TW202125283A - Customer behavior prediction method and system thereof obtain the future prediction sequence to greatly increase the accuracy of selecting target customer groups - Google Patents

Abstract

A customer behavior prediction system includes a storage module and a processing module electrically connected to the storage module. The storage module stores multiple previous behavior data related to a variety of different behaviors previously performed by a customer to be predicted and respectively corresponding to previous time points. The processing module obtains a previous sequence containing multiple previous parameters corresponding to the previous behavior data according to the previous behavior data. The previous parameters, according to the previous time points sequence and based upon the previous sequence, uses a customer behavior prediction model for predicting the future behavior of the customer to obtain a prediction sequence including a plurality of prediction parameters indicating future possible various behaviors of the customer to be predicted.

Description

Customer behavior prediction method and system

The present invention relates to a behavior prediction method and system, in particular to a behavior prediction method and system using machine learning.

The existing marketing methods in the banking industry are mainly based on the experience of product managers, which manually retrieve customers’ past consumer behavior data and select target customer groups, and verify the selected target customer groups through subsequent marketing activities to determine all Select the pros and cons of the target customer group, and the product manager will finally improve his selection experience based on the feedback of the pros and cons of the target customer group.

However, this manual method of selecting target customer groups not only does not have a fixed standard, it also consumes too much time and cost in training product managers, and the accuracy of the selected target customer groups is not enough to achieve the purpose of financial product promotion.

In view of this, it is necessary to propose a new solution to improve the accuracy of selecting target customer groups.

Therefore, the purpose of the present invention is to provide a method for predicting customer behavior based on machine learning to improve the accuracy of selected target customer groups.

Therefore, the customer behavior prediction method of the present invention is implemented by a computer device, and includes a step (A) and a step (B).

Step (A) is to use the computer device to obtain a number of previous behavior data corresponding to a previous time point based on a plurality of different behaviors related to a customer to be predicted in the past. The previous sequence of the previous parameters of the data, and the previous parameters are sorted according to the previous time points.

Step (B) is to use a customer behavior prediction model for predicting the future behavior of the customer by the computer device based on the previous sequence to obtain a prediction that includes a plurality of prediction parameters indicating the possible future behaviors of the customer to be predicted sequence.

Another object of the present invention is to provide a customer behavior prediction system based on machine learning to improve the accuracy of selecting target customer groups.

Therefore, the customer behavior prediction system of the present invention includes a storage module and a processing module electrically connected to the storage module.

The storage module stores multiple pieces of previous behavior data related to a variety of different behaviors previously performed by a customer to be predicted and each corresponding to a previous point in time.

The processing module obtains a previous sequence containing a plurality of previous parameters corresponding to the previous behavior data according to the previous behavior data, and the previous parameters are sorted according to the previous time points. According to the previous sequence, a The customer behavior prediction model for predicting the future behavior of the customer obtains a prediction sequence that includes a plurality of prediction parameters indicating possible various behaviors of the customer to be predicted in the future.

The effect of the present invention is to use the customer behavior prediction model for predicting the future behavior of the customer by using the previous sequence corresponding to the previous behavior of the customer to be predicted to obtain the future prediction sequence, and the future prediction sequence includes Each of the forecast parameters indicates the possible behavior of the customer to be predicted in the future, and the bank can determine whether the customer to be predicted is the target customer based on the possible behavior of the customer to be predicted in the future, which greatly improves the accuracy of selecting the target customer group Spend.

Referring to FIG. 1, an embodiment of the customer behavior prediction system 1 of the present invention is connected to a client 2 via a communication network 200.

The customer behavior prediction system 1 includes a system-side communication module 11 connected to the communication network 200, a system-side storage module 12, and a system that electrically connects the system-side communication module 11 and the system-side storage module 12 End processing module 13.

The system-side storage module 12 stores multiple training data corresponding to multiple customers, and each training data includes multiple training previous behavior data and multiple training prediction behavior data of the corresponding customer before performing the different behaviors. , And multiple recommendable parameters corresponding to various target customer behaviors and corresponding recommendation information. Among them, each piece of training previous behavior data corresponds to a previous training time point, each piece of training prediction behavior data corresponds to a training prediction time point, and any training prediction time point of the same training data is later than the previous training time points . Among them, each recommended information is financial product information, but it is not limited to this.

The client 2 includes a client communication module 21 connected to the communication network 200, a client display module 22, and a client processing that electrically connects the client communication module 21 and the client display module 22 Module 23.

In this embodiment, the implementation of the customer behavior prediction system 1 is, for example, a personal computer, a server, or a cloud host, but it is not limited to this.

In this embodiment, the implementation aspect of the client terminal 2 is, for example, a personal computer, a smart phone or a tablet computer, but it is not limited to this.

Hereinafter, the system-side communication module 11, the system-side storage module 12, the system-side processing module 13, and the components of the client 2 will be explained by using the client behavior prediction system 1 of the present invention to execute a method for predicting client behavior. Operational details. The customer behavior prediction method includes a model training program and a prediction recommendation program.

Referring to Figure 2, the model training procedure includes steps 51 to 53.

In step 51, for each training data, the system-side processing module 13 obtains a training previous that includes a plurality of training previous parameters corresponding to the training previous behavior data according to the training previous behavior data in the training data sequence. Among them, the previous training parameters are sorted according to the previous training time points.

In step 52, for each training data, the system-side processing module 13 obtains a training prediction including a plurality of training prediction parameters corresponding to the training prediction behavior data according to the training prediction behavior data in the training data Sequence, the training prediction parameters are sorted according to the training prediction time points.

It is worth noting that for each training data, when the previous training behavior data and the training predicted behavior data are customer credit card consumption information, the training previous parameters (input of training data) and the When the training prediction parameters (output of training data) are all merchant category codes (MCC CODE, Merchant Category Code, used to indicate the type of goods or services provided by the merchant); and, when the training previous behavior data and the training When the predicted behavior data is the customer's Internet browsing information, the previous training parameters (input of training data) and the training prediction parameters (output of training data) are all website codes, but the above examples are not limited . Among them, each URL in the web browsing information can correspond to a website code.

In step 53, the system-side processing module 13 uses the cyclic neural network to obtain a customer behavior prediction model for predicting the future behavior of the customer based on the training data. Among them, the recurrent neural network is a recurrent neural network (RNN, Recurrent Neural Networks) or a long short-term memory model (LSTM, Long Short-Term Memory), but it is not limited to this.

Referring to Figure 3, the predictive recommendation program includes steps 61 to 65.

In step 61, the system-side processing module 13 obtains a plurality of previous behavior data corresponding to a previous time point based on a plurality of previous behavior data related to a plurality of different behaviors previously performed by a customer to be predicted. The previous sequence of the previous parameter of the behavior data. Among them, the previous parameters are sorted according to the previous time points.

In step 62, the system-side processing module 13 uses the trained customer behavior prediction model based on the previous sequence to obtain a prediction sequence containing multiple prediction parameters indicating possible various behaviors of the client to be predicted in the future.

It is worth noting that when the previous behavior data is customer credit card consumption information, and the previous parameters are merchant category codes, then the forecast parameters are merchant category codes, and the forecast parameters mean The types of stores that customers may consume in the future; and when the previous behavioral data is the customer’s web browsing information, and the previous parameters are the website codes, the predicted parameters are the website codes. The prediction parameters indicate the websites that customers may click to browse in the future, but are not limited to the above examples. Among them, each URL in the web browsing information can correspond to a website code.

In step 63, the system-side processing module 13 determines whether there is at least one parameter to be recommended that is the same as any one of the recommendable parameters among the prediction parameters of the prediction sequence. When the system-side processing module 13 determines that there is no parameter to be recommended, the predictive recommendation procedure ends; when the system-side processing module 13 determines that there is at least one parameter to be recommended, the process step 64 is performed.

In step 64, the system-side processing module 13 transmits the recommendation information corresponding to each parameter to be recommended to the client 2 through the system-side communication module 11.

In step 65, the client processing module 23, after receiving the recommendation information corresponding to each parameter to be recommended through the client communication module 21, displays the recommendation information corresponding to each parameter to be recommended to the client端display module 22.

In summary, the client behavior prediction system 1 of the present invention uses the client processing module 23 to pre-process each training data obtained in advance to obtain the training corresponding to the training data and including the training previous parameters. The previous sequence, and the training prediction sequence corresponding to the training data and containing the training prediction parameters, and using the recurrent neural network to use the training previous sequence as the input and the training prediction sequence as the expected output to train the client Behavior prediction model; then, the client processing module 23 uses the trained customer behavior prediction model to obtain the future prediction sequence according to the previous sequence corresponding to the previous behavior of the client to be predicted, and the future prediction sequence Each prediction parameter included indicates the possible future behavior of the customer to be predicted, and recommends corresponding financial products based on the possible behavior of the customer to be predicted in the future, and the bank can determine the possible behavior of the customer to be predicted in the future Whether the customer to be predicted is a target customer, thereby greatly improving the accuracy of selecting the target customer group, so that the recommended financial products can better meet the needs of the customer to be predicted. Therefore, it can indeed achieve the purpose of the invention.

However, the above are only examples of the present invention. When the scope of implementation of the present invention cannot be limited by this, all simple equivalent changes and modifications made in accordance with the scope of the patent application of the present invention and the content of the patent specification still belong to Within the scope covered by the patent of the present invention.

200: Communication network 1: Customer behavior prediction system 11: System side communication module 12: System-side storage module 13: System-side processing module 2: client 21: Client communication module 22: Client display module 23: Client processing module 51~53: Steps 61~65: Steps

Other features and effects of the present invention will be clearly presented in the embodiments with reference to the drawings, in which: Figure 1 is a block diagram illustrating an embodiment of the customer behavior prediction system of the present invention; FIG. 2 is a flowchart illustrating a model training procedure of a method for predicting customer behavior executed in this embodiment; and Fig. 3 is a flowchart illustrating a prediction recommendation procedure of the customer behavior prediction method.

