CN109583966A - A kind of high value customer recognition methods, system, equipment and storage medium - Google Patents
A kind of high value customer recognition methods, system, equipment and storage medium Download PDFInfo
- Publication number
- CN109583966A CN109583966A CN201811494415.3A CN201811494415A CN109583966A CN 109583966 A CN109583966 A CN 109583966A CN 201811494415 A CN201811494415 A CN 201811494415A CN 109583966 A CN109583966 A CN 109583966A
- Authority
- CN
- China
- Prior art keywords
- information
- value
- customer
- historical
- group
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012549 training Methods 0.000 claims description 44
- 238000012360 testing method Methods 0.000 claims description 36
- 238000012216 screening Methods 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 9
- 230000001419 dependent effect Effects 0.000 claims description 9
- 238000005259 measurement Methods 0.000 abstract description 5
- 230000007306 turnover Effects 0.000 abstract 1
- 239000003550 marker Substances 0.000 description 14
- 238000004422 calculation algorithm Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 238000004140 cleaning Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 230000000630 rising effect Effects 0.000 description 4
- 238000012163 sequencing technique Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of high value customer recognition methods, system, equipment and storage medium, it is generated and the corresponding whether matched recognition result of history mark post community information by the way that customer information to be identified is input to the similar crowd recognition model of mark post group, accuracy of identification is high, quantitative measurement is really converted by qualitative description by the value height of client, by being categorized into the high value potentiality client similar with corresponding mark post group to high value customer, determine its stretchable asset space and can management path, targetedly, it is directive to launch more management resources, improve the satisfaction of client, and then promote management level.It can precisely identify the client of high value but low contribution, promote business efficiency, provide more good service for the client of high value but low contribution, optimize customer experience, promotes contribution margin, and then promote result of management, more good service is provided for high net value client, reduces the turnover rate of high net value client.
Description
Technical Field
The invention relates to the field of data mining, in particular to a high-value customer identification method, a high-value customer identification system, high-value customer identification equipment and a storage medium.
Background
The high-value client contributes great value to operating enterprises and brings rich social relations and strong social influence to the enterprises, so the high-value client becomes the key point of enterprise market competition. However, no effective method is available so far, which can accurately locate and identify high-value customers from a large number of customers, and realize that the value of the customers is converted from qualitative description to quantitative measurement, so that enterprises can utilize high-value customer resources to the maximum extent, analyze and predict the behaviors of the high-value customers are developed, the high-value customers are classified, stretchable asset space and operable paths of the high-value customers are determined, more operating resources are put in a targeted and directional manner, personalized services are developed, the satisfaction degree of the customers is improved, and the operating level is further improved, which is a problem that the operating enterprises need to explore and solve on operating roads all the time.
Disclosure of Invention
In order to solve the above technical problems, an object of the present invention is to provide a high-value customer identification method, system, device and storage medium.
According to an aspect of the present invention, there is provided a high-value customer identification method, including the steps of:
inputting the customer information to be identified into a benchmark group similar population identification model to generate an identification result whether the customer information is matched with the corresponding historical benchmark group information, wherein the benchmark group similar population identification model is used for representing the corresponding relation between the customer information to be identified and the identification result,
wherein,
the benchmark group similarity crowd identification model acquires historical high-value customer information with contribution values exceeding a first preset threshold value to form historical high-value customer group sample information;
clustering historical high-value customer group sample information based on preset customer characteristic information to generate corresponding historical clustered customer group information, and collecting the historical high-value customer information exceeding a second preset threshold value in the historical clustered customer group information to form historical benchmarking group information;
and screening corresponding historical high-value client information from the historical benchmarking group information as corresponding training data information and testing data information based on screening conditions of preset training data and testing data, training the initial model based on a first preset relation between a label and preset client characteristic information in the historical benchmarking group information and the corresponding training data, testing the initial model by the corresponding testing data and adjusting the initial model according to actual business requirement indexes.
Further, the preset client characteristic information comprises at least one of assets, running characteristics, liability conditions, consumption ability and preference, investment product characteristics and personal basic characteristic information.
Furthermore, the first preset relationship between the label in the historical benchmarking group information and the preset customer feature information is that the label of the historical benchmarking group information is used as a dependent variable, and the preset customer feature information is used as an independent variable.
Further, the identification result is whether the matching degree of the customer information to be identified and the corresponding historical benchmarking group information exceeds a fourth preset threshold, if so, the customer information to be identified is classified into the corresponding historical benchmarking group information to serve as the high-value potential customer of the corresponding benchmarking group.
Further, the high-value customer identification method further includes:
generating the liftable asset level of the high-value potential customer based on a second preset relation between corresponding asset information in the historical benchmarking group information and the liftable asset level of the high-value potential customer in a first preset historical period.
Further, the high-value customer identification method further includes:
and generating the liftable contribution level of the high-value potential customer based on a third preset relation between corresponding contribution value information in the historical benchmarking group information and the liftable contribution level of the high-value potential customer in a second preset historical period.
Further, the high-value customer identification method further includes:
and generating the operation product information and the corresponding product operation path information of the high-value potential customer based on the product purchase preference information and the product purchase path information of the corresponding customer in the historical benchmarking group information in a third preset historical period.
Further, generating the operation product information and the corresponding product operation path information of the high-value potential customer based on the product purchase preference information and the product purchase path information of the corresponding customer in the historical benchmarking group information in a third preset historical period, including:
and sequencing the purchased products of the customer based on the product category information and the corresponding deal amount of the purchased products of the customer in the historical benchmarking group information in a third preset historical period to determine the product purchase preference information of the corresponding customer.
The product category information comprises at least one of financing class, credit card, vehicle loan, house loan, consumption loan, micro loan, credit loan and other information classified according to business conditions.
The first preset threshold, the second preset threshold, the third preset threshold and the fourth preset threshold are all determined according to actual service conditions.
The high-value customer identification method further comprises the following steps: generating business target and business direction information aiming at the high-value potential customer based on the liftable asset level, the liftable contribution level, the business product information and the corresponding product business path information of the high-value potential customer.
According to another aspect of the present invention, there is provided a high value customer identification system comprising:
the identification unit is configured to input the client information to be identified into a benchmark population similar population identification model to generate an identification result whether the client information to be identified is matched with the corresponding historical benchmark population information, the benchmark population similar population identification model is used for representing the corresponding relation between the client information to be identified and the identification result,
the benchmark group similarity population identification model generation unit is configured for acquiring historical high-value customer information of which the contribution value exceeds a first preset threshold value to form historical high-value customer group sample information;
clustering historical high-value customer group sample information based on preset customer characteristic information to generate corresponding historical clustered customer group information, and collecting the historical high-value customer information exceeding a second preset threshold value in the historical clustered customer group information to form historical benchmarking group information;
and screening corresponding historical high-value client information from the historical benchmarking group information as corresponding training data information and testing data information based on screening conditions of preset training data and testing data, training the initial model based on a first preset relation between a label and preset client characteristic information in the historical benchmarking group information and the corresponding training data, testing the initial model by the corresponding testing data and adjusting the initial model according to actual business requirement indexes.
Further, the preset client characteristic information comprises at least one of assets, running characteristics, liability conditions, consumption ability and preference, investment product characteristics and personal basic characteristic information.
The first preset relationship between the label in the historical benchmarking group information and the preset customer characteristic information is that the label of the historical benchmarking group information is used as a dependent variable, and the preset customer characteristic information is used as an independent variable.
Further, the identification result in the identification unit is whether the matching degree between the customer information to be identified and the corresponding historical benchmarking group information exceeds a fourth preset threshold, and if yes, the high-value potential customer information generation unit is driven to be configured to classify the customer information to be identified into the corresponding historical benchmarking group information to serve as the high-value potential customer of the corresponding benchmarking group.
The high value potential customer information generation unit is further configured to:
generating the liftable asset level of the high-value potential customer based on a second preset relation between corresponding asset information in the historical benchmarking group information and the liftable asset level of the high-value potential customer in a first preset historical period.
The high value potential customer information generation unit is further configured to:
and generating the liftable contribution level of the high-value potential customer based on a third preset relation between corresponding contribution value information in the historical benchmarking group information and the liftable contribution level of the high-value potential customer in a second preset historical period.
The high value potential customer information generation unit is further configured to:
and generating the operation product information and the corresponding product operation path information of the high-value potential customer based on the product purchase preference information and the product purchase path information of the corresponding customer in the historical benchmarking group information in a third preset historical period.
The high value potential customer information generation unit is further configured to:
and sequencing the purchased products of the customer based on the product category information and the corresponding deal amount of the purchased products of the customer in the historical benchmarking group information in a third preset historical period to determine the product purchase preference information of the corresponding customer.
The product category information comprises at least one of financing class, credit card, vehicle loan, house loan, consumption loan, micro loan, credit loan and other information classified according to business conditions.
The first preset threshold, the second preset threshold, the third preset threshold and the fourth preset threshold are all determined according to actual service conditions.
The high value potential customer information generation unit is further configured to: and generating operation targets and operation direction information aiming at the customers based on the liftable asset level, the liftable contribution level, the operation product information and the corresponding product operation path information of the high-value potential customers.
According to another aspect of the present invention, there is provided an apparatus comprising:
one or more processors;
a memory for storing one or more programs,
the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of the above.
According to another aspect of the invention, there is provided a computer readable storage medium storing a computer program which, when executed by a processor, implements a method as defined in any one of the above.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the high-value customer identification method disclosed by the invention, historical high-value customer information with contribution values exceeding a first preset threshold value is collected to form historical high-value customer group sample information; clustering historical high-value customer group sample information based on preset customer characteristic information to generate corresponding historical clustered customer group information, and collecting the historical high-value customer information exceeding a second preset threshold value in the historical clustered customer group information to form historical benchmarking group information; based on the screening conditions of preset training data and test data, screening corresponding historical high-value client information from the historical benchmarking group information to be respectively used as corresponding training data information and test data information, training an initial model based on a first preset relation between labels in the historical benchmarking group information and preset client characteristic information and the corresponding training data, testing the corresponding test data and adjusting according to actual business requirement indexes to generate a benchmarking group similar population recognition model, wherein the benchmarking group similar population recognition model is strong in recognition pertinence and relevant to actual client conditions because the benchmarking group similar population recognition model is obtained according to the historical numbers, and a recognition result whether the client information to be recognized is matched with the corresponding historical benchmarking group information or not is generated by inputting the client information to be recognized into the benchmarking group similar population recognition model, so that the recognition precision is high and is suitable for the actual business conditions, the value of the customers is really converted into quantitative measurement from qualitative description, the high-value customers are classified into high-value potential customers of the same type as the corresponding benchmarking groups, stretchable asset space and operable paths of the customers are determined, more operating resources are put in a targeted and directional manner, the satisfaction degree of the customers is improved, and the operating level is further improved. The method has the advantages that the high-value and low-contribution customers can be accurately identified, the operation efficiency is improved, the higher-quality service is provided for the high-value and low-contribution customers, the customer experience is optimized, the contribution value is improved, the operation effect is further improved, the higher-quality service is provided for the high-net-value customers, and the loss rate of the high-net-value customers is reduced.
2. The high-value client identification system disclosed by the invention is simple in composition, the identification result whether the high-value client identification system is matched with the corresponding historical benchmarking group information or not is generated through the mutual cooperation of each composition system and unit, the identification precision is high, the high-value client identification system corresponds to the corresponding historical benchmarking group information, the value of the client is really converted into quantitative measurement from qualitative description, and the high-value client identification system is convenient for an enterprise to predict the liftable asset level, the liftable contribution level, the operation product information and the corresponding product operation path information of the high-value potential client of the corresponding benchmarking group through analyzing the corresponding historical benchmarking group information in the later stage so as to generate the operation target and the operation direction information aiming at the client. More operation resources are put in a targeted and directional manner, the satisfaction degree of customers is improved, and the operation level is further improved.
3. The equipment and the computer readable storage medium storing the computer program generate the identification result whether the information is matched with the corresponding historical benchmarking group information, really convert the value of the client from qualitative description into quantitative measurement, have high identification precision, and are convenient for enterprises to predict the liftable asset level, the liftable contribution level, the operating product information and the corresponding product operating path information of the high-value potential client of the corresponding benchmarking group by analyzing the corresponding historical benchmarking group information at the later stage to generate the operating target and the operating direction information aiming at the client. The system is beneficial to enterprises to put more operation resources in a targeted and directional manner, improves the satisfaction degree of customers and further improves the operation level.
Drawings
FIG. 1 is a flow diagram of a high value customer identification method according to an embodiment;
FIG. 2 is a graph of the relationship between the contour coefficient and the number of clusters according to the embodiment.
Detailed Description
In order to better understand the technical scheme of the invention, the invention is further explained by combining the specific embodiment and the attached drawings of the specification.
The first embodiment is as follows:
the embodiment provides a high-value customer identification method, as shown in fig. 1, including the following steps:
s1, collecting historical high-value customer information with contribution value exceeding a first preset threshold value to form historical high-value customer group sample information, and specifically selecting n top-ranked contributing income (high value) in the past year1Forming historical high-value customer group sample information and confirming high-value crowds, wherein the first step is determined according to actual business conditionsA predetermined threshold n1Taking a value (which can be 5-10%);
s2, clustering and clustering historical high-value customer group sample information based on preset customer characteristic information, wherein the preset customer characteristic information comprises at least one of assets, flow characteristics, liability conditions, consumption capacity and preference, investment product characteristics and personal basic characteristic information, generating corresponding historical clustering customer group information, collecting the historical high-value customer information exceeding a second preset threshold value in the historical clustering customer group information, and forming historical benchmarking group information, and the specific process can be as follows:
s21, cleaning population characteristics: the preset client characteristic information, namely the crowd characteristic, including the asset, the stream characteristic, the liability condition, the consumption ability and the preference, the investment product characteristic and the information of the personal basic characteristic (academic calendar, age, gender and the like) is determined, as shown in the preset client characteristic information in table 1.
Table 1: presetting customer characteristic information
S22, processing the preset client characteristics:
s221, discarding the high-loss-rate characteristic, wherein the cleaning characteristic may not be complete due to different service conditions, and the removal loss rate is greater than n3Is characterized by n310% -30% is recommended.
S222, carrying out data standardization operation on the characteristics.
S23, for TOP n1And clustering the customers, namely historical high-value customer group samples to obtain k historical clustering customer groups. Taking a K-means algorithm as an example, determining the number of clusters by using an outline coefficient method, that is, calculating the outline coefficient of the K-means parameter K (which can be 3 to 15) under each value by probing, and taking the rising inflection point of the outline coefficient as the number of guest clusters finally confirmed by clustering, as shown in fig. 2.
After the parameter K is determined, clustering is carried out on the historical high-value customer group samples (high-value customer groups) to obtain K historical clustering customer groups (high-value customer groups). For the clustering result, when the method is applied, the characteristics of the clustering group can be described by adopting a mean value calculation method for the characteristics in each cluster, and the group needing to be operated is selected.
S24, selecting each cluster group (selected to be operated), namely the historical cluster passenger group with the contribution income (high value) ranked at the top n in the past year2The client forms a historical benchmarking group as a benchmarking group of the group, and determines a second preset threshold n according to actual business conditions2Taking values (can be 5% -10%).
S3, based on the screening condition of preset training data and test data, screening corresponding historical high-value customer information from historical benchmarking group information to be respectively used as corresponding training data information and test data information, training an initial model based on a first preset relation between a label in the historical benchmarking group information and preset customer feature information and the corresponding training data, testing the corresponding test data and adjusting according to actual business requirement indexes to generate a benchmarking group similar population recognition model, wherein the first preset relation between the label in the historical benchmarking group information and the preset customer feature information is that the label of the historical benchmarking group information is used as a dependent variable, the preset customer feature information is used as an independent variable,
specifically, for example, the XGBoost algorithm is used for model training and learning, a historical benchmarking group, such as a high-value benchmarking customer group a, a high-value customer group B, a high-value customer group C and the like, is divided into a training data set (accounting for 80%) and a test data set (accounting for 20%), based on a historical benchmarking group information label, such as a customer group label, as a dependent variable, customer characteristic information, i.e., user characteristics, is preset as an independent variable, the initial model is trained by using the training data set under optimized initial model parameters (wherein, the main parameters of the initial model include a learning rate, a maximum depth of a tree, a minimum leaf node sample weight, a minimum loss function required by node classification and the like), and after a training result is determined, the model is verified by using the test data set. Finally, according to actual service demand indexes (recall rate, accuracy rate and the like), adjusting the model to determine a benchmark group similar population recognition model;
s4, inputting the information of the client to be identified into the marker post group similar population identification model to generate an identification result whether the information of the client to be identified is matched with the information of the corresponding historical marker post group, wherein the identification result is whether the matching degree of the information of the client to be identified and the information of the corresponding historical marker post group exceeds a fourth preset threshold value, if yes, the information of the client to be identified is classified into the information of the corresponding historical marker post group to serve as the high-value potential client of the corresponding marker post group, and the marker post group similar population identification model is used for representing the corresponding relation between the information of the client to be identified and the identification result.
The method specifically comprises the following steps: after the benchmark group similar population recognition model is determined, the model is applied to a target object needing to be detected, namely a client to be recognized, and n before each client group is taken to predict scores according to the number of clients capable of being operated by actual business4Whether the matching degree of the information of the client to be identified and the information of the corresponding historical benchmarking group exceeds a fourth preset threshold value or not is judged, the similar population of each high-value client group, namely the high-value potential client of the corresponding benchmarking group is obtained, wherein the fourth preset threshold value n is used for judging whether the matching degree of the information of the client to be identified and the information of the corresponding historical benchmarking group exceeds the fourth preset threshold value4According to the actual business requirements.
S5, generating the liftable asset level of the high-value potential customer based on the corresponding asset information in the historical benchmarking group information in the first preset historical period and the second preset relation of the liftable asset level of the high-value potential customer,
specifically, the liftable asset level of the high-similarity high-value crowd is determined by taking a first preset history period, such as a mean median of a month of an asset, in an inner layer level of a history benchmarking group, namely each high-value benchmarking group, as the benchmarking asset level in the level, namely the liftable asset level of the high-value potential customer is generated;
generating a liftable contribution level of the high-value potential customer based on a third preset relationship between corresponding contribution value information in the historical benchmarking group information and the liftable contribution level of the high-value potential customer in a second preset history period, and specifically determining the liftable contribution level of the high-similarity high-value crowd, namely generating the liftable contribution level of the high-value potential customer, by using the historical benchmarking group, namely the median of the contribution level of the next year in the second preset history period, such as the next year, in the level in each high-value benchmarking group;
generating the operation product information and the corresponding product operation path information of the high-value potential customer based on the product purchase preference information and the product purchase path information of the corresponding customer in the historical benchmarking group information in a third preset historical period, which specifically comprises the following steps:
and sequencing the purchased products of the client based on the product category information of the purchased products of the client and the corresponding deal amount in the historical benchmarking group information in the third preset historical period to determine the product purchase preference information of the corresponding client, wherein the product category information comprises at least one item of financing category, credit card, car loan, house loan, consumption loan, micro loan, credit loan and other classification information according to the service condition, as shown in table 2.
Table 2: define product categories (categorized according to business' existing products):
specifically, products and operation paths of similar high-value groups, namely the operation product information of the high-value potential customer and the corresponding product operation path information, are determined by counting third preset history periods such as product purchase preference and product purchase paths in the last year in the history marker post groups, namely the high-value marker post groups.
The method specifically comprises the following steps:
(1) according to the product categories defined above, such as financial product categories, the purchasing preference of the products in the last year in each high-value benchmarking group is calculated according to the payment amount ranking. Such as shown in table 3.
Table 3: purchasing preference of financial products in each high-value benchmarking group
(2) The third preset history period in the historical benchmarking groups, i.e. each high-value benchmarking group, is counted, as shown in table 4, the product purchase path in the last year.
Table 4: product purchase path:
(3) and determining products and operation paths of similar high-value crowds through a third preset history period such as product purchase preference and product purchase path in a last year in the history benchmarking group, namely each high-value benchmarking group.
Generating business target and business direction information aiming at the high-value potential customer based on the liftable asset level, the liftable contribution level, the business product information and the corresponding product business path information of the high-value potential customer.
The first preset threshold, the second preset threshold, the third preset threshold and the fourth preset threshold are all determined according to actual service conditions.
By searching high-value sample crowd, more high-value but low-contribution crowd (high potential) are identified by using a data mining similarity learning technology, and the operation target and the operation direction of the similar crowd are determined and applied to other clients.
The present embodiment further provides a high-value customer identification system, including:
the system comprises an identification unit, a benchmark population similarity model and a benchmark population similarity model, wherein the identification unit is used for inputting customer information to be identified into the benchmark population similarity model to generate an identification result whether the customer information is matched with corresponding historical benchmark population information or not, and the benchmark population similarity model is used for representing the corresponding relation between the customer information to be identified and the identification result;
the benchmark group similarity population identification model generation unit is configured for acquiring historical high-value customer information of which the contribution value exceeds a first preset threshold value to form historical high-value customer group sample information;
clustering historical high-value customer group sample information based on preset customer characteristic information to generate corresponding historical clustered customer group information, and collecting the historical high-value customer information exceeding a second preset threshold value in the historical clustered customer group information to form historical benchmarking group information;
and screening corresponding historical high-value client information from the historical benchmarking group information as corresponding training data information and testing data information based on screening conditions of preset training data and testing data, training the initial model based on a first preset relation between a label and preset client characteristic information in the historical benchmarking group information and the corresponding training data, testing the initial model by the corresponding testing data and adjusting the initial model according to actual business requirement indexes.
The preset client characteristic information comprises at least one of assets, stream characteristics, liability conditions, consumption capacity and preference, investment product characteristics and personal basic characteristic information.
The first preset relationship between the label in the historical benchmarking group information and the preset customer characteristic information is that the label of the historical benchmarking group information is used as a dependent variable, and the preset customer characteristic information is used as an independent variable.
And the identification result in the identification unit is whether the matching degree of the customer information to be identified and the corresponding historical benchmarking group information exceeds a fourth preset threshold, if so, the high-value potential customer information generation unit is driven to be configured to classify the customer information to be identified into the corresponding historical benchmarking group information to serve as the high-value potential customer of the corresponding benchmarking group.
The high value potential customer information generation unit is further configured to:
generating the liftable asset level of the high-value potential customer based on a second preset relation between corresponding asset information in the historical benchmarking group information and the liftable asset level of the high-value potential customer in a first preset historical period; generating an upgradable contribution level of the high-value potential customer based on a third preset relationship between corresponding contribution value information in the historical benchmarking group information and the upgradable contribution level of the high-value potential customer in a second preset historical period; generating operation product information and corresponding product operation path information of the high-value potential customer based on product purchase preference information and product purchase path information of a corresponding customer in historical benchmarking group information in a third preset historical period, specifically sequencing purchased products of the customer to determine the product purchase preference information of the corresponding customer based on product category information and corresponding transaction amount of the purchased products of the corresponding customer in the historical benchmarking group information in the third preset historical period, wherein the product category information comprises at least one item of financing category, credit card, car loan, house loan, consumption loan, micro loan, credit loan and other classification information according to business conditions; and generating operation targets and operation direction information aiming at the customers based on the liftable asset level, the liftable contribution level, the operation product information and the corresponding product operation path information of the high-value potential customers.
It should be understood that the subsystems or elements described in the high value customer identification system correspond to the steps described in the high value customer identification method. Thus, the operations and features described above with respect to the method are equally applicable to the subsystems of the high value subscriber identity system and the elements contained therein, and are not described in detail herein.
As another aspect, the present embodiment also provides an apparatus adapted to implement the embodiments of the present application, the apparatus including a computer system including a Central Processing Unit (CPU) that can perform various appropriate actions and processes according to a corresponding program stored in a Read Only Memory (ROM) for executing the respective steps described in the above-described high-value client identifying method or a corresponding program loaded from a storage section into a Random Access Memory (RAM) for executing the respective steps described in the above-described high-value client identifying method. In the RAM, various programs and data necessary for system operation are also stored. The CPU, ROM, and RAM are connected to each other via a bus. An input/output (I/O) interface is also connected to the bus.
The following components are connected to the I/O interface: an input section including a keyboard, a mouse, and the like; an output section including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section including a hard disk and the like; and a communication section including a network interface card such as a LAN card, a modem, or the like. The communication section performs communication processing via a network such as the internet. The drive is also connected to the I/O interface as needed. A removable medium such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive as necessary, so that a computer program read out therefrom is mounted into the storage section as necessary.
In particular, according to an embodiment of the present disclosure, the processes described by the respective steps described in the above-described high-value customer identification method may be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program tangibly embodied on a machine-readable medium, the computer program comprising program code for performing the high value customer identification method described above. In such an embodiment, the computer program may be downloaded and installed from a network via the communication section, and/or installed from a removable medium.
The flowcharts in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the flowchart illustrations, and combinations of blocks in the flowchart illustrations, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or by combinations of special purpose hardware and computer instructions.
The units or modules described in the embodiments of the present application may be implemented by software or hardware. The described units or modules may also be provided in a processor. The names of these units or modules do not in some cases constitute a limitation of the unit or module itself.
As another aspect, the present embodiment also provides a computer-readable storage medium, which may be the computer-readable storage medium included in the system in the foregoing embodiment; or it may be a separate computer readable storage medium not incorporated into the device. The computer readable storage medium stores one or more programs for use by one or more processors in performing the high value customer identification method described herein.
Example two:
the same features of this embodiment and the first embodiment are not described again, and the different features of this embodiment and the first embodiment are:
the specific value can be selected to rank the contributing income (high value) of the last half year at the top n1Forming historical high-value customer group sample information and confirming high-value crowds, wherein a first preset threshold value n is determined according to actual business conditions1Taking values (can be 5%);
s221, discarding the high-loss-rate characteristic, wherein the cleaning characteristic may not be complete due to different service conditions, and the removal loss rate is greater than n3Is characterized by n3May be 10%.
S23, for TOP n1And clustering the customers, namely historical high-value customer group samples to obtain k historical clustering customer groups. Taking a K-means algorithm as an example, determining the clustering number by using an outline coefficient method, namely calculating the outline coefficient of the K-means parameter K (which can be 3) under each value by probing, and taking the rising inflection point of the outline coefficient as the number of guest groups finally confirmed by clustering.
S24, selecting each cluster group (selected to be operated), namely the historical cluster passenger group with the contribution income (high value) ranked at the top n in the past half year2The client forms a historical benchmarking group as a benchmarking group of the group, and determines a second preset threshold n according to actual business conditions2Take value (can be 5%).
Specifically, for example, the XGBoost algorithm is used for model training and learning, a historical benchmarking group, such as a high-value benchmarking customer group a, a high-value customer group B, a high-value customer group C and the like, is divided into a training data set (accounting for 75%) and a test data set (accounting for 25%), based on a historical benchmarking group information label, such as a customer group label, as a dependent variable, customer characteristic information, i.e., user characteristics, is preset as an independent variable, the initial model is trained by using the training data set under optimized initial model parameters (wherein, the main parameters of the initial model include a learning rate, a maximum depth of a tree, a minimum leaf node sample weight, a minimum loss function required by node classification and the like), and after a training result is determined, the model is verified by using the test data set. Finally, according to actual service demand indexes (recall rate, accuracy rate and the like), adjusting the model to determine a benchmark group similar population recognition model;
determining the liftable asset level of the high-similarity high-value crowd by taking a first preset history period, such as a mean median of a month of a recent asset, in an inner-layer level of a history benchmarking group, namely each high-value benchmarking group, as the benchmarking asset level in the level, namely generating the liftable asset level of the high-value potential customer;
determining the liftable contribution level of the high-similarity high-value crowd, namely generating the liftable contribution level of the high-value potential customer, by using a second preset history period, such as the median of the contribution level of the last half year, in the level in the history benchmarking group, namely each high-value benchmarking group as the benchmarking contribution level in the level;
and determining products and operation paths of similar high-value groups, namely the operation product information of the high-value potential customer and the corresponding product operation path information, by counting third preset history periods such as near half year product purchase preference and product purchase paths in the history marker post groups, namely all high-value marker post groups.
The method specifically comprises the following steps:
(1) according to the product categories defined above, such as financial product categories, the purchasing preference of the products in the last half year in each high-value benchmarking group is calculated according to the sequence of the transaction amount.
(2) And (4) counting a third preset history period in the history benchmarking group, namely each high-value benchmarking group, such as a product purchase path in the last half year.
(3) And determining products and operation paths of similar high-value crowds through a third preset history period such as near half year product purchase preference and product purchase paths in the history benchmarking group, namely each high-value benchmarking group.
Example three:
the same features of this embodiment and the first embodiment are not described again, and the different features of this embodiment and the first embodiment are:
the specific value can be selected to rank the contributing income (high value) in the past year at the top n1Forming historical high-value customer group sample information and confirming high-value crowds, wherein a first preset threshold value n is determined according to actual business conditions1Taking values (which can be 10%);
s221, discarding the high-loss-rate characteristic, wherein the cleaning characteristic may not be complete due to different service conditions, and the removal loss rate is greater than n3Is characterized by n3May be 30%.
S23, for TOP n1And clustering the customers, namely historical high-value customer group samples to obtain k historical clustering customer groups. Taking a K-means algorithm as an example, determining the clustering number by using an outline coefficient method, namely calculating the outline coefficient of the K-means parameter K (which can be 15) under each value by probing, and taking the rising inflection point of the outline coefficient as the number of guest groups finally confirmed by clustering.
S24, selecting each cluster group (selected to be operated), namely the historical cluster passenger group with the contribution income (high value) ranked at the top n in the past year2The client forms a historical benchmarking group as a benchmarking group of the group, and determines a second preset threshold n according to actual business conditions2Take values (can be 10%).
Specifically, for example, the XGBoost algorithm is used for model training and learning, a historical benchmarking group, such as a high-value benchmarking customer group a, a high-value customer group B, a high-value customer group C and the like, is divided into a training data set (accounting for 85%) and a test data set (accounting for 15%), based on a historical benchmarking group information label, such as a customer group label, as a dependent variable, customer characteristic information, i.e., user characteristics, is preset as an independent variable, the initial model is trained by using the training data set under optimized initial model parameters (wherein, the main parameters of the initial model include a learning rate, a maximum depth of a tree, a minimum leaf node sample weight, a minimum loss function required by node classification and the like), and after a training result is determined, the model is verified by using the test data set. Finally, according to actual service demand indexes (recall rate, accuracy rate and the like), adjusting the model to determine a benchmark group similar population recognition model;
determining the liftable asset level of the high-similarity high-value crowd by taking a first preset history period, such as a mean median of a month of a recent asset, in an inner-layer level of a history benchmarking group, namely each high-value benchmarking group, as the benchmarking asset level in the level, namely generating the liftable asset level of the high-value potential customer;
determining the liftable contribution level of the high-similarity high-value crowd, namely generating the liftable contribution level of the high-value potential customer, by using a second preset history period, such as the median of the contribution level of the last year, in the level in the history benchmarking group, namely each high-value benchmarking group as the benchmarking contribution level in the level;
and determining products and operation paths of similar high-value groups, namely the operation product information of the high-value potential customer and the corresponding product operation path information, by counting third preset history periods such as product purchase preference and product purchase path in the last year in the history marker post groups, namely the high-value marker post groups.
The method specifically comprises the following steps:
(1) according to the product categories defined above, such as financial product categories, the purchasing preference of the products in the last year in each high-value benchmarking group is calculated according to the payment amount ranking.
(2) And counting a third preset history period in the history benchmarking group, namely each high-value benchmarking group, such as a product purchase path in the last year.
(3) And determining products and operation paths of similar high-value crowds through a third preset history period such as product purchase preference and product purchase path in a last year in the history benchmarking group, namely each high-value benchmarking group.
Example four:
the same features of this embodiment and the first embodiment are not described again, and the different features of this embodiment and the first embodiment are:
the specific value can be selected to rank the contributing income (high value) of the last half year at the top n1Forming historical high-value customer group sample information and confirming high-value crowds, wherein a first preset threshold value n is determined according to actual business conditions1Taking values (can be 7%);
s221, discarding the high-loss-rate characteristic, wherein the cleaning characteristic may not be complete due to different service conditions, and the removal loss rate is greater than n3Is characterized by n3May be 20%.
S23, for TOP n1And clustering the customers, namely historical high-value customer group samples to obtain k historical clustering customer groups. Taking a K-means algorithm as an example, determining the clustering number by using an outline coefficient method, namely calculating the outline coefficient of the K-means parameter K (which can be 10) under each value by probing, and taking the rising inflection point of the outline coefficient as the number of guest groups finally confirmed by clustering.
S24, selecting each cluster group (selected to be operated), namely the historical cluster passenger group with the contribution income (high value) ranked at the top n in the past half year2The client forms a historical benchmarking group as a benchmarking group of the group, and determines a second preset threshold n according to actual business conditions2Take values (can be 8%).
Specifically, for example, the XGBoost algorithm is used for model training and learning, a historical benchmarking group, such as a high-value benchmarking customer group a, a high-value customer group B, a high-value customer group C and the like, is divided into a training data set (accounting for 80%) and a test data set (accounting for 20%), based on a historical benchmarking group information label, such as a customer group label, as a dependent variable, customer characteristic information, i.e., user characteristics, is preset as an independent variable, the initial model is trained by using the training data set under optimized initial model parameters (wherein, the main parameters of the initial model include a learning rate, a maximum depth of a tree, a minimum leaf node sample weight, a minimum loss function required by node classification and the like), and after a training result is determined, the model is verified by using the test data set. Finally, according to actual service demand indexes (recall rate, accuracy rate and the like), adjusting the model to determine a benchmark group similar population recognition model;
determining the liftable asset level of the high-similarity high-value crowd by taking a first preset history period, such as a mean median of a month of a recent asset, in an inner-layer level of a history benchmarking group, namely each high-value benchmarking group, as the benchmarking asset level in the level, namely generating the liftable asset level of the high-value potential customer;
determining the liftable contribution level of the high-similarity high-value crowd, namely generating the liftable contribution level of the high-value potential customer, by using a second preset history period, such as the median of the contribution level of the last half year, in the level in the history benchmarking group, namely each high-value benchmarking group as the benchmarking contribution level in the level;
and determining products and operation paths of similar high-value groups, namely the operation product information of the high-value potential customer and the corresponding product operation path information, by counting third preset history periods such as near half year product purchase preference and product purchase paths in the history marker post groups, namely all high-value marker post groups.
The method specifically comprises the following steps:
(1) according to the product categories defined above, such as financial product categories, the purchasing preference of the products in the last half year in each high-value benchmarking group is calculated according to the sequence of the transaction amount.
(2) And (4) counting a third preset history period in the history benchmarking group, namely each high-value benchmarking group, such as a product purchase path in the last half year.
(3) And determining products and operation paths of similar high-value crowds through a third preset history period such as near half year product purchase preference and product purchase paths in the history benchmarking group, namely each high-value benchmarking group.
The above description is only a preferred embodiment of the application and is illustrative of the principles of the technology employed. It will be appreciated by a person skilled in the art that the scope of the invention as referred to in the present application is not limited to the embodiments with a specific combination of the above-mentioned features, but also covers other embodiments with any combination of the above-mentioned features or their equivalents without departing from the inventive concept. For example, the features described above have similar functions to (but are not limited to) those disclosed in this application.
Claims (10)
1. A high-value customer identification method is characterized by comprising the following steps:
inputting the customer information to be identified into a benchmark group similar population identification model to generate an identification result whether the customer information is matched with the corresponding historical benchmark group information, wherein the benchmark group similar population identification model is used for representing the corresponding relation between the customer information to be identified and the identification result,
wherein,
the benchmark group similarity crowd identification model acquires historical high-value customer information with contribution values exceeding a first preset threshold value to form historical high-value customer group sample information;
clustering historical high-value customer group sample information based on preset customer characteristic information to generate corresponding historical clustered customer group information, and collecting the historical high-value customer information exceeding a second preset threshold value in the historical clustered customer group information to form historical benchmarking group information;
and screening corresponding historical high-value client information from the historical benchmarking group information as corresponding training data information and testing data information based on screening conditions of preset training data and testing data, training the initial model based on a first preset relation between a label and preset client characteristic information in the historical benchmarking group information and the corresponding training data, testing the initial model by the corresponding testing data and adjusting the initial model according to actual business requirement indexes.
2. The high-value client identification method as claimed in claim 1, wherein the first predetermined relationship between the tag in the historical benchmarking group information and the predetermined client characteristic information is that the tag in the historical benchmarking group information is used as a dependent variable and the predetermined client characteristic information is used as an independent variable.
3. The method for identifying high-value customers as claimed in claim 1, wherein the identification result is whether the matching degree of the customer information to be identified and the corresponding historical benchmarking group information exceeds a fourth preset threshold, if yes, the customer information to be identified is classified into the corresponding historical benchmarking group information to be used as the high-value potential customer of the corresponding benchmarking group.
4. A high value customer identification method according to claim 3 further comprising:
generating the liftable asset level of the high-value potential customer based on a second preset relation between corresponding asset information in the historical benchmarking group information and the liftable asset level of the high-value potential customer in a first preset historical period.
5. The high value customer identification method of claim 4, further comprising:
and generating the liftable contribution level of the high-value potential customer based on a third preset relation between corresponding contribution value information in the historical benchmarking group information and the liftable contribution level of the high-value potential customer in a second preset historical period.
6. The high value customer identification method of claim 5, further comprising:
and generating the operation product information and the corresponding product operation path information of the high-value potential customer based on the product purchase preference information and the product purchase path information of the corresponding customer in the historical benchmarking group information in a third preset historical period.
7. The high value customer identification method of claim 6, further comprising:
generating business target and business direction information aiming at the high-value potential customer based on the liftable asset level, the liftable contribution level, the business product information and the corresponding product business path information of the high-value potential customer.
8. A high value customer identification system, comprising:
the identification unit is configured to input the client information to be identified into a benchmark population similar population identification model to generate an identification result whether the client information to be identified is matched with the corresponding historical benchmark population information, the benchmark population similar population identification model is used for representing the corresponding relation between the client information to be identified and the identification result,
the benchmark group similarity population identification model generation unit is configured for acquiring historical high-value customer information of which the contribution value exceeds a first preset threshold value to form historical high-value customer group sample information;
clustering historical high-value customer group sample information based on preset customer characteristic information to generate corresponding historical clustered customer group information, and collecting the historical high-value customer information exceeding a second preset threshold value in the historical clustered customer group information to form historical benchmarking group information;
and screening corresponding historical high-value client information from the historical benchmarking group information as corresponding training data information and testing data information based on screening conditions of preset training data and testing data, training the initial model based on a first preset relation between a label and preset client characteristic information in the historical benchmarking group information and the corresponding training data, testing the initial model by the corresponding testing data and adjusting the initial model according to actual business requirement indexes.
9. An apparatus, characterized in that the apparatus comprises:
one or more processors;
a memory for storing one or more programs,
the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method recited in any of claims 1-7.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811494415.3A CN109583966B (en) | 2018-12-07 | 2018-12-07 | High-value customer identification method, system, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811494415.3A CN109583966B (en) | 2018-12-07 | 2018-12-07 | High-value customer identification method, system, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109583966A true CN109583966A (en) | 2019-04-05 |
| CN109583966B CN109583966B (en) | 2021-01-22 |
Family
ID=65929151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811494415.3A Active CN109583966B (en) | 2018-12-07 | 2018-12-07 | High-value customer identification method, system, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109583966B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110264274A (en) * | 2019-06-21 | 2019-09-20 | 深圳前海微众银行股份有限公司 | Objective group's division methods, model generating method, device, equipment and storage medium |
| CN110837931A (en) * | 2019-11-08 | 2020-02-25 | 中国农业银行股份有限公司 | Customer churn prediction method, device and storage medium |
| CN111105266A (en) * | 2019-11-11 | 2020-05-05 | 中国建设银行股份有限公司 | Client grouping method and device based on improved decision tree |
| CN111311338A (en) * | 2020-03-30 | 2020-06-19 | 网易（杭州）网络有限公司 | User value prediction method and user value prediction model training method |
| CN112529236A (en) * | 2019-09-18 | 2021-03-19 | 泰康保险集团股份有限公司 | Target object identification method and device, electronic equipment and storage medium |
| CN112529628A (en) * | 2020-12-16 | 2021-03-19 | 平安科技（深圳）有限公司 | Client label generation method and device, computer equipment and storage medium |
| CN112669091A (en) * | 2021-01-06 | 2021-04-16 | 腾讯科技（深圳）有限公司 | Data processing method, device and storage medium |
| CN113159972A (en) * | 2021-05-20 | 2021-07-23 | 深圳前海微众银行股份有限公司 | Combination determination method, combination determination device, electronic equipment and computer readable storage medium |
| CN113822715A (en) * | 2021-09-23 | 2021-12-21 | 安徽数据堂科技有限公司 | Data acquisition, training and processing integrated platform analysis method |
| CN114022198A (en) * | 2021-10-27 | 2022-02-08 | 中国人民财产保险股份有限公司 | Customer value scoring method and device |
| CN116308507A (en) * | 2023-02-07 | 2023-06-23 | 网舟联合科技(北京)有限公司 | Potential customer mining method based on customer value, storage medium and equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105225135A (en) * | 2015-10-31 | 2016-01-06 | 广州华多网络科技有限公司 | Potentiality customer recognition method and device |
| CN105760957A (en) * | 2016-02-23 | 2016-07-13 | 国元证券股份有限公司 | Securities soft lost customer prediction method |
-
2018
- 2018-12-07 CN CN201811494415.3A patent/CN109583966B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105225135A (en) * | 2015-10-31 | 2016-01-06 | 广州华多网络科技有限公司 | Potentiality customer recognition method and device |
| CN105760957A (en) * | 2016-02-23 | 2016-07-13 | 国元证券股份有限公司 | Securities soft lost customer prediction method |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110264274B (en) * | 2019-06-21 | 2023-12-29 | 深圳前海微众银行股份有限公司 | Guest group dividing method, model generating method, device, equipment and storage medium |
| CN110264274A (en) * | 2019-06-21 | 2019-09-20 | 深圳前海微众银行股份有限公司 | Objective group's division methods, model generating method, device, equipment and storage medium |
| CN112529236A (en) * | 2019-09-18 | 2021-03-19 | 泰康保险集团股份有限公司 | Target object identification method and device, electronic equipment and storage medium |
| CN110837931B (en) * | 2019-11-08 | 2023-01-31 | 中国农业银行股份有限公司 | Customer churn prediction method, device and storage medium |
| CN110837931A (en) * | 2019-11-08 | 2020-02-25 | 中国农业银行股份有限公司 | Customer churn prediction method, device and storage medium |
| CN111105266B (en) * | 2019-11-11 | 2023-10-27 | 建信金融科技有限责任公司 | Client grouping method and device based on improved decision tree |
| CN111105266A (en) * | 2019-11-11 | 2020-05-05 | 中国建设银行股份有限公司 | Client grouping method and device based on improved decision tree |
| CN111311338A (en) * | 2020-03-30 | 2020-06-19 | 网易（杭州）网络有限公司 | User value prediction method and user value prediction model training method |
| CN112529628A (en) * | 2020-12-16 | 2021-03-19 | 平安科技（深圳）有限公司 | Client label generation method and device, computer equipment and storage medium |
| CN112529628B (en) * | 2020-12-16 | 2024-04-09 | 平安科技（深圳）有限公司 | Client label generation method and device, computer equipment and storage medium |
| CN112669091A (en) * | 2021-01-06 | 2021-04-16 | 腾讯科技（深圳）有限公司 | Data processing method, device and storage medium |
| CN112669091B (en) * | 2021-01-06 | 2023-12-12 | 腾讯科技（深圳）有限公司 | Data processing method, device and storage medium |
| CN113159972A (en) * | 2021-05-20 | 2021-07-23 | 深圳前海微众银行股份有限公司 | Combination determination method, combination determination device, electronic equipment and computer readable storage medium |
| CN113159972B (en) * | 2021-05-20 | 2024-05-24 | 深圳前海微众银行股份有限公司 | Combination determination method, device, electronic equipment and computer readable storage medium |
| CN113822715A (en) * | 2021-09-23 | 2021-12-21 | 安徽数据堂科技有限公司 | Data acquisition, training and processing integrated platform analysis method |
| CN114022198A (en) * | 2021-10-27 | 2022-02-08 | 中国人民财产保险股份有限公司 | Customer value scoring method and device |
| CN116308507A (en) * | 2023-02-07 | 2023-06-23 | 网舟联合科技(北京)有限公司 | Potential customer mining method based on customer value, storage medium and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109583966B (en) | 2021-01-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109583966B (en) | High-value customer identification method, system, equipment and storage medium | |
| CN111291816B (en) | Method and device for carrying out feature processing aiming at user classification model | |
| CN108665159A (en) | A kind of methods of risk assessment, device, terminal device and storage medium | |
| CN110704730B (en) | Product data pushing method and system based on big data and computer equipment | |
| Bravo et al. | Granting and managing loans for micro-entrepreneurs: New developments and practical experiences | |
| US20200234218A1 (en) | Systems and methods for entity performance and risk scoring | |
| CN112529628B (en) | Client label generation method and device, computer equipment and storage medium | |
| CN112102073A (en) | Credit risk control method and system, electronic device and readable storage medium | |
| CN104321794A (en) | A system and method using multi-dimensional rating to determine an entity's future commercial viability | |
| US11423442B2 (en) | Method and system for predicting relevant offerings for users of data management systems using machine learning processes | |
| CN111882420A (en) | Generation method of response rate, marketing method, model training method and device | |
| CN110766481A (en) | Client data processing method and device, electronic equipment and computer readable medium | |
| CN111861521A (en) | Data processing method and device, computer readable medium and electronic equipment | |
| CN112950347A (en) | Resource data processing optimization method and device, storage medium and terminal | |
| CN115545886A (en) | Overdue risk identification method, overdue risk identification device, overdue risk identification equipment and storage medium | |
| CN112990989B (en) | Value prediction model input data generation method, device, equipment and medium | |
| CN113627997A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN117217812A (en) | User behavior prediction method and device, computer equipment and storage medium | |
| CN116821759A (en) | Identification prediction method and device for category labels, processor and electronic equipment | |
| CN114626940A (en) | Data analysis method and device and electronic equipment | |
| CN110570301B (en) | Risk identification method, device, equipment and medium | |
| CN113706258A (en) | Product recommendation method, device, equipment and storage medium based on combined model | |
| CN113537666B (en) | Evaluation model training method, evaluation and business auditing method, device and equipment | |
| CN117994017A (en) | Method for constructing retail credit risk prediction model and online credit service Scoredelta model | |
| CN118333739A (en) | Method for constructing retail credit risk prediction model and retail credit business Scoremult model |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |