, equipment and storage medium

    CN111612167A

  • Aviation service customization system and method based on federated learning technology platform

    CN111899076A

  • Urban traffic travel data analysis method based on federated learning

    CN112163979A