200: Communication network

1: Customer behavior prediction system

11: System side communication module

12: System-side storage module

13: System-side processing module

2: client

21: Client communication module

22: Client display module

23: Client processing module

Claims (10)

A method for predicting customer behavior is implemented by a computer device. The method for predicting customer behavior includes the following steps: (A) Using the computer device, according to multiple pieces of previous behavior data related to a variety of different behaviors previously performed by a customer to be predicted and each corresponding to a previous point in time, a data containing multiple data corresponding to the previous behavior is obtained The previous sequence of previous parameters, and the previous parameters are sorted according to the previous time points; and (B) Using the computer device, according to the previous sequence, a customer behavior prediction model for predicting the future behavior of the customer is used to obtain a prediction sequence containing a plurality of prediction parameters indicating the possible future behaviors of the customer to be predicted. According to the customer behavior prediction method of claim 1, the computer device is also connected to a client via a communication network, and the computer device stores a plurality of recommendable parameters corresponding to various target customer behaviors and corresponding recommendation information. Among them, the customer behavior prediction method also includes the following steps: (C) Using the computer device, determine whether there is at least one parameter to be recommended that is the same as any one of the recommendable parameters among the predictive parameters of the predictive sequence; and (D) When it is determined that the at least one parameter to be recommended exists, the computer device transmits the recommendation information corresponding to each parameter to be recommended to the client. The method for predicting customer behavior according to claim 1, wherein the computer device stores a plurality of training data corresponding to a plurality of customers, and each training data includes a plurality of trainings of the corresponding customer before performing the different behaviors Previous behavior data and multiple training prediction behavior data, each training previous behavior data corresponds to a previous training time point, each training prediction behavior data corresponds to a training prediction time point, and any training prediction time point of the same training data All are later than the previous time points of these trainings, and before step (A), the following steps are also included: (E) For each training data, by the computer device, according to the training previous behavior data in the training data, a training previous sequence containing a plurality of training previous parameters corresponding to the training previous behavior data is obtained, the Prior training parameters are sorted according to the previous training time points; (F) For each training data, use the computer device to obtain a training prediction sequence containing a plurality of training prediction parameters corresponding to the training prediction behavior data based on the training prediction behavior data in the training data, the The training prediction parameters are sorted according to the training prediction time points; and (G) Through the computer device, according to the training data, the cyclic neural network is used to obtain the customer behavior prediction model. The customer behavior prediction method according to claim 1, wherein: In step (A), the previous behavior data is the customer's credit card consumption information, and the previous parameters are the merchant category code; and In step (B), the prediction parameters are merchant category codes. The customer behavior prediction method according to claim 1, wherein: In step (A), the previous behavior data is the customer's Internet browsing information, and the previous parameters are the website code; and In step (B), these prediction parameters are website codes. A customer behavior prediction system, including: A storage module, storing multiple pieces of previous behavior data related to a variety of different behaviors previously performed by a customer to be predicted and each corresponding to a previous point in time; and A processing module, electrically connected to the storage module, obtains a previous sequence containing multiple previous parameters corresponding to the previous behavior data based on the previous behavior data, and the previous parameters are sorted according to the previous time points. The previous sequence uses a customer behavior prediction model for predicting the future behavior of the customer to obtain a prediction sequence that includes a plurality of prediction parameters indicating possible various behaviors of the customer to be predicted in the future. The customer behavior prediction system according to claim 6, further comprising a communication module connected to a client via a communication network and electrically connected to the processing module, and the storage module also stores a plurality of behaviors corresponding to various target customers Recommendable parameters and corresponding recommendation information, the processing module determines whether there is at least one parameter to be recommended that is the same as any one of the recommendable parameters in the prediction parameters of the prediction sequence, and when the processing When the module determines that the at least one parameter to be recommended exists, the processing module transmits the recommendation information corresponding to each parameter to be recommended to the client through the communication module. The customer behavior prediction system according to claim 6, wherein the storage module also stores a plurality of training data corresponding to multiple customers, and each training data includes the corresponding customer's previous multiple behaviors. Previous training behavior data and multiple training prediction behavior data. Each previous training behavior data corresponds to a previous training time point. Each training prediction behavior data corresponds to a training prediction time point. Any training prediction of the same training data The time points are all later than the training previous time points. For each training data, the processing module obtains a training previous behavior data corresponding to the training previous behavior data according to the training previous behavior data in the training data. The training previous sequence of parameters, the training previous parameters are sorted according to the training previous time points, for each training data, the processing module obtains a data containing multiple correspondences based on the training prediction behavior data in the training data The training prediction sequence of the training prediction parameters of the training prediction behavior data, the training prediction parameters are sorted according to the training prediction time points, and the processing module uses the recurrent neural network to obtain the customer behavior prediction model based on the training data . For example, the customer behavior prediction system according to claim 6, wherein the previous behavior data is customer credit card consumption information, the previous parameters are merchant category codes, and the prediction parameters are merchant category codes. For example, the customer behavior prediction system according to claim 6, wherein the previous behavior data is customer web browsing information, the previous parameters are website codes, and the prediction parameters are website codes.

Images