Credit card recommendation willingness identification method and related device
By extracting basic information and behavioral data of credit card customers, and using a neural network model based on Hilbert transform to identify credit card recommendation intentions, the problem of inaccurate identification of credit card old customers in the existing technology is solved, and accurate recommendation intention recognition and targeted invitations are achieved.
Patent Information
- Application Number
- CN202411742867.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-29
- Publication Date
- 2025-02-25
AI Technical Summary
The existing technology cannot accurately identify whether old credit card customers are willing to recommend credit cards to new customers, resulting in banks being unable to accurately target when recommending credit cards, causing trouble to old customers who do not have the intention to recommend.
By obtaining the basic information and behavioral data of the target credit card customer, the characteristics that have predictive ability of credit card recommendation intention are extracted and input them into the neural network model based on the Hilbert transform for identification. If the probability of recommendation intention is greater than the threshold, it is determined that the customer has credit card recommendation intention.
It realizes accurate identification of credit card recommendation intentions of credit card customers, which facilitates accurate targeted invitations to old customers who are willing to recommend, and improves the efficiency of recommendation activities and customer satisfaction.
Smart Images
Figure CN119515534A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of artificial intelligence technology, and in particular to a credit card recommendation intention recognition method and related devices. Background Art
[0002] Recommending new customers by old credit card customers is an important strategy for banks to achieve new customer growth, that is, encouraging old credit card customers to recommend new customers to apply for credit cards.
[0003] At present, since it is impossible to accurately identify whether old credit card customers are willing to recommend credit cards to new customers, banks generally randomly or widely invite old credit card customers to recommend credit cards to new customers, such as calling old credit card customers or sending event invitation information to encourage old credit card customers to participate in new customer recommendation activities. However, this method cannot accurately send event invitation information to old customers who are willing to recommend credit cards, and it will cause trouble to old customers who are not willing to recommend credit cards. Summary of the invention
[0004] In view of the above problems, this application provides a credit card recommendation intention identification method and related devices to accurately identify the credit card recommendation intention of target credit card customers, so as to facilitate the subsequent precise and targeted invitation to old customers with credit card recommendation intention. The specific scheme is as follows:
[0005] The first aspect of the present application provides a method for identifying credit card recommendation intentions, comprising:
[0006] Obtain basic information and behavioral data of target credit card customers;
[0007] Extracting a first preset feature from the basic information and behavior data of the target credit card customer, wherein the first preset feature is a feature that has predictive power for credit card recommendation willingness;
[0008] Inputting the first preset feature into a credit card recommendation intention recognition model to obtain a recommendation intention probability output by the credit card recommendation intention recognition model, wherein the credit card recommendation intention recognition model is obtained by pre-training a Hilbert transform-based neural network model using a training set, and the Hilbert transform-based neural network model includes a first neural network, a plurality of Hilbert layers, and a second neural network;
[0009] If the recommendation willingness probability is greater than a threshold, it is determined that the target credit card customer has a credit card recommendation willingness.
[0010] In a possible implementation, a method for determining the first preset feature includes:
[0011] Input a test sample into the XGBoost model, wherein the test sample includes basic information and behavior data of a sample credit card customer and is pre-labeled with a label of having a recommendation intention or having no recommendation intention;
[0012] Determine the number of times each feature in the test sample is used as a split node in all trees of the XGBoost model;
[0013] Sort the number of times each feature is used as a split node from high to low;
[0014] The first N features in the ranking are determined as the first preset features, where N is an integer greater than 1.
[0015] In a possible implementation, inputting the first preset feature into a credit card recommendation intention recognition model to obtain a recommendation intention probability output by the credit card recommendation intention recognition model includes:
[0016] Inputting the first preset feature into the credit card recommendation intention recognition model;
[0017] Using the first neural network to convert the first preset feature into high-dimensional information;
[0018] Using a plurality of the Hilbert layers in sequence to perform feature selection on the high-dimensional information;
[0019] The second neural network is used to convert the data output by the last Hilbert layer into the recommendation willingness probability.
[0020] In a possible implementation, after determining that the target credit card customer has a credit card recommendation willingness, the credit card recommendation willingness identification method further includes:
[0021] Extracting a second preset feature from the basic information and behavior data of the target credit card customer, wherein the second preset feature is a feature having predictive capability for credit card customer classification;
[0022] The second preset feature is input into the credit card customer classification model to obtain the target customer type output by the credit card customer classification model.
[0023] In a possible implementation, after obtaining the target customer type output by the credit card customer classification model, the credit card recommendation intention identification method further includes:
[0024] Determine the target invitation method corresponding to the target customer type according to the preset correspondence between the customer type and the invitation method;
[0025] The target invitation method is used to invite the target credit card customer to recommend a credit card to a new customer.
[0026] A second aspect of the present application provides a credit card recommendation intention identification device, comprising:
[0027] Data acquisition module, used to obtain basic information and behavior data of target credit card customers;
[0028] A feature extraction module, used to extract a first preset feature from the basic information and behavior data of the target credit card customer, wherein the first preset feature is a feature that has predictive power for credit card recommendation willingness;
[0029] a model processing module, configured to input the first preset feature into a credit card recommendation intention recognition model to obtain a recommendation intention probability output by the credit card recommendation intention recognition model, wherein the credit card recommendation intention recognition model is obtained by pre-training a Hilbert transform-based neural network model using a training set, and the Hilbert transform-based neural network model includes a first neural network, a plurality of Hilbert layers, and a second neural network;
[0030] The determination module is used to determine that the target credit card customer has a credit card recommendation intention if the recommendation intention probability is greater than a threshold.
[0031] In a possible implementation, the credit card recommendation intention identification device further includes:
[0032] The feature determination module is used to input a test sample into the XGBoost model, wherein the test sample includes basic information and behavioral data of a sample credit card customer and is pre-labeled with a label of having a recommendation intention or not having a recommendation intention; determine the number of times each feature in the test sample is used as a split node in all trees of the XGBoost model; sort the number of times each feature is used as a split node in order from high to low; and determine the first N features in the sorting as the first preset features, where N is an integer greater than 1.
[0033] A third aspect of the present application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:
[0034] The memory is used to store computer programs;
[0035] The processor is used to execute the computer program so that the electronic device can implement the credit card recommendation intention identification method of the above-mentioned first aspect or any implementation method of the first aspect.
[0036] The fourth aspect of the present application provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements the credit card recommendation intention identification method of the above-mentioned first aspect or any implementation method of the first aspect.
[0037] In a fifth aspect, the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the credit card recommendation intention identification method of the above-mentioned first aspect or any implementation method of the first aspect.
[0038] Through the above technical scheme, the present application provides a method and related devices for identifying the willingness to recommend a credit card. After acquiring the basic information and behavioral data of the target credit card customer, preset features with predictive capabilities for the willingness to recommend a credit card are extracted therefrom, and the preset features are input into a credit card recommendation willingness identification model. Each Hilbert layer in the credit card recommendation willingness identification model performs feature selection through the Hilbert transform, eliminates redundant features, and achieves accurate identification of the target credit card customer's willingness to recommend a credit card, so as to facilitate subsequent precise and targeted invitations to old customers with the willingness to recommend a credit card. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.
[0040] Figure 1 A system architecture diagram provided for an embodiment of the present application;
[0041] Figure 2 A schematic diagram of a terminal structure provided in an embodiment of the present application;
[0042] Figure 3 A schematic diagram of a server structure provided in an embodiment of the present application;
[0043] Figure 4 A flowchart of a method for identifying credit card recommendation intention provided in an embodiment of the present application;
[0044] Figure 5 A schematic diagram of the structure of a neural network model provided in an embodiment of the present application;
[0045] Figure 6 A schematic diagram of the structure of a neural network model based on Hilbert transform provided in an embodiment of the present application;
[0046] Figure 7A flowchart of a credit card recommendation intention recognition model training method provided in an embodiment of the present application;
[0047] Figure 8 A schematic diagram of the structure of a device for identifying credit card recommendation intention provided in an embodiment of the present application. DETAILED DESCRIPTION
[0048] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.
[0049] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.
[0050] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.
[0051] See also Figure 1 , Figure 1 A schematic diagram of a system architecture is shown. The system may include a terminal 100 and a server 200. The server 200 may include one or more servers ( Figure 1 In the example, a server is included, and the server 200 can provide the credit card recommendation intention identification method provided in the embodiment of the present application for one or more terminals.
[0052] Among them, an application can be installed on the terminal 100, and the above application can provide an interface. The terminal 100 can receive relevant parameters entered by the user on the interface, such as the target credit card customer or the basic information and behavior data of the target credit card customer, and send the above parameters to the server 200. The server 200 can obtain processing results based on the received parameters, such as whether the target credit card customer has the intention to recommend a credit card, and return the processing results to the terminal 100.
