CN113869423A - Marketing response model construction method, equipment and medium - Google Patents

Abstract

The application discloses a marketing response model construction method, equipment and a medium, wherein the method comprises the following steps: determining enterprise data of a sample enterprise which is acquired in advance, and establishing a standard database according to the enterprise data of the sample enterprise; determining enterprise indexes of the sample enterprises according to the standard database; establishing a multi-dimensional clustering analysis model of a sample enterprise through a K-means clustering algorithm according to enterprise indexes; determining clustering clusters of sample enterprises under different index dimensions according to a multi-dimensional clustering analysis model; and determining the marketing response probability of the client enterprises through a K nearest neighbor algorithm according to the clustering clusters respectively corresponding to the sample enterprises under different index dimensions. The unsupervised machine learning model adopts characteristic indexes of enterprises to form categories, and the problem of positioning target customers under the condition of no expression sample is solved, so that the marketing cost is reduced, and the marketing investment return rate is improved.

Description

Marketing response model construction method, equipment and medium

Technical Field

The application relates to the field of marketing response, in particular to a marketing response model construction method, equipment and medium.

Background

The application of financial science and technology in the credit finance field is more and more extensive, advanced artificial intelligence, machine learning and big data technology are used for risk assessment of enterprise credit, and a financial wind control system which runs through the whole credit process of marketing, pre-credit, mid-credit and post-credit is a business target for developing credit investigation service.

In the financial credit field, the main customers faced by bank credit products are small and micro enterprises, the conditions of small and micro enterprise objects served by different credit products are different, how to accurately divide the small and micro enterprise object, different bank credit products are recommended for different groups, enterprise customers with greater interest in the bank credit products and response probability are identified by using a model, and the method is one of important contents of financial science and technology assisted small and micro enterprise credit services.

Therefore, a need exists for a customer response model that accurately reflects the probability that a customer business is interested in a bank credit product.

Disclosure of Invention

In order to solve the above problems, the present application proposes a method, an apparatus, and a medium for constructing a marketing response model, wherein the method includes: determining enterprise data of a sample enterprise which is acquired in advance, and establishing a standard database according to the enterprise data of the sample enterprise; determining enterprise indexes of the sample enterprises according to the standard database; establishing a multi-dimensional clustering analysis model of the sample enterprise through a K-means clustering algorithm according to the enterprise index; determining clustering clusters of the sample enterprises under different index dimensions according to the multi-dimensional clustering analysis model; and determining the marketing response probability of the client enterprises through a K nearest neighbor algorithm according to the clustering clusters respectively corresponding to the sample enterprises under different index dimensions.

In one example, establishing a standard database according to the enterprise data of the sample enterprise specifically includes: determining a data structure of standard data in the standard database; performing data processing on the enterprise data to convert the enterprise data into the standard data; performing content identification on the standard data, and determining a content repetition rate corresponding to the standard data; and if the content repetition rate is higher than a preset threshold value, deleting the standard data.

In one example, determining the enterprise metrics of the sample enterprise according to the criteria database specifically includes: determining the standard data belonging to a preset data dimension in the standard database; the type of the preset data dimension is determined by the type of the marketing task; and performing text recognition on the standard data belonging to the preset data dimension to generate an enterprise index of the sample enterprise.

In one example, the establishing the enterprise multidimensional clustering analysis model through a K-means clustering algorithm according to the enterprise index specifically includes: determining a module entering index in the enterprise indexes according to the marketing task type; carrying out data cleaning on the mold entering indexes to obtain a training sample; and determining a K value in the K mean value clustering algorithm according to the clustering evaluation index and the training sample.

In one example, performing data cleaning on the mold-entering index to obtain a training sample specifically includes: determining a missing value corresponding to each index in the mold-entering indexes, and deleting the mold-entering indexes as invalid values if the missing values are larger than a preset proportion; performing multiple collinearity inspection on the rest mold-entering indexes, and removing the mold-entering indexes with collinearity in the rest mold-entering indexes; and carrying out standardization processing on the residual mold-entering indexes to obtain a training sample.

In one example, determining a K value in the K-means clustering algorithm according to a clustering evaluation index and the training examples specifically includes: traversing the K value in a preset interval, and carrying out K mean value clustering analysis on the training samples according to different values of the K; drawing a corresponding clustering analysis result graph according to the different values of the K, and obtaining corresponding CH measurement index values; and selecting the K value corresponding to the maximum CH metric index value as the K value in the K-means clustering algorithm.

In one example, after determining cluster clusters respectively corresponding to the sample enterprise under different dimensions according to the multidimensional cluster analysis model, the method further includes: and selecting the cluster with the highest probability of responding to the marketing task as a target cluster according to the indexes corresponding to the cluster clusters respectively.

In one example, determining the marketing response probability of the client enterprise according to the clustering clusters respectively corresponding to the sample enterprise under different dimensions through a K-nearest neighbor algorithm specifically includes: determining input client enterprise data, and calling the multidimensional clustering analysis model to predict the probability that the client enterprise belongs to the target clustering cluster; and taking the probability that the client enterprise belongs to the target cluster as the marketing response probability of the client enterprise.

The present application further provides a marketing response model building apparatus, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to: determining enterprise data of a sample enterprise which is acquired in advance, and establishing a standard database according to the enterprise data of the sample enterprise; determining enterprise indexes of the sample enterprises according to the standard database; establishing a multi-dimensional clustering analysis model of the sample enterprise through a K-means clustering algorithm according to the enterprise index; determining clustering clusters of the sample enterprises under different index dimensions according to the multi-dimensional clustering analysis model; and determining the marketing response probability of the client enterprises through a K nearest neighbor algorithm according to the clustering clusters respectively corresponding to the sample enterprises under different index dimensions.

The present application further provides a non-volatile computer storage medium storing computer-executable instructions configured to: determining enterprise data of a sample enterprise which is acquired in advance, and establishing a standard database according to the enterprise data of the sample enterprise; determining enterprise indexes of the sample enterprises according to the standard database; establishing a multi-dimensional clustering analysis model of the sample enterprise through a K-means clustering algorithm according to the enterprise index; determining clustering clusters of the sample enterprises under different index dimensions according to the multi-dimensional clustering analysis model; and determining the marketing response probability of the client enterprises through a K nearest neighbor algorithm according to the clustering clusters respectively corresponding to the sample enterprises under different index dimensions.

Compared with the traditional customer marketing response model method based on customer behavior analysis, the method provided by the application is based on information of enterprises such as enterprise workers and merchants, social security and tax, and the like, and performs operations such as data merging, data alignment, data fusion and the like on multi-source data through multi-source data fusion, so that the analysis scene of the customer marketing response model is widened. The method can perform clustering analysis on the clients under the condition of no sample, has simple and quick algorithm, and solves the problem of less data accumulation at the initial stage of service development. The marketing strategy is determined according to the characteristic setting after the customer group is classified, so that the marketing efficiency can be further improved, and the cost is reduced.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:

FIG. 1 is a schematic flow chart of a method for constructing a marketing response model according to an embodiment of the present application;

fig. 2 is a schematic diagram of a marketing response model building device in an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

Fig. 1 is a schematic flowchart of a marketing response model construction method according to an embodiment of the present disclosure. Certain input parameters or intermediate results in the flow process allow for manual intervention adjustments to help improve accuracy.

The analysis method according to the embodiment of the present application may be implemented by a terminal device or a server, and the present application is not limited to this. For convenience of understanding and description, the following embodiments are described in detail by taking a server as an example.

As shown in fig. 1, an embodiment of the present application provides a marketing response model building method, including:

s101: enterprise data of sample enterprises collected in advance are determined, and a standard database is established according to the enterprise data of the sample enterprises.

The enterprise data is from a plurality of data sources, and can be from enterprise government data, social data, third party data and the like, including relevant data such as market supervision, human society, committee for issuing and modifying, public security, and housing bureaus, and the third party data of the enterprise includes information such as judicial laws, intellectual property rights and blacklists. Meanwhile, the data dimension of the enterprise data of the sample enterprise at least comprises the following steps: the enterprise data can reflect the state information of the sample enterprise more comprehensively. A standard database is then built based on the enterprise data for the sample enterprise.

S102: and determining the enterprise indexes of the sample enterprises according to the standard database.

And the server combs the enterprise data in the standard database based on the standard database to establish an enterprise index system. The enterprise index system covers 8 dimensions in total, namely enterprise repayment ability, management ability, repayment willingness, enterprise qualification, profitability, operation ability, innovation ability and development ability. The management capacity mainly comprises the indexes of treatment perfection, the change times of the director in about 6 months, the change times of the director in about one year, the change times of legal representatives in about one year, the same-ratio increase rate of financial expenses, the ratio of financial expenses, the per-employee income, the personnel loss rate and the like. The repayment capacity mainly comprises indexes such as an asset liability rate, a equity liability ratio, a net asset industry level, a total liability share same-ratio increase rate, a quick-action ratio, a flowing ratio, a cash flowing liability ratio and the like, and indexes such as a tax payment credit level, whether normal tax payment is carried out, the accumulated tax payment times of the enterprise in the last six months, the current tax payment time from the last tax payment, the tax payment times of the non-payment currently due, and the illegal times of major tax payment in the last six months. But when establishing the marketing response model, the index dimensions of the enterprise at least comprise: the system comprises five first-level dimensions, namely enterprise background, enterprise stability, enterprise operation capacity, enterprise development capacity and enterprise technological innovation capacity, wherein the five first-level dimensions are closely related to marketing response.

S103: and establishing a multi-dimensional clustering analysis model of the sample enterprise through a K-means clustering algorithm according to the enterprise index.

And the server establishes a multidimensional clustering analysis model obtained by enterprise data of the sample enterprise by using a K-means clustering algorithm according to enterprise indexes. The K-means clustering algorithm is a clustering method which is common, wide in application range and rich in use scenes. The clustering algorithm considered by Du Wei, Zhao Chunrong, Huang Wei and the like can effectively divide customers in the customer segmentation model, so that enterprises can respectively carry out marketing according to the characteristics of different customers, the marketing effect of the enterprises is improved, and the investment return rate is improved. The method improves the K-means clustering algorithm, establishes a customer segmentation model for tourism customers, and divides the tourism customers according to different requirements of the customers, so that the obtained model has a good effect.

S104: and determining the clustering clusters of the sample enterprises under different index dimensions according to the multi-dimensional clustering analysis model.

After the server verifies that the multidimensional clustering analysis model tends to be stable under each index dimension, enterprise description suitable for each clustering cluster is determined, for example, the enterprise repayment willingness under the cluster partition is strong, the enterprise debt paying capability is strong, the enterprise qualification is good, and the like. And comprehensively evaluating the enterprise clustering effect under each dimension from two aspects of service and clustering effect performance, and finally determining the optimal clustering division result.

S105: and determining the marketing response probability of the client enterprises through a K nearest neighbor algorithm according to the clustering clusters respectively corresponding to the sample enterprises under different index dimensions.

And the server determines the marketing response probability of the client enterprises by using a K nearest neighbor algorithm through clustering clusters of the sample enterprises under different index dimensions. The core idea of the K-Nearest Neighbor (KNN) algorithm is that if most of K Nearest Neighbor samples of a sample in the feature space belong to a certain class, the sample also belongs to this class. Where K is usually an integer no greater than 20, to distinguish it from the value of K in the K-means clustering algorithm, where the value of M is used instead of the value of K in the K-nearest neighbor algorithm.

In one example, after obtaining enterprise data of a sample enterprise, a server first needs to establish a uniform data standard to perform standardized management on the multi-source data in storage in a standardized manner; secondly, the enterprise multi-source data is treated and processed through ETL and other data treatment tools, if the enterprise data is storable data such as internet data, the enterprise data can be regularly pulled at intervals, real-time interface data are processed through a memory, and data processing, data standardization, index calculation and light characteristic mining are carried out on the enterprise data in combination with a batch processing mode; and then, performing content identification on the standard data, determining the repetition rate of the content corresponding to the standard data, and deleting the standard data corresponding to the repeated content if the repetition rate is higher than a preset threshold value so as to delete the repeated data and reduce the consumption of a storage space. And finally, fusing and converging the enterprise data into a unified data warehouse through transverse and longitudinal data fusion, wherein the data warehouse stores information such as standard library data, a processed index library, a processed feature library and the like after multi-source data fusion.

In one example, after the standard database is established by the server, a preset data dimension needs to be selected according to the type of the marketing task, so as to preliminarily screen the standard data in the standard database. And performing text recognition on the screened standard data to generate enterprise indexes of the sample enterprises. Generally, when generating the index, the staff is required to participate, and the enterprise index is generated by help of experience.

In one example, when the server establishes a multidimensional clustering analysis model of an enterprise, firstly, a model entering index of a plurality of enterprise indexes is determined according to a marketing task type, exploratory data analysis is performed on the model entering index, then, data cleaning is performed on the model entering index, so that a training sample is obtained, and then, a K value in the K-means clustering algorithm is determined according to a clustering evaluation index and the training sample.

When exploratory data analysis is carried out on the in-mold indexes, simple description statistics is carried out on the screened in-mold indexes, the variance, the mean value, the median, the data distribution and the like of each index are analyzed, after the data are subjected to simple statistical analysis, data segmentation is carried out on specific index data, and the dynamic change condition of the data and the value taking condition under a certain specific condition are deeply analyzed; and performing visual analysis on the model-entering sample indexes by drawing a histogram curve of the univariate, a relation curve of the univariate and the target variable and the like.

Further, when data cleaning is performed on standard data corresponding to the modulus-entering index, an invalid value in the index needs to be processed, firstly, numerical quantization needs to be performed on a part of quantifiable index, then, missing value statistics needs to be performed on the modulus-entering index, and a modulus-entering index with a missing value larger than a preset proportion is removed, for example, a training index with a missing value larger than 90% is removed. The same-value rate statistics can be carried out on the remaining indexes, and the characteristic that the removal attribute has only one value, for example, the removal attribute has the same-value rate of more than 85 percent; and performing multiple collinearity test on the mould-entering indexes subjected to the missing homonymy filtering, and removing the collinearity indexes in the rest mould-entering indexes. And finally, carrying out standardization processing on the remaining model-entering indexes by adopting a standardization processing method to obtain training examples for the clustering process.

In an embodiment, when the K value in the K-means clustering algorithm is determined by using the CH metric index as a clustering evaluation index and using the CH metric index and a training example, the K value may be traversed within a preset value interval, for example, a value range of the K value is set to take values within an interval of 1 to 10, K-means clustering analysis is respectively performed, a clustering analysis result graph is drawn, CH metric index values under different K values are sequentially calculated, and the larger the CH value is, the better the clustering effect is. And selecting a K value result with the optimal clustering effect by combining a visual result graph of the clustering analysis and different values of CH.

Further, determining a K value of enterprise clustering analysis under each dimension through the CH measurement index, training the clustering analysis model by using the corresponding optimal K value after verifying that the clustering model under each dimension tends to be stable, drawing a cluster division visualization graph formed by the enterprise clustering analysis results under each dimension, and observing the effect of the enterprise clustering analysis under each dimension. And analyzing the distribution condition of each index value aiming at each cluster, determining enterprise description suitable for the cluster division, comprehensively evaluating the enterprise clustering effect under each dimension from two aspects of service and clustering effect performance, and finally determining the optimal clustering division result. And selecting the cluster with the highest probability of responding to the marketing task as a target cluster according to the indexes corresponding to the cluster clusters respectively.

In one embodiment, after obtaining cluster clusters respectively corresponding to sample enterprises in different dimensions, a marketing response model is established by a K nearest neighbor algorithm, namely a supervised classification KNN method, with a cluster model entering index as X and a cluster result cluster as Y, after the sample enterprises in each dimension are clustered and divided, enterprise data of the client enterprises are input, and after a cluster analysis model is called to predict cluster division of the client enterprises, the cluster division labels of the enterprises are given by the model. And (3) adopting each dimension modeling index as X used in KNN classification model training, adopting the result category of cluster analysis cluster division as Y with supervision classification, and training the KNN classification model, wherein the probability that the multi-dimensional cluster analysis model predicts that the enterprise belongs to the target cluster is the enterprise marketing response probability.

As shown in fig. 2, an embodiment of the present application further provides a marketing response model building apparatus, including:

at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to: determining enterprise data of a sample enterprise which is acquired in advance, and establishing a standard database according to the enterprise data of the sample enterprise; determining enterprise indexes of the sample enterprises according to the standard database; establishing a multi-dimensional clustering analysis model of the sample enterprise through a K-means clustering algorithm according to the enterprise index; determining clustering clusters of the sample enterprises under different index dimensions according to the multi-dimensional clustering analysis model; and determining the marketing response probability of the client enterprises through a K nearest neighbor algorithm according to the clustering clusters respectively corresponding to the sample enterprises under different index dimensions.

An embodiment of the present application further provides a non-volatile computer storage medium storing computer-executable instructions, where the computer-executable instructions are configured to:

at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to: determining enterprise data of a sample enterprise which is acquired in advance, and establishing a standard database according to the enterprise data of the sample enterprise; determining enterprise indexes of the sample enterprises according to the standard database; establishing a multi-dimensional clustering analysis model of the sample enterprise through a K-means clustering algorithm according to the enterprise index; determining clustering clusters of the sample enterprises under different index dimensions according to the multi-dimensional clustering analysis model; and determining the marketing response probability of the client enterprises through a K nearest neighbor algorithm according to the clustering clusters respectively corresponding to the sample enterprises under different index dimensions.

The embodiments in the present application are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the device and media embodiments, the description is relatively simple as it is substantially similar to the method embodiments, and reference may be made to some descriptions of the method embodiments for relevant points.

The device and the medium provided by the embodiment of the application correspond to the method one to one, so the device and the medium also have the similar beneficial technical effects as the corresponding method, and the beneficial technical effects of the method are explained in detail above, so the beneficial technical effects of the device and the medium are not repeated herein.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application 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 application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. 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.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims (10)

1. A marketing response model construction method, the method comprising:

determining enterprise data of a sample enterprise which is acquired in advance, and establishing a standard database according to the enterprise data of the sample enterprise;

determining enterprise indexes of the sample enterprises according to the standard database;

establishing a multi-dimensional clustering analysis model of the sample enterprise through a K-means clustering algorithm according to the enterprise index;

determining clustering clusters of the sample enterprises under different index dimensions according to the multi-dimensional clustering analysis model;

and determining the marketing response probability of the client enterprises through a K nearest neighbor algorithm according to the clustering clusters respectively corresponding to the sample enterprises under different index dimensions.

2. The method according to claim 1, wherein building a standard database based on the enterprise data of the sample enterprise comprises:

determining a data structure of standard data in the standard database;

performing data processing on the enterprise data to convert the enterprise data into the standard data;

performing content identification on the standard data, and determining a content repetition rate corresponding to the standard data;

and if the content repetition rate is higher than a preset threshold value, deleting the standard data.

3. The method of claim 2, wherein determining the business indicator for the sample business from the criteria database comprises:

determining the standard data belonging to a preset data dimension in the standard database; the type of the preset data dimension is determined by the type of the marketing task;

and performing text recognition on the standard data belonging to the preset data dimension to generate an enterprise index of the sample enterprise.

4. The method according to claim 1, wherein the establishing the enterprise multidimensional clustering analysis model through a K-means clustering algorithm according to the enterprise index specifically comprises:

determining a module entering index in the enterprise indexes according to the marketing task type;

carrying out data cleaning on the mold entering indexes to obtain a training sample;

and determining a K value in the K mean value clustering algorithm according to the clustering evaluation index and the training sample.

5. The method according to claim 4, wherein the data cleaning of the mold-entering index to obtain a training sample specifically comprises:

determining a missing value corresponding to each index in the mold-entering indexes, and deleting the mold-entering indexes as invalid values if the missing values are larger than a preset proportion;

performing multiple collinearity inspection on the rest mold-entering indexes, and removing the mold-entering indexes with collinearity in the rest mold-entering indexes;

and carrying out standardization processing on the residual mold-entering indexes to obtain a training sample.

6. The method according to claim 5, wherein determining the K value in the K-means clustering algorithm according to the clustering evaluation index and the training examples specifically comprises:

traversing the K value in a preset interval, and carrying out K mean value clustering analysis on the training samples according to different values of the K value;

drawing a corresponding clustering analysis result graph according to different values of the K value, and obtaining corresponding CH measurement index values;

and selecting the K value corresponding to the maximum CH metric index value as the K value in the K-means clustering algorithm.

7. The method of claim 5, wherein after determining cluster clusters respectively corresponding to the sample enterprise in different dimensions according to the multidimensional cluster analysis model, the method further comprises:

and selecting the cluster with the highest probability of responding to the marketing task as a target cluster according to the indexes corresponding to the cluster clusters respectively.

8. The method according to claim 7, wherein determining the marketing response probability of the client enterprise according to the clustering clusters respectively corresponding to the sample enterprise in different dimensions by using a K-nearest neighbor algorithm specifically comprises:

determining input client enterprise data, and calling the multidimensional clustering analysis model to predict the probability that the client enterprise belongs to the target clustering cluster;

and taking the probability that the client enterprise belongs to the target cluster as the marketing response probability of the client enterprise.

9. A marketing response model building apparatus, comprising:

at least one processor; and a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to cause the at least one processor to perform:

determining enterprise data of a sample enterprise which is acquired in advance, and establishing a standard database according to the enterprise data of the sample enterprise;

determining enterprise indexes of the sample enterprises according to the standard database;

establishing a multi-dimensional clustering analysis model of the sample enterprise through a K-means clustering algorithm according to the enterprise index;

determining clustering clusters of the sample enterprises under different index dimensions according to the multi-dimensional clustering analysis model;

and determining the marketing response probability of the client enterprises through a K nearest neighbor algorithm according to the clustering clusters respectively corresponding to the sample enterprises under different index dimensions.

10. A non-transitory computer storage medium storing computer-executable instructions, the computer-executable instructions configured to:

determining enterprise data of a sample enterprise which is acquired in advance, and establishing a standard database according to the enterprise data of the sample enterprise;

determining enterprise indexes of the sample enterprises according to the standard database;

establishing a multi-dimensional clustering analysis model of the sample enterprise through a K-means clustering algorithm according to the enterprise index;

determining clustering clusters of the sample enterprises under different index dimensions according to the multi-dimensional clustering analysis model;

and determining the marketing response probability of the client enterprises through a K nearest neighbor algorithm according to the clustering clusters respectively corresponding to the sample enterprises under different index dimensions.

Images