CN115439221A - Bank customer data analysis method and device - Google Patents
Bank customer data analysis method and device Download PDFInfo
- Publication number
- CN115439221A CN115439221A CN202211114719.9A CN202211114719A CN115439221A CN 115439221 A CN115439221 A CN 115439221A CN 202211114719 A CN202211114719 A CN 202211114719A CN 115439221 A CN115439221 A CN 115439221A
- Authority
- CN
- China
- Prior art keywords
- client
- loan
- deposit
- data
- historical
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a bank customer data analysis method and a device, relating to the technical field of artificial intelligence, wherein the method comprises the following steps: under the condition of user authorization, acquiring personal attribute data and consumption data of a plurality of deposit clients and loan clients; analyzing personal attribute data and consumption data of a plurality of deposit clients and loan clients to obtain client characteristics of each deposit client and each loan client; generating client portrait data of each deposit client and each loan client according to the client characteristics of each deposit client and each loan client; inputting the client portrait data of each deposit client and each loan client and the service correlation data between the loan services into a multilayer cascade integration model to obtain the probability of converting each deposit client into a loan client; potential loan clients are determined from the plurality of deposit clients based on the probability of each deposit client converting to a loan client. The method can quickly and accurately determine the potential loan clients and improve the conversion rate of the loan storage users.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a bank customer data analysis method and device. It should be noted that the bank customer data analysis method and apparatus of the present invention can be used in the technical field of artificial intelligence, and can also be used in any field except the technical field of artificial intelligence.
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, the analysis of bank customer data is inferred by bank business personnel according to experience, the problems of low efficiency and low accuracy of analysis results exist, for example, the loan business is the core business of a bank, and how to accurately convert a deposit user into a loan user is a problem which is not easy to solve, and the conventional loan user acquisition mainly depends on popularization of business personnel or purchase of a vehicle room by the user, so that the loan user conversion cannot be accurately realized.
Disclosure of Invention
The embodiment of the invention provides a bank customer data analysis method, which is used for improving the efficiency and the accuracy of bank customer data analysis, accurately determining potential loan customers and conveniently and quickly converting deposit users into loan users, and comprises the following steps:
acquiring personal attribute data and consumption data of a plurality of deposit customers and loan customers under the authorization of a user;
analyzing personal attribute data and consumption data of a plurality of deposit clients and loan clients to obtain client characteristics of each deposit client and each loan client;
generating client portrait data of each deposit client and each loan client according to the client characteristics of each deposit client and each loan client;
inputting the client image data of each deposit client and each loan client and the service correlation data among the loan deposit services into a multi-layer cascade integration model to obtain the probability of converting each deposit client into a loan client; the multilayer cascade integration model is a machine learning model, comprises a plurality of groups of parallel feature extraction networks and is obtained by training the customer image data of historical deposit customers and historical loan customers, the business association data among historical loan-saving businesses and the probability of converting the historical deposit customers into the historical loan customers;
potential loan clients are determined from the plurality of deposit clients based on the probability of each deposit client converting to a loan client.
The embodiment of the invention also provides a bank customer data analysis device, which is used for improving the efficiency and the accuracy of bank customer data analysis, accurately determining potential loan customers and conveniently and quickly converting deposit users into loan users, and comprises:
the data acquisition module is used for acquiring personal attribute data and consumption data of a plurality of deposit customers and loan customers under the authorization of a user;
the client portrait generation module is used for analyzing personal attribute data and consumption data of a plurality of deposit clients and loan clients to obtain client characteristics of each deposit client and each loan client; generating client portrait data of each deposit client and each loan client according to the client characteristics of each deposit client and each loan client;
the potential loan client determining module is used for inputting the client image data of each deposit client and each loan client and the service correlation data among the loan deposit services into the multi-layer cascade integration model to obtain the probability of converting each deposit client into a loan client; the multilayer cascade integration model is a machine learning model and comprises a plurality of groups of parallel feature extraction networks, and is obtained by training the customer image data of the historical deposit customers and the historical loan customers, the business association data between the historical loan stores and businesses and the probability of converting the historical deposit customers into the historical loan customers; potential loan clients are determined from a plurality of deposit clients based on the probability of each deposit client converting to a loan client.
The embodiment of the invention also provides computer equipment which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor realizes the bank customer data analysis method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, wherein a computer program is stored in the computer readable storage medium, and when the computer program is executed by a processor, the bank customer data analysis method is realized.
An embodiment of the present invention further provides a computer program product, where the computer program product includes a computer program, and when the computer program is executed by a processor, the method for analyzing bank customer data is implemented.
In the embodiment of the invention, personal attribute data and consumption data of a plurality of deposit customers and loan customers are obtained under the condition of user authorization; analyzing personal attribute data and consumption data of a plurality of deposit clients and loan clients to obtain client characteristics of each deposit client and each loan client; generating client portrait data of each deposit client and each loan client according to the client characteristics of each deposit client and each loan client; inputting the client image data of each deposit client and each loan client and the service correlation data among the loan deposit services into a multi-layer cascade integration model to obtain the probability of converting each deposit client into a loan client; the multilayer cascade integration model is a machine learning model and comprises a plurality of groups of parallel feature extraction networks, and is obtained by training the customer image data of the historical deposit customers and the historical loan customers, the business association data between the historical loan stores and businesses and the probability of converting the historical deposit customers into the historical loan customers; potential loan clients are determined from the plurality of deposit clients based on the probability of each deposit client converting to a loan client. In the embodiment of the invention, personal attribute data and consumption data of a plurality of deposit clients and loan clients are used as basic data to generate client portrait data, and business association data between the client portrait data and the loan saving business is input into a multilayer cascade integration model to obtain the probability of converting each deposit client into a loan client, so that potential loan clients are determined from the plurality of deposit clients; the multi-layer cascade integration model comprises a plurality of groups of parallel machine learning models of the feature extraction network, data features can be fully mined and processed, and the accuracy of the probability of converting output deposit clients into loan clients is improved, so that potential loan clients in the deposit clients are accurately determined, targeted marketing can be performed on the potential loan clients, deposit users are quickly and accurately converted into loan users, and the conversion rate of the deposit users and the loan users is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:
FIG. 1 is a schematic flow chart of a method for analyzing bank customer data according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating an embodiment of a method for analyzing bank customer data according to the present invention;
FIG. 3 is a schematic diagram of a bank customer data analysis device according to an embodiment of the present invention;
FIG. 4 is a diagram of an embodiment of a bank customer data analysis device according to the present invention;
FIG. 5 is a diagram of a computer device in an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
The inventor finds that, at present, the analysis of bank customer data is concluded by bank business personnel according to experience, and the problems of low efficiency and low accuracy of analysis results exist, for example, the loan business is the core business of a bank, and how to accurately convert a deposit user into a loan user is a problem which is not easy to solve, and the conventional loan user acquisition mainly depends on popularization of business personnel or the user purchases a vehicle room by himself, so that the loan user conversion cannot be accurately realized.
Fig. 1 is a schematic flow chart of a method for analyzing bank customer data according to an embodiment of the present invention, as shown in fig. 1, the method includes:
102, analyzing personal attribute data and consumption data of a plurality of deposit clients and loan clients to obtain client characteristics of each deposit client and each loan client;
and step 105, determining potential loan clients from a plurality of deposit clients according to the probability that each deposit client is converted into a loan client.
As can be seen from the flow shown in fig. 1, in the embodiment of the present invention, personal attribute data and consumption data of a plurality of deposit clients and loan clients are used as basic data to generate client portrait data, and business association data between the client portrait data and loan saving business is input into a multi-layer cascade integration model to obtain a probability that each deposit client is converted into a loan client, so as to determine potential loan clients from the plurality of deposit clients; the multi-layer cascade integration model comprises a plurality of groups of parallel machine learning models of the feature extraction network, data features can be fully mined and processed, and the accuracy of the probability of converting output deposit clients into loan clients is improved, so that potential loan clients in the deposit clients are accurately determined, targeted marketing can be performed on the potential loan clients, the deposit clients are quickly and accurately converted into loan clients, and the conversion rate of the loan clients is improved.
It should be noted that, in the technical solution of the present application, the acquisition, storage, use, processing, etc. of data all conform to the relevant regulations of the national laws and regulations.
Each step is explained in detail below.
In step 101, personal attribute data and consumption data of a plurality of deposit and loan customers are obtained with the authorization of a user.
For example, the personal attribute data and consumption data of a plurality of deposit customers and loan customers can be obtained from a certain marketing activity under the authorization of a user, or the personal attribute data and consumption data of a plurality of deposit customers and loan customers can be obtained based on a big data platform; the personal attribute data includes, but is not limited to, information such as name, gender, academic calendar, occupation, income, deposit, movable property and real property condition, and the consumption data includes, but is not limited to, information such as financial record, consumption preference, group buying activity, community activity, forwarding and sharing.
Fig. 2 is a specific embodiment of a bank customer data analysis method according to an embodiment of the present invention, and as shown in fig. 2, in the flow diagram of the bank customer data analysis method shown in fig. 1, before analyzing personal attribute data and consumption data of a plurality of deposit customers and loan customers, the method further includes:
In implementation, the personal attribute data and consumption data of a plurality of deposit clients and loan clients are preprocessed, for example, the operations of screening, cleaning, processing, field arrangement, duplication removal, abnormal data elimination, flushing and the like are performed on the consumption data such as the types and the amounts of purchased products, the personal attribute data such as the occupation, the academic calendar and the deposit of the clients, and the like.
In step 102, analyzing personal attribute data and consumption data of a plurality of deposit clients and loan clients to obtain client characteristics of each deposit client and each loan client; in step 103, client figure data of each of the deposit client and the loan client is generated based on the client characteristics of each of the deposit client and the loan client.
During implementation, analysis operations such as preprocessing and statistics are carried out on personal attribute data and consumption data of deposit customers and loan customers, customer characteristics of each deposit customer and loan customer are obtained, for example, attribute characteristics, consumption preference characteristics and behavior characteristics of each deposit customer and loan customer, a label system of customer figures is designed, such as attribute labels, behavior labels, group labels and prediction labels, and customer figure data of each deposit customer and loan customer are generated.
In one embodiment, before inputting the client portrait data of each deposit client and loan client, and the business association data between the loan businesses into the multi-layer cascading integration model, the method further comprises:
and carrying out correlation analysis on the client characteristics of each deposit client and each loan client to obtain service correlation data between the deposit and loan services.
For example, based on attribute characteristics, consumption preference characteristics, behavior characteristics and the like of each deposit client and loan client, a term set is constructed, such as { fractional loan, luxury goods, credit card }, { house, vehicle and large loan }, service association data such as confidence degree, support degree and the like of each term set or each term set are calculated by adopting an association rule mining Apriori algorithm, and finally service association data among loan-in services are obtained through arrangement.
In one embodiment, the correlation analysis is performed on the client characteristics of each deposit client and loan client to obtain the service correlation data between the loan services, and the method comprises the following steps:
analyzing the correlation between the loan-saving service of each deposit client and loan client and the marketing campaign according to the frequency and/or activity of each deposit client and loan client in participating in the marketing campaign and the state of the loan-saving service;
and acquiring service correlation data among the loan-saving services according to the correlation.
In implementation, all available data of each deposit client and loan client are fully considered, especially the frequency and/or activity of each deposit client and loan client participating in marketing activities and the loan saving business state, such as the number of times of participating in marketing activities of products with larger amount, sharing, approval and collection conditions, loan repayment conditions are taken into consideration, the association between the loan saving business and the marketing activities of each deposit client and loan client is analyzed, for example, users who often participate in automobile exhibition, automobile 4S shop marketing activities, browse automobile product webpages and have a small amount of deposits may loan, and business association data between the loan saving businesses is obtained according to the association between the loan saving businesses and the marketing activities.
Executing to step 104, inputting the client portrait data of each deposit client and loan client and the service correlation data between the loan deposit services into a multi-layer cascade integration model to obtain the probability of converting each deposit client into a loan client; for example, the percentage of probability that a deposit client translates into a loan client is output;
the multi-layer cascade integration model is a machine learning model and comprises a plurality of groups of parallel feature extraction networks, and the multi-layer cascade integration model is obtained by training the customer image data of the historical deposit customers and the historical loan customers, the business association data between the historical loan storage businesses and the probability of converting the historical deposit customers into the historical loan customers.
In the implementation process, a plurality of groups of parallel feature extraction networks in the multilayer cascade integration model can be used for carrying out depth feature coding on input data, depth feature information is fully utilized in the multilayer cascade integration model, different integration features are learned, the relationship among the integration features is mined, and the probability that each deposit client is converted into a loan client is obtained.
In one embodiment, in the multilayer cascade integration model, different kinds of feature extraction networks are respectively used as different network model groups; or, the same kind of feature extraction network is respectively used as different network model groups after adopting different training modes; or, the same kind of feature extraction network is respectively used as different network model groups after adopting different data sampling modes.
During implementation, firstly, a plurality of feature extraction networks are constructed, including but not limited to feature extraction networks such as a convolutional neural network CNN, a convolutional neural network AlexNet, a convolutional neural network VGG Net, a deep neural network GoogleNet, a deep residual error network ResNet, a deep residual error network ResNeXt, a deep neural network YOLO, a lightweight neural network SqueezeNet, a deep neural network SegNet, an antagonistic neural network GAN and the like; then, grouping the extracted networks based on a plurality of characteristics; different types of feature extraction networks can be used as a network model group, and the feature extraction networks in the same group have similar functions and can extract similar features; or the same kind of feature extraction network can be used as a network model group according to a training mode; or the same kind of feature extraction network can be used as a network model group according to the data sampling mode; for example, there may be multiple groups in the multi-layer cascading integration model, with multiple CNN feature extraction networks within each group. Therefore, the advantages of each feature extraction network can be better utilized, data feature information can be deeply mined through the calculation of the multiple groups of feature extraction networks, result prediction is better carried out, and the accuracy of the probability that output deposit customers are converted into loan customers is improved.
In one embodiment, the different kinds of feature extraction networks are respectively used as different network model groups, and include:
extracting the features of the ResNet category to form a ResNet series network model group;
the feature extraction network of the GoogleNet category is used as a GoogleNet series network model group;
and (3) taking the feature extraction network of the EfficientNet category as an EfficientNet series network model group.
In one embodiment, the feature extraction networks of the same kind are respectively used as different network model groups after adopting different training modes, and the method includes:
the same kind of feature extraction network respectively serves as: the Xception network model group and the DenseNet network model group.
In one embodiment, the multi-layer cascade integration model is trained and tested as follows:
acquiring client image data of a historical deposit client and a historical loan client, business association data among historical loan storage businesses and the probability of converting the historical deposit client into the historical loan client, and constructing a training set and a testing set;
training the machine learning model by using a training set to obtain a multilayer cascade integration model;
and testing the multilayer cascade integration model by using the test set.
For example, obtaining client image data of a historical deposit client and a historical loan client, business association data between historical loan deposit businesses and probability of converting the historical deposit client into the historical loan client, and using the business association data as sample data to construct a training set and a test set; training the machine learning model by using a training set to obtain a multilayer cascade integration model; and testing the multilayer cascade integration model by using the test set. And then, sample data, a training set and a test set can be expanded periodically, and iterative training is carried out on the multilayer cascade integration model.
Finally, in step 105, potential loan clients are determined from the plurality of deposit clients based on the probability of each deposit client converting to a loan client.
In implementation, a probability threshold value can be preset, after the probability of each deposit client converted into a loan client is obtained, the deposit client corresponding to the probability exceeding the threshold value is determined as a potential loan client, and a potential loan client list can be directly output, so that the potential loan client is accurately marketed, for example, a proper loan product is recommended to the client, and the deposit client is converted into the loan client.
In summary, in the embodiment of the present invention, first, personal attribute data and consumption data of a plurality of deposit customers and loan customers are obtained; then, analyzing based on the acquired data to generate client portrait data and business associated data; finally, according to the customer portrait data and the business associated data, the probability of converting the deposit customers into loan customers is obtained by using a multilayer cascade integration model, and potential loan customers are determined from a plurality of deposit customers; therefore, accurate marketing is carried out, the deposit users are quickly converted into loan users, and the conversion rate of the deposit users is improved.
The embodiment of the invention also provides a bank customer data analysis device, which is described in the following embodiment. Because the principle of the device for solving the problems is similar to the bank customer data analysis method, the implementation of the device can refer to the implementation of the method, and repeated details are not repeated.
Fig. 3 is a schematic diagram of a bank customer data analysis device in an embodiment of the present invention, as shown in fig. 3, the device includes:
a data acquisition module 301, configured to acquire personal attribute data and consumption data of a plurality of deposit customers and loan customers under the authorization of a user;
a client figure generation module 302, for analyzing the personal attribute data and consumption data of a plurality of deposit clients and loan clients, and obtaining the client characteristics of each deposit client and loan client; generating client portrait data of each deposit client and each loan client according to the client characteristics of each deposit client and each loan client;
the potential loan client determining module 303 is used for inputting the client portrait data of each deposit client and each loan client and the service correlation data among the loan saving services into the multi-layer cascade integration model to obtain the probability of converting each deposit client into a loan client; the multilayer cascade integration model is a machine learning model and comprises a plurality of groups of parallel feature extraction networks, and is obtained by training the customer image data of the historical deposit customers and the historical loan customers, the business association data between the historical loan stores and businesses and the probability of converting the historical deposit customers into the historical loan customers; potential loan clients are determined from a plurality of deposit clients based on the probability of each deposit client converting to a loan client.
Fig. 4 is a specific embodiment of a bank customer data analysis apparatus according to an embodiment of the present invention, and as shown in fig. 4, the apparatus shown in fig. 3 further includes:
and the data preprocessing module 401 is configured to preprocess the personal attribute data and consumption data of the plurality of deposit customers and loan customers before the client representation generation module 302 analyzes the personal attribute data and consumption data of the plurality of deposit customers and loan customers, where the preprocessing includes one or any combination of flushing, screening, field warping, deduplication and abnormal data elimination.
In one embodiment, the bank customer data analysis device further comprises:
and the business association analysis module is used for performing correlation analysis on the client characteristics of each deposit client and each loan client to obtain business association data between loan-saving businesses before the potential loan client determination module 303 inputs the client image data of each deposit client and each loan client and the business association data between loan-saving businesses into the multi-layer cascade integration model.
In an embodiment, the service association analysis module is specifically configured to:
analyzing the correlation between the loan reservation service and the marketing campaign of each deposit client and loan client according to the frequency and/or activity of each deposit client and loan client participating in the marketing campaign and the state of the loan reservation service;
and acquiring service correlation data among the loan-saving services according to the correlation.
In one embodiment, in the multilayer cascade integration model, different kinds of feature extraction networks are respectively used as different network model groups; or, the same kind of feature extraction network is respectively used as different network model groups after adopting different training modes; or, the same kind of feature extraction network is respectively used as different network model groups after adopting different data sampling modes.
In one embodiment, the different kinds of feature extraction networks are respectively used as different network model groups, and the method comprises the following steps:
extracting the features of the ResNet category to form a ResNet series network model group;
the characteristic extraction network of the GoogleNet category is used as a GoogleNet series network model group;
and taking the feature extraction network of the EfficientNet category as an EfficientNet series network model group.
In one embodiment, the feature extraction networks of the same kind are respectively used as different network model groups after adopting different training modes, and the method includes:
the same kind of feature extraction network is respectively used as: the Xception network model group and the DenseNet network model group.
In one embodiment, the multi-layer cascade integration model is trained and tested as follows:
acquiring client image data of a historical deposit client and a historical loan client, business association data among historical loan storage businesses and probability of converting the historical deposit client into the historical loan client, and constructing a training set and a testing set;
training the machine learning model by using a training set to obtain a multilayer cascade integration model;
and testing the multilayer cascade integration model by using the test set.
Fig. 5 is a schematic diagram of a computer device according to an embodiment of the present invention, and as shown in fig. 5, an embodiment of the present invention further provides a computer device 500, which includes a processor 501, a memory 502, and a computer program 503 stored in the memory 502 and capable of running on the processor 501, and when the processor 501 executes the computer program 503, the bank customer data analysis method is implemented.
The embodiment of the invention also provides a computer readable storage medium, wherein a computer program is stored in the computer readable storage medium, and when the computer program is executed by a processor, the bank customer data analysis method is realized.
The embodiment of the invention also provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to realize the bank customer data analysis method.
In the embodiment of the invention, personal attribute data and consumption data of a plurality of deposit clients and loan clients are obtained; analyzing personal attribute data and consumption data of a plurality of deposit clients and loan clients to obtain client characteristics of each deposit client and each loan client; generating client portrait data of each deposit client and each loan client according to the client characteristics of each deposit client and each loan client; inputting the client portrait data of each deposit client and each loan client and the service correlation data between the loan services into a multilayer cascade integration model to obtain the probability of converting each deposit client into a loan client; the multilayer cascade integration model comprises a plurality of groups of parallel machine learning models of a feature extraction network, and is obtained by training customer image data of historical deposit customers and historical loan customers, business association data among historical loan storage businesses and probability of converting the historical deposit customers into the historical loan customers; potential loan clients are determined from a plurality of deposit clients based on the probability of each deposit client converting to a loan client. In the embodiment of the invention, personal attribute data and consumption data of a plurality of deposit clients and loan clients are used as basic data to generate client portrait data, and business association data between the client portrait data and the loan saving business is input into a multilayer cascade integration model to obtain the probability of converting each deposit client into a loan client, so that potential loan clients are determined from the plurality of deposit clients; the multi-layer cascade integration model comprises a plurality of groups of parallel machine learning models of the feature extraction network, data features can be fully mined and processed, and the accuracy of the probability of converting output deposit clients into loan clients is improved, so that potential loan clients in the deposit clients are accurately determined, targeted marketing can be performed on the potential loan clients, the deposit clients are quickly and accurately converted into loan clients, and the conversion rate of the loan clients is improved.
As will be appreciated by one skilled in the art, 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 flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams 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 above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (19)
1. A bank customer data analysis method is characterized by comprising the following steps:
acquiring personal attribute data and consumption data of a plurality of deposit customers and loan customers under the authorization of a user;
analyzing personal attribute data and consumption data of a plurality of deposit clients and loan clients to obtain client characteristics of each deposit client and each loan client;
generating client image data of each deposit client and each loan client according to the client characteristics of each deposit client and each loan client;
inputting the client image data of each deposit client and each loan client and the service correlation data among the loan deposit services into a multi-layer cascade integration model to obtain the probability of converting each deposit client into a loan client; the multilayer cascade integration model is a machine learning model and comprises a plurality of groups of parallel feature extraction networks, and is obtained by training the customer image data of the historical deposit customers and the historical loan customers, the business association data between the historical loan stores and businesses and the probability of converting the historical deposit customers into the historical loan customers;
potential loan clients are determined from the plurality of deposit clients based on the probability of each deposit client converting to a loan client.
2. The method of claim 1, wherein prior to analyzing the personal attribute data and consumption data of the plurality of depositors and borrowers, further comprising:
and preprocessing personal attribute data and consumption data of a plurality of deposit clients and loan clients, wherein the preprocessing comprises one or any combination of flushing, screening, field warping, duplication removing and abnormal data elimination.
3. The method as claimed in claim 1, wherein before inputting the client drawing data of each of the depositor and the loan client, the business association data between the depositor and the loan business into the multi-layer cascade integration model, further comprising:
and carrying out correlation analysis on the client characteristics of each deposit client and each loan client to obtain service correlation data between the deposit and loan services.
4. The method of claim 3, wherein performing a correlation analysis on the client characteristics of each of the depository and the loan client to obtain service correlation data between the loan-saving services comprises:
analyzing the correlation between the loan reservation service and the marketing campaign of each deposit client and loan client according to the frequency and/or activity of each deposit client and loan client participating in the marketing campaign and the state of the loan reservation service;
and acquiring service correlation data among the loan-saving services according to the correlation.
5. The method according to any one of claims 1 to 4, wherein in the multi-layer cascade integration model, different kinds of feature extraction networks are respectively used as different network model groups; or, the same kind of feature extraction network is respectively used as different network model groups after adopting different training modes; or, the same kind of feature extraction network is respectively used as different network model groups after adopting different data sampling modes.
6. The method of claim 5, wherein the different kinds of feature extraction networks are respectively used as different network model groups, and the method comprises the following steps:
the feature extraction network of the ResNet category is used as a ResNet series network model group;
the feature extraction network of the GoogleNet category is used as a GoogleNet series network model group;
and (3) taking the feature extraction network of the EfficientNet category as an EfficientNet series network model group.
7. The method of claim 5, wherein the feature extraction networks of the same kind are respectively used as different network model groups after adopting different training modes, and the method comprises the following steps:
the same kind of feature extraction network respectively serves as: the Xception network model group and the DenseNet network model group.
8. The method of claim 1, wherein the multi-layer cascaded integration model is trained and tested as follows:
acquiring client image data of a historical deposit client and a historical loan client, business association data among historical loan storage businesses and the probability of converting the historical deposit client into the historical loan client, and constructing a training set and a testing set;
training the machine learning model by using a training set to obtain a multilayer cascade integration model;
and testing the multilayer cascade integration model by using the test set.
9. A bank customer data analysis device, comprising:
the data acquisition module is used for acquiring personal attribute data and consumption data of a plurality of deposit customers and loan customers under the authorization of a user;
the client portrait generating module is used for analyzing personal attribute data and consumption data of a plurality of deposit clients and loan clients to obtain client characteristics of each deposit client and each loan client; generating client portrait data of each deposit client and each loan client according to the client characteristics of each deposit client and each loan client;
the potential loan client determining module is used for inputting the client portrait data of each deposit client and each loan client and the service correlation data among the loan saving services into the multi-layer cascade integration model to obtain the probability of converting each deposit client into a loan client; the multilayer cascade integration model is a machine learning model, comprises a plurality of groups of parallel feature extraction networks and is obtained by training the customer image data of historical deposit customers and historical loan customers, the business association data among historical loan-saving businesses and the probability of converting the historical deposit customers into the historical loan customers; potential loan clients are determined from the plurality of deposit clients based on the probability of each deposit client converting to a loan client.
10. The apparatus of claim 9, further comprising:
and the data preprocessing module is used for preprocessing the personal attribute data and the consumption data of the plurality of deposit customers and loan customers before the customer portrait generating module analyzes the personal attribute data and the consumption data of the plurality of deposit customers and loan customers, and the preprocessing comprises one or any combination of flushing, screening, field regularizing, duplicate removal and abnormal data elimination.
11. The apparatus of claim 9, further comprising:
and the business correlation analysis module is used for performing correlation analysis on the client characteristics of each deposit client and each loan client before the potential loan client determination module inputs the client image data of each deposit client and each loan client and the business correlation data between the deposit and loan businesses into the multi-layer cascade integration model to obtain the business correlation data between the deposit and loan businesses.
12. The apparatus of claim 11, wherein the service association analysis module is specifically configured to:
analyzing the correlation between the loan reservation service and the marketing campaign of each deposit client and loan client according to the frequency and/or activity of each deposit client and loan client participating in the marketing campaign and the state of the loan reservation service;
and acquiring service correlation data among the loan-saving services according to the correlation.
13. The apparatus according to any one of claims 9 to 12, wherein in the multi-layer cascade integration model, different kinds of feature extraction networks are respectively used as different network model groups; or, the same kind of feature extraction network is respectively used as different network model groups after adopting different training modes; or, the same kind of feature extraction network is respectively used as different network model groups after adopting different data sampling modes.
14. The apparatus of claim 13, wherein the different kinds of feature extraction networks are respectively as different network model groups, and comprise:
the feature extraction network of the ResNet category is used as a ResNet series network model group;
the characteristic extraction network of the GoogleNet category is used as a GoogleNet series network model group;
and (3) taking the feature extraction network of the EfficientNet category as an EfficientNet series network model group.
15. The apparatus of claim 13, wherein the feature extraction networks of the same kind are respectively used as different network model groups after adopting different training modes, and the method comprises:
the same kind of feature extraction network is respectively used as: the Xception network model group and the DenseNet network model group.
16. The apparatus of claim 9, wherein the multi-layer cascaded integration model is trained and tested as follows:
acquiring client image data of a historical deposit client and a historical loan client, business association data among historical loan storage businesses and the probability of converting the historical deposit client into the historical loan client, and constructing a training set and a testing set;
training the machine learning model by using a training set to obtain a multilayer cascade integration model;
and testing the multilayer cascade integration model by using the test set.
17. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any one of claims 1 to 8 when executing the computer program.
18. 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 8.
19. A computer program product, characterized in that the computer program product comprises a computer program which, when being executed by a processor, carries out the method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211114719.9A CN115439221A (en) | 2022-09-14 | 2022-09-14 | Bank customer data analysis method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211114719.9A CN115439221A (en) | 2022-09-14 | 2022-09-14 | Bank customer data analysis method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115439221A true CN115439221A (en) | 2022-12-06 |
Family
ID=84247026
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211114719.9A Pending CN115439221A (en) | 2022-09-14 | 2022-09-14 | Bank customer data analysis method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115439221A (en) |
-
2022
- 2022-09-14 CN CN202211114719.9A patent/CN115439221A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111291816B (en) | Method and device for carrying out feature processing aiming at user classification model | |
| CN110992167B (en) | Bank customer business intention recognition method and device | |
| CN111199343A (en) | Multi-model fusion tobacco market supervision abnormal data mining method | |
| CN109583966B (en) | High-value customer identification method, system, equipment and storage medium | |
| CN112070615B (en) | Financial product recommendation method and device based on knowledge graph | |
| Sadikin et al. | Comparative study of classification method on customer candidate data to predict its potential risk | |
| CN109635010B (en) | User characteristic and characteristic factor extraction and query method and system | |
| CN114612251A (en) | Risk assessment method, device, equipment and storage medium | |
| CN109711424A (en) | A kind of rule of conduct acquisition methods, device and equipment based on decision tree | |
| CN109977977B (en) | Method for identifying potential user and corresponding device | |
| Bouzidi et al. | Deep learning-based automated learning environment using smart data to improve corporate marketing, business strategies, fraud detection in financial services, and financial time series forecasting | |
| CN115545886A (en) | Overdue risk identification method, overdue risk identification device, overdue risk identification equipment and storage medium | |
| CN115730947A (en) | Bank customer loss prediction method and device | |
| CN115063035A (en) | Customer evaluation method, system, equipment and storage medium based on neural network | |
| CN117611335A (en) | Financial risk identification method, apparatus, electronic device and storage medium | |
| Bier et al. | Variable-length multivariate time series classification using ROCKET: a case study of incident detection | |
| US20210073247A1 (en) | System and method for machine learning architecture for interdependence detection | |
| CN110457329B (en) | Method and device for realizing personalized recommendation | |
| CN111951099A (en) | Credit card issuing model and application method thereof | |
| CN115439221A (en) | Bank customer data analysis method and device | |
| CN114626940A (en) | Data analysis method and device and electronic equipment | |
| CN114756685A (en) | Complaint risk identification method and device for complaint sheet | |
| CN114202428A (en) | Asset data processing method, device and equipment based on graph convolution neural network | |
| CN113065892A (en) | Information pushing method, device, equipment and storage medium | |
| CN112950392A (en) | Information display method, posterior information determination method and device and related equipment |
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 |