[0053] It should be understood that in some optional implementations, the terminal 100 can also complete the action of obtaining the processing result based on the received parameters by itself. Specifically, it can identify whether the target credit card customer has the intention to recommend a credit card based on the basic information and behavior data of the target credit card customer, without the need for server cooperation, and the embodiments of the present application are not limited thereto.
[0054] Next describe Figure 1 The product form of the mid-terminal 100;
[0055] The terminal 100 in the embodiment of the present application can be a mobile phone, a tablet computer, a wearable device, an augmented reality (AR) / virtual reality (VR) device, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), etc., and the embodiment of the present application does not impose any limitation on this.
[0056] Figure 2 An optional hardware structure diagram of the terminal 100 is shown.
[0057] refer to Figure 2 As shown, the terminal 100 may include a radio frequency unit 110, a memory 120, an input unit 130, a display unit 140, a camera 150 (optional), an audio circuit 160 (optional), a speaker 161 (optional), a microphone 162 (optional), an earphone jack 163 (optional), a processor 170, an external interface 180, a power supply 190 and other components. Those skilled in the art will appreciate that Figure 2 This is merely an example of a terminal and does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown in the figure, or a combination of certain components, or different components.
[0058] The input unit 130 can be used to receive input digital or character information, and generate key signal input related to the user settings and function control of the terminal 100. Specifically, the input unit 130 may include a touch screen 131 (optional) and / or other input devices 132. The touch screen 131 can collect the user's touch operations on or near it (such as the user's operation on or near the touch screen using any suitable object such as fingers, joints, stylus, etc.), and drive the corresponding connection device according to a pre-set program. The touch screen can detect the user's touch action on the touch screen, convert the touch action into a touch signal and send it to the processor 170, and can receive and execute the command sent by the processor 170; the touch signal at least includes the touch point coordinate information. The touch screen 131 can provide an input interface and an output interface between the terminal 100 and the user. In addition, the touch screen can be implemented using multiple types such as resistive, capacitive, infrared, and surface acoustic wave. In addition to the touch screen 131, the input unit 130 can also include other input devices. Specifically, other input devices 132 may include, but are not limited to, one or more of a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, a joystick, and the like.
[0059] Among them, the input device 132 can receive input data and the like.
[0060] The display unit 140 can be used to display information input by the user or information provided to the user, various menus of the terminal 100, interactive interfaces, file display and / or playback of any multimedia file. In the embodiment of the present application, the display unit 140 can be used to display the input interface, processing results, etc. of the target credit card customer or the target credit card customer's basic information and behavior data.
[0061] The memory 120 can be used to store instructions and data. The memory 120 can mainly include an instruction storage area and a data storage area. The data storage area can store various data, such as multimedia files, texts, etc.; the instruction storage area can store software units such as operating systems, applications, instructions required for at least one function, or their subsets and extensions. It can also include a non-volatile random access memory; provide the processor 170 with hardware, software and data resources including management of computing and processing equipment, and support control software and applications. It is also used for the storage of multimedia files, and the storage of running programs and applications.
[0062] The processor 170 is the control center of the terminal 100. It uses various interfaces and lines to connect various parts of the entire terminal 100. By running or executing instructions stored in the memory 120 and calling data stored in the memory 120, it executes various functions of the terminal 100 and processes data, thereby controlling the terminal device as a whole. Optionally, the processor 170 may include one or more processing units; preferably, the processor 170 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface and application program, and the modem processor mainly processes wireless communication. It is understandable that the above-mentioned modem processor may not be integrated into the processor 170. In some embodiments, the processor and the memory may be implemented on a single chip, and in some embodiments, they may also be implemented separately on separate chips. The processor 170 may also be used to generate corresponding operation control signals, send them to corresponding components of the computing and processing device, read and process data in the software, especially read and process data and programs in the memory 120, so that each functional module therein performs corresponding functions, thereby controlling the corresponding components to act according to the requirements of the instructions.
[0063] Among them, the memory 120 can be used to store software codes related to the credit card recommendation intention identification method, the processor 170 can execute the steps of the credit card recommendation intention identification method, and can also schedule other units (such as the above-mentioned input unit 130 and display unit 140) to realize corresponding functions.
[0064] The radio frequency unit 110 (optional) can be used for receiving and sending information or receiving and sending signals during a call, for example, after receiving the downlink information of the base station, it is sent to the processor 170 for processing; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (Low Noise Amplifier, LNA), a duplexer, etc. In addition, the radio frequency unit 110 can also communicate with network devices and other devices through wireless communication. The wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile communication (Global System of Mobile communication, GSM), General Packet Radio Service (General Packet Radio Service, GPRS), Code Division Multiple Access (Code Division Multiple Access, CDMA), Wideband Code Division Multiple Access (Wideband Code Division Multiple Access, WCDMA), Long Term Evolution (Long Term Evolution, LTE), email, Short Messaging Service (SMS), etc.
[0065] In this embodiment of the present application, the RF unit 110 can send data to the server 200 and receive processing results sent by the server 200.
[0066] It should be understood that the radio frequency unit 110 is optional and can be replaced by other communication interfaces, such as a network port.
[0067] The terminal 100 also includes a power supply 190 (such as a battery) for supplying power to various components. Preferably, the power supply can be logically connected to the processor 170 through a power management system, so that the power management system can manage functions such as charging, discharging, and power consumption.
[0068] The terminal 100 further includes an external interface 180 , which may be a standard Micro USB interface or a multi-pin connector, and may be used to connect the terminal 100 to communicate with other devices, or to connect a charger to charge the terminal 100 .
[0069] Although not shown, the terminal 100 may also include a flashlight, a wireless fidelity (WiFi) module, a Bluetooth module, sensors with different functions, etc., which are not described in detail here. Some or all of the methods described below may be applied in the following embodiments. Figure 2 In the terminal 100 shown.
[0070] Next describe Figure 1 The product form of the server 200;
[0071] Figure 3 A schematic diagram of the structure of a server 200 is provided, such as Figure 3 As shown, the server 200 includes a bus 201, a processor 202, a communication interface 203, and a memory 204. The processor 202, the memory 204, and the communication interface 203 communicate with each other via the bus 201.
[0072] The bus 201 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.
[0073] The processor 202 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).
[0074] The memory 204 may include a volatile memory, such as a random access memory (RAM). The memory 204 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).
[0075] Among them, the memory 204 can be used to store software codes related to the credit card recommendation intention identification method, the processor 202 can execute the steps of the credit card recommendation intention identification method, and can also schedule other units to achieve corresponding functions.
[0076] It should be understood that the above-mentioned terminal 100 and server 200 can be centralized or distributed devices, and the processors in the above-mentioned terminal 100 and server 200 (such as processor 170 and processor 202) can be hardware circuits (such as application specific integrated circuit (ASIC), field-programmable gate array (FPGA), general-purpose processor, digital signal processor (DSP), microprocessor or microcontroller, etc.), or a combination of these hardware circuits. For example, the processor can be a hardware system with an instruction execution function, such as a CPU, DSP, etc., or a hardware system without an instruction execution function, such as an ASIC, FPGA, etc., or a combination of the above-mentioned hardware systems without an instruction execution function and hardware systems with an instruction execution function.
[0077] The present application embodiment provides a credit card recommendation intention recognition method. The credit card recommendation intention recognition method of the present application embodiment is described in detail below with reference to the accompanying drawings.
[0078] Reference Figure 4 , Figure 4 A flow chart of a credit card recommendation intention identification method provided in an embodiment of the present application is shown as follows: Figure 4 As shown, a credit card recommendation intention identification method provided in an embodiment of the present application may include steps 401 to 404, and these steps are described in detail below.
[0079] 401: Obtain the basic information and behavior data of the target credit card customer;
[0080] The target credit card customer can be any credit card customer, that is, an old credit card customer.
[0081] The basic information of target credit card customers includes gender, age, average AUM (Average Asset UnderManagement), etc.
[0082] The behavioral data of target credit card customers comes from consumption repayment behavior, customer behavior tracking points and other external data.
[0083] 402: extracting a first preset feature from the basic information and behavior data of the target credit card customer, where the first preset feature is a feature that has predictive power for credit card recommendation willingness;
[0084] The basic information and behavioral data of the target credit card customers are preprocessed, such as data cleaning, feature extraction and data normalization, to obtain the key data of the target credit card customers.
[0085] Key data include: consumption habits, repayment habits, bill amounts, activity participation data, behavioral data, consumption preferences, frequently used social networks, etc. Among them, consumption habits include: number of consumptions, consumption amounts, consumption frequency, etc., repayment habits include: whether to pay in installments, whether to repay on time, etc., activity participation data include: activity participation types, such as credit card activities, bank activities, etc., consumption preferences include: food, daily necessities, travel, etc.
[0086] Then, a first preset feature is extracted from the key data. The first preset feature is a predetermined feature that has predictive power for credit card recommendation willingness.
[0087] 403: Inputting the first preset feature into the credit card recommendation intention recognition model to obtain the recommendation intention probability output by the credit card recommendation intention recognition model;
[0088] The credit card recommendation intention recognition model is obtained by pre-training a neural network model based on Hilbert transform using a training set, and the training set includes training samples that are pre-labeled with recommendation intention and without recommendation intention.
[0089] 404: If the recommendation intention probability is greater than the threshold, it is determined that the target credit card customer has a credit card recommendation intention.
[0090] The threshold is set according to the accuracy requirement in the actual application scenario. The larger the threshold, the higher the reliability of the target credit card customer's willingness to be recommended a credit card.
[0091] The present embodiment discloses a method for identifying the willingness to recommend a credit card. After acquiring the basic information and behavior data of the target credit card customer, preset features with predictive capabilities for the willingness to recommend a credit card are extracted therefrom, and the preset features are input into a credit card recommendation willingness identification model. Each Hilbert layer in the credit card recommendation willingness identification model performs feature selection through Hilbert transform, removes redundant features, and achieves accurate identification of the target credit card customer's willingness to recommend a credit card, so as to facilitate subsequent precise and targeted invitations to old customers with the willingness to recommend a credit card.
[0092] In a possible implementation, the method for determining the first preset feature in the above embodiment includes the following steps A1-A4:
[0093] A1: Input the test sample into the XGBoost model. The test sample includes the basic information and behavior data of the sample credit card customers, and is pre-labeled with the label of recommendation intention or non-recommendation intention;
[0094] A2: Determine the number of times each feature in the test sample is used as a split node in all trees of the XGBoost model. The more times, the more important the feature is.
[0095] A3: Sort the number of times each feature is used as a split node from high to low;
[0096] A4: Determine the first N features in the above ranking as the first preset features, where N is an integer greater than 1.
[0097] In a possible implementation, the average depth of the splitting point of each feature in the test sample in all trees of the XGBoost model can also be determined. The larger the average depth, the more important the feature is. Then, the average depth of the splitting point of each feature in all trees of the XGBoost model is sorted from high to low, and the first N features in the sorting are determined as the first preset features.
[0098] In a possible implementation, the coverage of samples by each feature in all trees of the XGBoost model can also be calculated, that is, the total number of samples covered when the feature is used to split nodes. The more samples are covered, the more important the feature is. Then, the coverage of samples by each feature in all trees of the XGBoost model is sorted from high to low, and the first N features in the sorting are determined as the first preset features.
[0099] In a possible implementation, the average gain brought by the splitting of each feature on each tree of the XGBoost model can also be determined. The larger the gain, the greater the contribution of the feature to the model and the more important it is. Then, the average gain brought by the splitting of each feature on each tree of the XGBoost model is sorted from high to low, and the first N features in the sorting are determined as the first preset features.
[0100] like Figure 5 The neural network model shown in the figure includes an input layer, 6 hidden layers and an output layer. Both the input layer and the output layer contain 20 neurons. The structure of the hidden layer has both vector and matrix forms. However, the basic mathematical expressions between layers are:
[0101] .
[0102] Among them, x is the input vector, y is the output vector, W and V are weight matrices, b is the bias vector, σ is the activation function, σ is a nonlinear function, commonly used are relu, sigmoid and tanh, etc. Based on this shallow neural network model, it can be naturally extended to a deep neural network model, where the expression of the L-layer neural network includes L-1 hidden layers, and its expression is as follows:
[0103] .
[0104] Among them, W L and V L is the weight matrix of the L-th layer neural network, W L-1 , W L-2 and W 1 are the weight matrices of the L-1, L-2 and 1st layers of the neural network, b L 、b L-1 b 1 They are the bias vectors of the Lth, L-1th and 1st layers of the neural network respectively.
[0105] Based on this, this embodiment proposes a neural network model based on Hilbert transform, such as Figure 6 As shown, the neural network model based on Hilbert transform includes a first neural network P, multiple Hilbert layers and a second neural network Q, a(x) represents the first preset feature input into the model, and u(x) represents the probability of recommendation willingness.
[0106] Each Hilbert layer performs feature selection and removes redundant features. The mathematical expression of the Hilbert layer is:
[0107]
[0108] Where W and V are weight matrices, represents the Hilbert transform.
[0109] The Hilbert transform is a linear operator that produces a function H(u)(t) with the same domain as the function u(t).
[0110] The Hilbert transform of u can be considered as the convolution of u(t) and the function h(t)=1 / (πt). Since h(t) is not integrable, the integral defining the convolution does not converge. Therefore, the Hilbert transform is calculated using the Cauchy principal value ( ). The Hilbert transform of the function u(t) (in this embodiment, representing the input data of the Hilbert layer) is:
[0111]
[0112] Compared with the basic neural network layer, the key point of the Hilbert transform layer is the introduction of a mapping operator. The Hilbert transform layer only retains specific features each time, so that it only focuses on the core part, so that the Hilbert transform layer can achieve good results with fewer iterations. At the same time, the Hilbert transform layer can also use the Hilbert transform to process input data of different lengths.
[0113] In a possible implementation, inputting the first preset feature into the credit card recommendation intention recognition model to obtain the recommendation intention probability output by the credit card recommendation intention recognition model includes the following steps 4031-4034:
[0114] 4031: inputting the first preset feature into a credit card recommendation intention recognition model;
[0115] 4032: Convert the first preset feature into high-dimensional information using a first neural network;
[0116] 4033: Use multiple Hilbert layers in sequence to select features of high-dimensional information;
[0117] 4034: Use the second neural network to convert the data output by the last Hilbert layer into the probability of recommendation willingness.
[0118] The second neural network includes an output layer that uses an activation function to convert the output of the second neural network into a probability distribution.
[0119] In a possible implementation, after determining that the target credit card customer has a willingness to recommend a credit card, an invitation may be sent to the target credit card customer to encourage the target credit card customer to participate in a new customer recommendation activity.
[0120] In another possible implementation, in order to further determine the target customer type of the target credit card customer, thereby initiating an invitation to the target credit card customer in an invitation manner corresponding to the target customer type, and increasing the probability of the target credit card customer participating in the new customer recommendation activity, after determining that the target credit card customer has a credit card recommendation intention, the credit card recommendation intention identification method provided in this embodiment further includes the following steps B1-B2:
[0121] B1: Extract the second preset feature from the basic information and behavior data of the target credit card customers;
[0122] The second preset feature is a feature that has predictive power for credit card customer classification.
[0123] The method for determining the second preset feature is consistent with the principle of the method for determining the first preset feature in the above embodiment.
[0124] B2: Input the second preset feature into the credit card customer classification model to obtain the target customer type output by the credit card customer classification model.
[0125] The credit card customer classification model is pre-trained using a training set, which includes training samples pre-labeled with customer types. The credit card customer classification model can be trained based on any classification model, such as a decision tree model, a naive Bayes model, a random forest model, a neural network model, etc.
[0126] Further, after determining the target customer type, the credit card recommendation intention identification method provided in this embodiment further includes the following steps C1-C2:
[0127] C1: Determine the target invitation method corresponding to the target customer type based on the pre-set correspondence between the customer type and the invitation method;
[0128] C2: Use targeted invitation method to invite target credit card customers to recommend credit cards to new customers.
[0129] Exemplarily, customer types include offline active type, credit card APP active type, etc. The offline active type refers to customers who actively participate in offline activities, and the credit card APP active type refers to customers who frequently use the credit card APP.
[0130] Exemplarily, the invitation method corresponding to the offline active type may be inviting customers to participate in new customer recommendation activities through offline advertising pages or inviting customers to participate in new customer recommendation activities through manual customer calls; the invitation method corresponding to the credit card APP active type may be displaying new customer recommendation activity information on the credit card APP homepage and / or the key process node page for customers to handle business. Taking the credit card repayment business as an example, the key process node page may be the repayment amount inquiry page.
[0131] By using targeted invitation methods to specifically invite target credit card customers to recommend credit cards to new customers, the success rate of new customer recommendations can be increased.
[0132] See also Figure 7 The flowchart of the credit card recommendation intention identification model training method shown in the figure first prepares the training set and the validation set. Specifically, the basic data of old credit card customers, the basic data of recommended new credit card customers and other external data are obtained, and the samples of old credit card customers with recommendation intention and the samples of old credit card customers without recommendation intention are extracted and labeled, and the samples are divided into the training set and the validation set. Exemplarily, the samples are divided into the training set and the validation set in a ratio of 4:1.
[0133] Then, a neural network model based on Hilbert transform is constructed, and the number of model network layers, optimizer and loss function are designed. The main function of the optimizer is to minimize the loss function by adjusting the network parameters. Common optimizers include stochastic gradient descent (SGD), adaptive moment estimation (Adam) and root mean square propagation (RMSProp). Selecting a suitable optimizer can accelerate the convergence speed of the network, improve training efficiency and generalization ability. In addition, the setting of hyperparameters such as the depth of the network, the number of neurons in each layer, and the learning rate also have a significant impact on the training process. Therefore, it is necessary to determine the best hyperparameter combination through repeated adjustment and verification to obtain a model with good generalization performance.
[0134] Then the training set is used to train the neural network model based on Hilbert transform. The initial parameters of the neural network are usually randomly initialized. In order to make the network fit the data better, multiple trainings are required, and the network weights are adjusted through the back propagation algorithm so that the output of the network is as close to the real result as possible. During the training process, the training set and the validation set are needed to monitor the training and generalization effects of the model respectively. Achieving good performance on the training set does not guarantee the generalization ability of the model on unseen data. Therefore, the validation set is needed to evaluate the generalization performance of the model. Only after the model shows good performance on both the training set and the validation set can it be applied to the test set to ensure the accuracy and reliability of the model.
[0135] The above introduces a credit card recommendation intention identification method provided by an embodiment of the present application. The following will introduce a device for executing the above credit card recommendation intention identification method.
[0136] See also Figure 8 , Figure 8 This is a schematic diagram of the structure of a credit card recommendation intention recognition device provided in an embodiment of the present application. Figure 8 As shown, the credit card recommendation intention identification device includes:
[0137] Data acquisition module 801, used to acquire basic information and behavior data of target credit card customers;
[0138] A feature extraction module 802 is used to extract a first preset feature from the basic information and behavior data of the target credit card customer, wherein the first preset feature is a feature that has predictive power for credit card recommendation willingness;
[0139] The model processing module 803 is used to input the first preset feature into a credit card recommendation intention recognition model to obtain a recommendation intention probability output by the credit card recommendation intention recognition model, wherein the credit card recommendation intention recognition model is obtained by pre-training a Hilbert transform-based neural network model using a training set, and the Hilbert transform-based neural network model includes a first neural network, a plurality of Hilbert layers, and a second neural network;
[0140] The determination module 804 is used to determine that the target credit card customer has a credit card recommendation intention if the recommendation intention probability is greater than a threshold.
[0141] In a possible implementation, the credit card recommendation intention identification device further includes:
[0142] The feature determination module is used to input a test sample into the XGBoost model, wherein the test sample includes basic information and behavioral data of a sample credit card customer and is pre-labeled with a label of having a recommendation intention or not having a recommendation intention; determine the number of times each feature in the test sample is used as a split node in all trees of the XGBoost model; sort the number of times each feature is used as a split node in order from high to low; and determine the first N features in the sorting as the first preset features, where N is an integer greater than 1.
[0143] In a possible implementation, the model processing module 803 is specifically used to input the first preset feature into the credit card recommendation intention recognition model; use the first neural network to convert the first preset feature into high-dimensional information; use multiple Hilbert layers in sequence to perform feature selection on the high-dimensional information; use the second neural network to convert the data output by the last Hilbert layer into the recommendation intention probability.
[0144] In a possible implementation, the credit card recommendation intention identification device further includes:
[0145] The customer classification module is used to extract a second preset feature from the basic information and behavior data of the target credit card customer, wherein the second preset feature is a feature with predictive ability for credit card customer classification; the second preset feature is input into a credit card customer classification model to obtain the target customer type output by the credit card customer classification model.
[0146] In a possible implementation, the credit card recommendation intention identification device further includes:
[0147] The invitation method determination module is used to determine the target invitation method corresponding to the target customer type according to the preset correspondence between the customer type and the invitation method; and use the target invitation method to invite the target credit card customer to recommend a credit card to a new customer.
[0148] This embodiment discloses a device for identifying the willingness to recommend a credit card. After acquiring the basic information and behavior data of a target credit card customer, preset features with predictive capabilities for the willingness to recommend a credit card are extracted therefrom, and the preset features are input into a credit card recommendation willingness identification model. Each Hilbert layer in the credit card recommendation willingness identification model performs feature selection through Hilbert transform, removes redundant features, and achieves accurate identification of the target credit card customer's willingness to recommend a credit card, so as to facilitate subsequent precise and targeted invitations to old customers with the willingness to recommend a credit card.
[0149] An embodiment of the present application also provides an electronic device, including at least one processor and a memory connected to the processor, wherein:
[0150] The memory is used to store computer programs;
[0151] The processor is used to execute a computer program so that the electronic device can implement any one of the credit card recommendation intention identification methods provided in the embodiments of the present application.
[0152] Also provided in an embodiment of the present application is a computer program product including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any one of the credit card recommendation intention identification methods provided in the embodiments of the present application.
[0153] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When one or more computer programs are executed by an electronic device, the electronic device can implement any one of the credit card recommendation intention identification methods provided in the embodiments of the present application.
[0154] It is understandable that before using the technical solutions disclosed in the embodiments of this application, the type, scope of use, usage scenarios, etc. of the personal information involved in this application should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.
[0155] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.
[0156] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.
[0157] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.
[0158] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.
Claims
1. A method for identifying credit card recommendation intention, characterized in that: include: Obtain basic information and behavioral data of target credit card customers; Extracting a first preset feature from the basic information and behavior data of the target credit card customer, wherein the first preset feature is a feature that has predictive power for credit card recommendation willingness; Inputting the first preset feature into a credit card recommendation intention recognition model to obtain a recommendation intention probability output by the credit card recommendation intention recognition model, wherein the credit card recommendation intention recognition model is obtained by pre-training a Hilbert transform-based neural network model using a training set, and the Hilbert transform-based neural network model includes a first neural network, a plurality of Hilbert layers, and a second neural network; If the recommendation willingness probability is greater than a threshold, it is determined that the target credit card customer has a credit card recommendation willingness.
2. The credit card recommendation intention identification method according to claim 1, characterized in that: The method for determining the first preset feature includes: Input a test sample into the XGBoost model, wherein the test sample includes basic information and behavior data of a sample credit card customer and is pre-labeled with a label of having a recommendation intention or having no recommendation intention; Determine the number of times each feature in the test sample is used as a split node in all trees of the XGBoost model; Sort the number of times each feature is used as a split node from high to low; The first N features in the ranking are determined as the first preset features, where N is an integer greater than 1.
3. The credit card recommendation intention identification method according to claim 1, characterized in that: The step of inputting the first preset feature into a credit card recommendation intention recognition model to obtain a recommendation intention probability output by the credit card recommendation intention recognition model includes: Inputting the first preset feature into the credit card recommendation intention recognition model; Using the first neural network to convert the first preset feature into high-dimensional information; Using a plurality of the Hilbert layers in sequence to perform feature selection on the high-dimensional information; The second neural network is used to convert the data output by the last Hilbert layer into the recommendation willingness probability.
4. The credit card recommendation intention identification method according to claim 1, characterized in that: After determining that the target credit card customer has a credit card recommendation intention, the credit card recommendation intention identification method further includes: Extracting a second preset feature from the basic information and behavior data of the target credit card customer, wherein the second preset feature is a feature having predictive capability for credit card customer classification; The second preset feature is input into the credit card customer classification model to obtain the target customer type output by the credit card customer classification model.
5. The credit card recommendation intention identification method according to claim 4, characterized in that: After obtaining the target customer type output by the credit card customer classification model, the credit card recommendation intention identification method further includes: Determine the target invitation method corresponding to the target customer type according to the preset correspondence between the customer type and the invitation method; The target invitation method is used to invite the target credit card customer to recommend a credit card to a new customer.
6. A credit card recommendation intention recognition device, characterized in that: include: Data acquisition module, used to obtain basic information and behavior data of target credit card customers; A feature extraction module, used to extract a first preset feature from the basic information and behavior data of the target credit card customer, wherein the first preset feature is a feature that has predictive power for credit card recommendation willingness; a model processing module, configured to input the first preset feature into a credit card recommendation intention recognition model to obtain a recommendation intention probability output by the credit card recommendation intention recognition model, wherein the credit card recommendation intention recognition model is obtained by pre-training a Hilbert transform-based neural network model using a training set, and the Hilbert transform-based neural network model includes a first neural network, a plurality of Hilbert layers, and a second neural network; The determination module is used to determine that the target credit card customer has a credit card recommendation intention if the recommendation intention probability is greater than a threshold.
7. The credit card recommendation intention identification device according to claim 6, characterized in that: The credit card recommendation intention identification device also includes: The feature determination module is used to input a test sample into the XGBoost model, wherein the test sample includes basic information and behavioral data of a sample credit card customer and is pre-labeled with a label of having a recommendation intention or not having a recommendation intention; determine the number of times each feature in the test sample is used as a split node in all trees of the XGBoost model; sort the number of times each feature is used as a split node in order from high to low; and determine the first N features in the sorting as the first preset features, where N is an integer greater than 1.
8. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the credit card recommendation intention identification method as described in any one of claims 1 to 5.
9. A computer program product, characterized in that It includes computer-readable instructions, which, when executed on an electronic device, enable the electronic device to implement the credit card recommendation intention identification method as described in any one of claims 1 to 5.
10. A computer storage medium, characterized in that: The storage medium carries one or more computer programs, which, when executed by an electronic device, enable the electronic device to implement the credit card recommendation intention identification method as described in any one of claims 1 to 5.