CN116091185A - Customer loss early warning method and device - Google Patents
Customer loss early warning method and device Download PDFInfo
- Publication number
- CN116091185A CN116091185A CN202211500041.8A CN202211500041A CN116091185A CN 116091185 A CN116091185 A CN 116091185A CN 202211500041 A CN202211500041 A CN 202211500041A CN 116091185 A CN116091185 A CN 116091185A
- Authority
- CN
- China
- Prior art keywords
- customer
- information
- data
- loss
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/02—Banking, e.g. interest calculation or account maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Engineering & Computer Science (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Software Systems (AREA)
- Human Resources & Organizations (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Technology Law (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a customer loss early warning method and a device, which relate to artificial intelligence, and the method comprises the following steps: acquiring customer portrait information and customer behavior information; establishing a BP neural network model, and presetting a desired output sample set; training the BP neural network model according to the customer portrait information and the customer behavior information to obtain a customer loss early warning model, and determining an actual output result; and determining customer loss risk information according to a preset expected output sample set and an actual output result. The invention can predict potential loss risk hidden trouble of the client in the service, so that the client manager can reduce the loss of the client in advance and take corresponding measures, thereby reducing the occurrence probability of the loss.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a customer loss early warning method and device.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
At present, a customer loss early warning block is generally obtained by running batches through a plurality of banking systems according to transaction information of customers to obtain a customer list which possibly has abnormality under different behaviors, and then a customer manager screens the customer list according to own experience so as to carry out customer maintenance; since most of the batch processed data are not real-time, the fact that the risk is problematic when the post-batch report is generated is existed, the customer manager cannot be notified in advance, and it is difficult to make a coping plan in advance to avoid or control the risk of customer loss.
In the process of maintaining clients, a current bank client manager mainly analyzes client behavior through report data after a plurality of systems are batched, and especially for the current situation of each system vertical shaft, the data acquisition of the part cannot be very timely; under more conditions, report data information can be simply sorted according to a unified standard, the difference among different client groups is not considered, a client manager cannot directly blindly use the report data information, screening and maintenance are required to be carried out by combining the client information, and the report data information can not predict the loss risk of the client in advance while time and labor are wasted.
Therefore, how to provide a new solution to the above technical problem is a technical problem to be solved in the art.
Disclosure of Invention
The embodiment of the invention provides a customer loss early warning method, which predicts potential loss risk hidden trouble of a customer in a service so as to facilitate a customer manager to reduce customer loss in advance and take countermeasures and reduce the occurrence probability of loss, and comprises the following steps:
acquiring customer portrait information and customer behavior information;
establishing a BP neural network model, and presetting a desired output sample set;
training the BP neural network model according to the customer portrait information and the customer behavior information to obtain a customer loss early warning model, and determining an actual output result;
and determining customer loss risk information according to a preset expected output sample set and an actual output result.
The embodiment of the invention also provides a customer loss early warning device, which comprises:
the information acquisition module is used for acquiring customer portrait information and customer behavior information;
the BP neural network model building module is used for building a BP neural network model and presetting a desired output sample set;
the model training module is used for training the BP neural network model according to the customer portrait information and the customer behavior information to obtain a customer loss early warning model and determining an actual output result;
and the client loss risk information determining module is used for determining client loss risk information according to a preset expected output sample set and an actual output result.
The embodiment of the invention also provides computer equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the client loss early warning method is realized when the processor executes the computer program.
The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program realizes the client loss early warning method when being executed by a processor.
The embodiment of the invention also provides a computer program product, which comprises a computer program, wherein the computer program realizes the client loss early warning method when being executed by a processor.
The client loss early warning method and device provided by the embodiment of the invention comprise the following steps: acquiring customer portrait information and customer behavior information; establishing a BP neural network model, and presetting a desired output sample set; training the BP neural network model according to the customer portrait information and the customer behavior information to obtain a customer loss early warning model, and determining an actual output result; and determining customer loss risk information according to a preset expected output sample set and an actual output result. A customer loss early warning method based on decision trees and BP neural networks aims at analyzing business behaviors of customers, such as transfer behaviors, payment behaviors, login intervals, asset changes and the like, through massive data of customer group portraits, establishing a customer loss model, predicting the risk rate of customer loss and timely notifying a customer manager of carrying out customer saving work. Establishing a BP neural network by collecting customer group information and a customer real-time behavior data set, and outputting a customer loss threshold index under multiple customer groups by training the BP network for multiple times; therefore, potential loss risk hidden danger of the client in the service is predicted, so that a client manager can reduce loss of the client in advance and take countermeasures, and the occurrence probability of loss is reduced.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
fig. 1 is a schematic diagram of a customer loss early warning method according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a process of establishing a BP neural network model of a client churn early warning method according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a training process of a BP neural network model of a client churn early warning method according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of a training process of a BP neural network model of a client churn early warning method according to an embodiment of the present invention.
Fig. 5 is a schematic diagram of a computer device running a customer churn early warning method implemented in the present invention.
Fig. 6 is a schematic diagram of a customer loss early warning device according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present invention and their descriptions herein are for the purpose of explaining the present invention, but are not to be construed as limiting the invention.
Fig. 1 is a schematic diagram of a customer loss early warning method according to an embodiment of the present invention, as shown in fig. 1, the embodiment of the present invention provides a customer loss early warning method, which predicts potential loss risk hidden danger of a customer in a service, so that a customer manager reduces customer loss in advance and takes countermeasures, and reduces the occurrence probability of loss, the method includes:
step 101: acquiring customer portrait information and customer behavior information;
step 102: establishing a BP neural network model, and presetting a desired output sample set;
step 103: training the BP neural network model according to the customer portrait information and the customer behavior information to obtain a customer loss early warning model, and determining an actual output result;
step 104: and determining customer loss risk information according to a preset expected output sample set and an actual output result.
The client loss early warning method provided by the embodiment of the invention comprises the following steps: acquiring customer portrait information and customer behavior information; establishing a BP neural network model, and presetting a desired output sample set; training the BP neural network model according to the customer portrait information and the customer behavior information to obtain a customer loss early warning model, and determining an actual output result; and determining customer loss risk information according to a preset expected output sample set and an actual output result. A customer loss early warning method based on decision trees and BP neural networks aims at analyzing business behaviors of customers, such as transfer behaviors, payment behaviors, login intervals, asset changes and the like, through massive data of customer group portraits, establishing a customer loss model, predicting the risk rate of customer loss and timely notifying a customer manager of carrying out customer saving work. Establishing a BP neural network by collecting customer group information and a customer real-time behavior data set, and outputting a customer loss threshold index under multiple customer groups by training the BP network for multiple times; therefore, potential loss risk hidden danger of the client in the service is predicted, so that a client manager can reduce loss of the client in advance and take countermeasures, and the occurrence probability of loss is reduced.
Neural network: the method is an algorithm mathematical model which imitates the behavior characteristics of a human neural network and performs distributed parallel information processing. BP neural network: an error backward propagation neural network is the most widely used type in a neural network model; it is divided into an input layer, a hidden layer and an output layer.
With the wide application of machine learning and big data analysis in banking industry, the acquisition of portrait attributes of clients and behavior attributes of clients in a system is more and more convenient. In order to ensure the liveness of the existing clients of the bank, the system can better serve the existing users, early warning of the loss of the clients can be carried out by utilizing the data of each dimension of the clients collected by big data, and risk intervention is carried out in advance. In the banking field, the diversity of customers can make it difficult to determine customer churn, so a machine learning method combined with a neural network is needed to analyze customer churn according to customer portrait information and behavior information.
The customer retention in the current commercial banking industry is taken as one of the most important concerns, a great amount of manpower and material resources are consumed in daily work to keep the customer activity, a post-processing mode is generally adopted in the face of customer loss, and the situation can be found after the customer activity is reduced, so that the customer loss can be better prevented by earlier finding the lost Miao of the customer, and the purpose of maintaining the customer activity is achieved. The invention provides a customer loss early warning method based on a decision tree and a BP neural network, which aims to analyze business behaviors of customers, such as transfer behaviors, payment behaviors, login intervals, asset changes and the like, through massive data of customer group portraits, establish a customer loss model, predict the risk rate of customer loss and timely inform a customer manager to carry out customer saving work.
When the client loss early warning method provided by the embodiment of the invention is implemented, in one embodiment, the method comprises the following steps:
acquiring customer portrait information and customer behavior information;
establishing a BP neural network model, and presetting a desired output sample set;
training the BP neural network model according to the customer portrait information and the customer behavior information to obtain a customer loss early warning model, and determining an actual output result;
and determining customer loss risk information according to a preset expected output sample set and an actual output result.
Because the customer's behavior information has real-time requirements, the customer's behavior information such as transfer behavior, purchasing behavior, etc. needs to utilize Kafka to asynchronously collect the message, and the information such as the time from last purchasing product, etc. of the customer's historical consumption record needs to be simply calculated according to the last purchasing time collected. The invention utilizes the quasi-real-time data to combine with the image information of the client to establish a BP neural network model, and trains the model for a plurality of times so as to output loss risk early warning thresholds of different guest groups; and early warning prompt is carried out on the clients exceeding the normal loss early warning threshold range, so that the loss risk is effectively advanced, and the maintenance cost of a client manager is reduced.
When the client loss early warning method provided by the embodiment of the invention is implemented, in one embodiment, the client image information is obtained, which comprises the following steps: customer learning data, customer age data, customer income data, customer asset data, and customer region data are acquired as customer portrait information.
When the client loss early warning method provided by the embodiment of the invention is implemented, in one embodiment, the client behavior information is obtained, which comprises the following steps: and acquiring login day data, consumption number data, transfer amount data and the time length data of the last purchased product as customer behavior information.
Customer group portrayal attributes such as: customer academia, customer age, customer income, customer assets, customer territories, etc., different portrayal groups may cause differences in the behavior of the customers. Different customer behaviors, such as: the number of login days, the number of consumption, the amount of transfer, the time length from the last purchase of the product, and the like, and the loss risk of the customers under different customer groups can be reflected. Many indicators of banking projects also reflect the health of the project itself and other potential risks or problems. In view of the above, the invention provides a customer loss early warning scheme based on a BP neural network model, which collects customer portrait information, acquires customer behavior attributes in real time, and can improve learning and memory functions of the neural network by utilizing the connection rights of multiple hidden layers of the BP neural network model, thus being one of the most widely used neural network models at present. Therefore, the BP network is used for customer loss early warning, the impending customer loss is reminded to save the customer, the customer loss is reduced, and the purpose of better service for the customer is achieved.
Fig. 2 is a schematic diagram of a process of establishing a BP neural network model of a client churn early warning method according to an embodiment of the present invention. When the client churn early warning method provided by the embodiment of the present invention is implemented as shown in fig. 2, in one embodiment, the establishing a BP neural network model includes:
step 201: presetting an input node information item and an output node information item, and setting the number of hidden layers;
step 202: and establishing a BP neural network model according to the preset input node information item, the preset output node information item and the preset hidden layer number.
In an embodiment, the BP neural network training is prepared in advance, and a data set is collected, and image information (such as a client school, a client age, a client income, a client asset and a client region) of a client and behavior information (such as a login day, a consumption number, a transfer amount and a time length from last purchase of a product) of the client are used as BP neural network training input; establishing a project risk early warning BP neural network model, and determining the proper hidden layer number in the model; and outputting loss early warning threshold conditions of different client groups through a multi-time training model, and prompting clients to loss risk early warning for the super threshold indexes. The most core content of the method is to determine the proper hidden layer number of the BP network, and write codes for establishing the BP network model, which is the most complex and difficult place in the text.
Fig. 3 is a schematic diagram of a training process of a BP neural network model of a client loss early-warning method according to an embodiment of the present invention, and as shown in fig. 3, when the client loss early-warning method provided by the embodiment of the present invention is implemented, in one embodiment, the BP neural network model is trained according to client portrait information and client behavior information, so as to obtain a client loss early-warning model, and determining an actual output result includes:
step 301: inputting customer portrait information and customer behavior information into a BP neural network model for training;
step 302: and adjusting the weight coefficient among the hidden layers for a plurality of times in the training process until the weight coefficient reaches a preset target, stopping training, taking the BP neural network model which is currently trained as a customer loss early warning model, and determining an actual output result.
Fig. 4 is a schematic diagram of a training process performed by a BP neural network model of a client loss early-warning method according to an embodiment of the present invention, as shown in fig. 4, when the client loss early-warning method provided by the embodiment of the present invention is implemented, in an embodiment, determining client loss risk information according to a preset expected output sample set and an actual output result includes:
step 401: determining loss risk thresholds of different guest groups according to a preset expected output sample set and an actual output result;
step 402: and marking the loss risk threshold clients exceeding the corresponding guest groups in the actual output results as loss risk clients, and determining the loss risk information of the clients.
When the client loss early warning method provided by the embodiment of the invention is implemented, in one embodiment, the method further comprises the following steps: and carrying out data filtering on the acquired customer portrait information and customer behavior information, filtering invalid data, and determining the customer portrait information and the customer behavior information with consistent source data.
The following describes briefly a modularized example of a customer loss early warning method according to an embodiment of the present invention in combination with a specific scenario:
1) Customer group information and customer behavior data acquisition module: collecting portrait information (such as client school, client age, client income, client asset and client region) of a client, and behavior information (such as login days, consumption times, transfer amount and the time length of last purchased product) of the client as BP neural network training input;
2) The BP neural network establishment module: determining an input node information item, selecting a proper hidden layer number, and selecting an output node information item; meanwhile, presetting a desired output sample set;
3) Training a BP network module: training the BP network by using the training data set prepared in the step 1), and adjusting the weight coefficient among layers for multiple times in the training process to obtain an actual output result;
4) Data comparison module: comparing the actual output with a preset output sample set to obtain a comparison result;
5) Risk early warning prompt module: and outputting the result of the super threshold value, and carrying out early warning prompt.
Further, the data stream includes: customer portrayal details information and behavior data set preparation: the following common key project data information is collected from a customer information system, a transaction system and a mobile banking APP, and comprises the following steps: such as customer academy, customer age, customer income, customer asset, customer region, login days, number of consumption, transfer amount, last time the product was purchased, etc.; item data set processing: processing the prepared project history data set, manually checking or filtering out invalid data by means of a program, ensuring the consistency of information of all source data, and taking the data as initial input for training a subsequent BP neural network; presetting a desired output sample set: presetting a desired output sample set according to historical data or through a mathematical statistics method, and comparing errors of actual output and sample output; establishing a BP neural network model: determining the number of hidden layers according to the input information and the expected output, and establishing a BP network model; BP network model training: the 4 steps are circularly executed, and each actual output result is obtained through training; error comparison: comparing the actual output with the expected sample, and deducing loss risk thresholds of different guest groups; project risk early warning: and for all clients exceeding the loss early warning threshold, the corresponding client manager carries out risk early warning prompt.
The client group information and the client real-time behavior data set are collected to be used as input for establishing the BP network model, and meanwhile, a desired result sample set is preset. By training the BP neural network for multiple times, errors of actual output and sample results are successively compared, so that probability of risk possibly occurring in the project implementation process is predicted, maintenance decision support is provided for a customer manager, and customer loss is timely reduced. Introducing BP neural network, which needs sample set input; therefore, the customer group information and the customer real-time behavior data set are collected and effectively screened to serve as training initial input of the BP network.
Fig. 5 is a schematic diagram of a computer device for running a client loss early warning method according to the present invention, and as shown in fig. 5, an embodiment of the present invention further provides a computer device 500, including a memory 510, a processor 520, and a computer program 530 stored in the memory and capable of running on the processor, where the processor implements the client loss early warning method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program realizes the client loss early warning method when being executed by a processor.
The embodiment of the invention also provides a computer program product, which comprises a computer program, wherein the computer program realizes the client loss early warning method when being executed by a processor.
The embodiment of the invention also provides a customer loss early warning device, which is described in the following embodiment. Because the principle of the device for solving the problems is similar to that of a customer loss early warning method, the implementation of the device can refer to the implementation of a customer loss early warning method, and the repetition is omitted.
Fig. 6 is a schematic diagram of a customer loss early-warning device according to an embodiment of the present invention, and as shown in fig. 6, the embodiment of the present invention further provides a customer loss early-warning device.
When the customer loss early warning device provided by the embodiment of the invention is implemented, in one embodiment, the device comprises:
an information acquisition module 601, configured to acquire customer portrait information and customer behavior information;
the BP neural network model building module 602 is configured to build a BP neural network model, and preset a desired output sample set;
the model training module 603 is configured to train the BP neural network model according to the customer portrait information and the customer behavior information, obtain a customer loss early warning model, and determine an actual output result;
the customer loss risk information determining module 604 is configured to determine customer loss risk information according to a preset expected output sample set and an actual output result.
When the customer loss early warning device provided by the embodiment of the invention is implemented, in one embodiment, the information acquisition module is specifically configured to: customer learning data, customer age data, customer income data, customer asset data, and customer region data are acquired as customer portrait information.
When the customer loss early warning device provided by the embodiment of the invention is implemented, in one embodiment, the information acquisition module is further used for: and acquiring login day data, consumption number data, transfer amount data and the time length data of the last purchased product as customer behavior information.
When the client loss early warning device provided by the embodiment of the invention is implemented, in one embodiment, the BP neural network model building module is specifically used for: presetting an input node information item and an output node information item, and setting the number of hidden layers;
and establishing a BP neural network model according to the preset input node information item, the preset output node information item and the preset hidden layer number.
When the customer loss early warning device provided by the embodiment of the invention is implemented, in one embodiment, the model training module is specifically used for:
inputting customer portrait information and customer behavior information into a BP neural network model for training;
and adjusting the weight coefficient among the hidden layers for a plurality of times in the training process until the weight coefficient reaches a preset target, stopping training, taking the BP neural network model which is currently trained as a customer loss early warning model, and determining an actual output result.
When the customer loss early warning device provided by the embodiment of the invention is implemented, in one embodiment, the customer loss risk information determining module is specifically configured to:
determining loss risk thresholds of different guest groups according to a preset expected output sample set and an actual output result;
and marking the loss risk threshold clients exceeding the corresponding guest groups in the actual output results as loss risk clients, and determining the loss risk information of the clients.
When the customer loss early warning device provided by the embodiment of the invention is implemented, in one embodiment, the information acquisition module is further used for: and carrying out data filtering on the acquired customer portrait information and customer behavior information, filtering invalid data, and determining the customer portrait information and the customer behavior information with consistent source data.
In summary, the method and the device for customer churn early warning provided by the embodiment of the invention comprise the following steps: acquiring customer portrait information and customer behavior information; establishing a BP neural network model, and presetting a desired output sample set; training the BP neural network model according to the customer portrait information and the customer behavior information to obtain a customer loss early warning model, and determining an actual output result; and determining customer loss risk information according to a preset expected output sample set and an actual output result. A customer loss early warning method based on decision trees and BP neural networks aims at analyzing business behaviors of customers, such as transfer behaviors, payment behaviors, login intervals, asset changes and the like, through massive data of customer group portraits, establishing a customer loss model, predicting the risk rate of customer loss and timely notifying a customer manager of carrying out customer saving work. Establishing a BP neural network by collecting customer group information and a customer real-time behavior data set, and outputting a customer loss threshold index under multiple customer groups by training the BP network for multiple times; therefore, potential loss risk hidden danger of the client in the service is predicted, so that a client manager can reduce loss of the client in advance and take countermeasures, and the occurrence probability of loss is reduced.
According to the technical scheme, the data acquisition, storage, use, processing and the like all meet the relevant regulations of national laws and regulations, and various types of data such as personal identity data, operation data, behavior data and the like related to individuals, clients, crowds and the like acquired by the method are authorized.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (17)
1. The customer churn early warning method is characterized by comprising the following steps:
acquiring customer portrait information and customer behavior information;
establishing a BP neural network model, and presetting a desired output sample set;
training the BP neural network model according to the customer portrait information and the customer behavior information to obtain a customer loss early warning model, and determining an actual output result;
and determining customer loss risk information according to a preset expected output sample set and an actual output result.
2. The method of claim 1, wherein obtaining customer representation information comprises: customer learning data, customer age data, customer income data, customer asset data, and customer region data are acquired as customer portrait information.
3. The method of claim 1, wherein obtaining customer behavior information comprises: and acquiring login day data, consumption number data, transfer amount data and the time length data of the last purchased product as customer behavior information.
4. The method of claim 1, wherein establishing a BP neural network model comprises: presetting an input node information item and an output node information item, and setting the number of hidden layers;
and establishing a BP neural network model according to the preset input node information item, the preset output node information item and the preset hidden layer number.
5. The method of claim 1, wherein training the BP neural network model based on the customer representation information and the customer behavior information to obtain a customer churn early warning model, determining the actual output result comprises:
inputting customer portrait information and customer behavior information into a BP neural network model for training;
and adjusting the weight coefficient among the hidden layers for a plurality of times in the training process until the weight coefficient reaches a preset target, stopping training, taking the BP neural network model which is currently trained as a customer loss early warning model, and determining an actual output result.
6. The method of claim 1, wherein determining customer churn risk information based on the preset desired output sample set and the actual output result comprises:
determining loss risk thresholds of different guest groups according to a preset expected output sample set and an actual output result;
and marking the loss risk threshold clients exceeding the corresponding guest groups in the actual output results as loss risk clients, and determining the loss risk information of the clients.
7. The method as recited in claim 1, further comprising: and carrying out data filtering on the acquired customer portrait information and customer behavior information, filtering invalid data, and determining the customer portrait information and the customer behavior information with consistent source data.
8. A customer churn warning device, comprising:
the information acquisition module is used for acquiring customer portrait information and customer behavior information;
the BP neural network model building module is used for building a BP neural network model and presetting a desired output sample set;
the model training module is used for training the BP neural network model according to the customer portrait information and the customer behavior information to obtain a customer loss early warning model and determining an actual output result;
and the client loss risk information determining module is used for determining client loss risk information according to a preset expected output sample set and an actual output result.
9. The apparatus of claim 8, wherein the information acquisition module is specifically configured to: customer learning data, customer age data, customer income data, customer asset data, and customer region data are acquired as customer portrait information.
10. The apparatus of claim 8, wherein the information acquisition module is further to: and acquiring login day data, consumption number data, transfer amount data and the time length data of the last purchased product as customer behavior information.
11. The apparatus of claim 8, wherein the BP neural network model building module is specifically configured to: presetting an input node information item and an output node information item, and setting the number of hidden layers;
and establishing a BP neural network model according to the preset input node information item, the preset output node information item and the preset hidden layer number.
12. The apparatus of claim 8, wherein the model training module is configured to:
inputting customer portrait information and customer behavior information into a BP neural network model for training;
and adjusting the weight coefficient among the hidden layers for a plurality of times in the training process until the weight coefficient reaches a preset target, stopping training, taking the BP neural network model which is currently trained as a customer loss early warning model, and determining an actual output result.
13. The apparatus of claim 8, wherein the customer churn risk information determination module is specifically configured to:
determining loss risk thresholds of different guest groups according to a preset expected output sample set and an actual output result;
and marking the loss risk threshold clients exceeding the corresponding guest groups in the actual output results as loss risk clients, and determining the loss risk information of the clients.
14. The apparatus of claim 8, wherein the information acquisition module is further to: and carrying out data filtering on the acquired customer portrait information and customer behavior information, filtering invalid data, and determining the customer portrait information and the customer behavior information with consistent source data.
15. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 7 when executing the computer program.
16. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, implements the method of any of claims 1 to 7.
17. A computer program product, characterized in that the computer program product comprises a computer program which, when executed by a processor, implements the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211500041.8A CN116091185A (en) | 2022-11-28 | 2022-11-28 | Customer loss early warning method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211500041.8A CN116091185A (en) | 2022-11-28 | 2022-11-28 | Customer loss early warning method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116091185A true CN116091185A (en) | 2023-05-09 |
Family
ID=86199966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211500041.8A Pending CN116091185A (en) | 2022-11-28 | 2022-11-28 | Customer loss early warning method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116091185A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116883070A (en) * | 2023-09-05 | 2023-10-13 | 上海银行股份有限公司 | Bank generation payroll customer loss early warning method |
| CN117422181A (en) * | 2023-12-15 | 2024-01-19 | 湖南三湘银行股份有限公司 | Fuzzy label-based method and system for early warning loss of issuing clients |
-
2022
- 2022-11-28 CN CN202211500041.8A patent/CN116091185A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116883070A (en) * | 2023-09-05 | 2023-10-13 | 上海银行股份有限公司 | Bank generation payroll customer loss early warning method |
| CN117422181A (en) * | 2023-12-15 | 2024-01-19 | 湖南三湘银行股份有限公司 | Fuzzy label-based method and system for early warning loss of issuing clients |
| CN117422181B (en) * | 2023-12-15 | 2024-04-02 | 湖南三湘银行股份有限公司 | Fuzzy label-based method and system for early warning loss of issuing clients |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116091185A (en) | Customer loss early warning method and device | |
| US11190535B2 (en) | Methods and systems for inferring behavior and vulnerabilities from process models | |
| CN106886481B (en) | Static analysis and prediction method and device for system health degree | |
| CN111967910A (en) | User passenger group classification method and device | |
| US20180276584A1 (en) | Facilitating organizational management using bug data | |
| US9799007B2 (en) | Method of collaborative software development | |
| CN111882420B (en) | Response rate generation method, marketing method, model training method and device | |
| CN112598443A (en) | Online channel business data processing method and system based on deep learning | |
| CN118333576B (en) | Business process management system and method based on closed-loop cooperation | |
| CN111986027A (en) | Abnormal transaction processing method and device based on artificial intelligence | |
| CN112308623A (en) | High-quality client loss prediction method and device based on supervised learning and storage medium | |
| CN112632179A (en) | Model construction method and device, storage medium and equipment | |
| CN118313643B (en) | Whole-process production monitoring and management method and system for granulized fruit beverage | |
| CN115730947A (en) | Bank customer loss prediction method and device | |
| CN115938600A (en) | Mental health state prediction method and system based on correlation analysis | |
| CN115185804A (en) | Server performance prediction method, system, terminal and storage medium | |
| CN113609393B (en) | Digital platform based on data service and data management | |
| CN113744890A (en) | Reworking and production-resuming analysis method, system and storage medium | |
| CN111882113B (en) | Enterprise mobile banking user prediction method and device | |
| AU2021204470A1 (en) | Benefit surrender prediction | |
| CN110413482B (en) | Detection method and device | |
| CN113824580A (en) | Network index early warning method and system | |
| CN116523244A (en) | Testing manpower risk early warning method based on outsourcing resources | |
| CN112232960B (en) | Transaction application system monitoring method and device | |
| d'Astous et al. | Empirical study of exchange patterns during software peer review meetings |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |