CN109828969A - The processing method and system of customer data - Google Patents
The processing method and system of customer data Download PDFInfo
- Publication number
- CN109828969A CN109828969A CN201910161595.1A CN201910161595A CN109828969A CN 109828969 A CN109828969 A CN 109828969A CN 201910161595 A CN201910161595 A CN 201910161595A CN 109828969 A CN109828969 A CN 109828969A
- Authority
- CN
- China
- Prior art keywords
- customer
- data
- segmentation
- library
- database
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of processing method and system of customer data.Wherein, which includes: Data Integration module, and the historical customer data for will acquire carries out data cleansing by customer value standard and customers' credit standard, the customer data that obtains that treated；Customer segmentation database, the customer data for being obtained according to Data Integration module construct the database based on dimension and theme；Customer segmentation data-mining module obtains algorithms library, model library and the knowledge base of corresponding customer data for being finely divided excavation based on the customer data in customer segmentation database；Customer segmentation application module obtains sorted customer data for carrying out client segmentation according to application target based on customer segmentation data-mining module.The present invention is solved due to lacking the technical issues of providing comprehensively solve method for power supply enterprise's customer segmentation demand in the prior art.
Description
Technical field
The present invention relates to technical field of electricity, in particular to a kind of processing method and system of customer data.
Background technique
Traditional power customer classification method is according to electricity consumption projected demand.Common several client segmentation methods include: basis
Electricity consumption industry is divided into industry, agricultural, communications and transportation etc., and each major class can be divided into several groups again；It is divided into according to electricity consumption classification
Big industry, business, resident, farming power etc.；It is divided into first order load, two stage loads, three stage loads according to client's importance；According to
Electricity consumption size is divided into large user, responsible consumer, general user, ordinary user.Electric power enterprise provide service be also for
Upper several customer types, define different fault category and service class, are taken with response time and failure result to evaluate
Business is horizontal.
With the continuous mature and popularization and application of big data technology, there is also much gone through based on power customer currently on the market
History data, using the data digging methods such as polymerization, decision tree, neural network and big data technology carry out customer segmentation method and
It attempts, but often rests on simple technology and algorithm application problem, realize that there are gaps with reality for efficiency and classification results, no
Comprehensively solve method can be provided for power supply enterprise's customer segmentation demand.
Asking for comprehensively solve method is provided for power supply enterprise's customer segmentation demand due to lacking in the prior art for above-mentioned
Topic, currently no effective solution has been proposed.
Summary of the invention
The embodiment of the invention provides a kind of processing method and system of customer data, at least to solve due to the prior art
The technical issues of middle shortage provides comprehensively solve method for power supply enterprise's customer segmentation demand.
According to an aspect of an embodiment of the present invention, a kind of processing system of customer data is provided, comprising: Data Integration
Module, customer segmentation database, customer segmentation data-mining module and customer segmentation application module, wherein Data Integration module,
Historical customer data for will acquire carries out data cleansing by customer value standard and customers' credit standard, after obtaining processing
Customer data；Customer segmentation database, the customer data building for being obtained according to Data Integration module are based on dimension and master
The database of topic；Customer segmentation data-mining module, for being finely divided digging based on the customer data in customer segmentation database
Pick obtains algorithms library, model library and the knowledge base of corresponding customer data；Customer segmentation application module, for being based on customer segmentation
Data-mining module carries out client segmentation according to application target, obtains sorted customer data.
Optionally, Data Integration module, for historical data needed for obtaining customer segmentation from electric power data library, history number
According to from multiple operation systems；Wherein, the storage of history data P being drawn into is in systems in spatial database, and thin according to client
Fractional dimension and index progress data merging, duplicate removal delete one of attribute or statistics reconstruct or a variety of operations.
Optionally, customer segmentation database, by being carried out based on customer value according to electricity, the electricity charge and its growth rate index
It calculates, obtains customer value score, and according to promptness rate four annual arrearage number, arrearage accounting, default electricity use number, payment passes
Key index carries out the calculating of customers' credit score, obtains customers' credit score；According to customer value score and customers' credit score, obtain
The database based on dimension and theme is constructed to customer data.
Optionally, customer segmentation data-mining module is also used in the algorithms library in customer data, model library and knowledge base
More new algorithm and/or model.
Optionally, customer segmentation application module is used for according to different application target, from algorithms library, module library, knowledge base
Corresponding algorithm and/or model are selected, client segmentation is carried out, obtains sorted customer data.
According to another aspect of an embodiment of the present invention, a kind of processing method of customer data is additionally provided, comprising: will acquire
Historical customer data data cleansing is carried out by customer value standard and customers' credit standard, the client's number that obtains that treated
According to；The database based on dimension and theme is constructed according to obtained customer data；It is finely divided excavation based on customer data, is obtained
Algorithms library, model library and the knowledge base of corresponding customer data；Client segmentation is carried out based on application target, obtains sorted client
Data.
Optionally, the historical customer data that will acquire is clear by customer value standard and customers' credit standard progress data
It washes, obtaining that treated, customer data includes: historical data needed for obtaining customer segmentation from electric power data library, and historical data is come
Derived from multiple operation systems；Wherein, the storage of history data P being drawn into is tieed up according to customer segmentation in systems in spatial database
Degree and index progress data merging, duplicate removal delete one of attribute or statistics reconstruct or a variety of operations；According to electricity, the electricity charge and
Its growth rate index carries out customer value calculating, obtains customer value score；And according to annual arrearage number, arrearage accounting, disobey
About electricity consumption number, payment four key indexes of promptness rate carry out the calculating of customers' credit score, obtain customers' credit score；According to visitor
Family value score and customers' credit score obtain that treated customer data.
It further, optionally, include: foundation according to obtained database of the customer data building based on dimension and theme
Customers' credit score and customer value score determine dimension；Dimensionality reduction, dimension transformation processing are carried out to historical data according to dimension, obtained
To the database of theme.
Optionally, this method further include: more new algorithm and/or model in update algorithms library, model library and knowledge base.
Optionally, client segmentation is carried out based on application target, obtaining sorted customer data includes: according to different application
Target selects corresponding algorithm and/or model from algorithms library, module library, knowledge base, client segmentation is carried out, after obtaining classification
Customer data.
In embodiments of the present invention, by the way of big data technology, pass through Data Integration module, customer segmentation data
Library, customer segmentation data-mining module and customer segmentation application module, wherein Data Integration module, the client for will acquire
Historical data carries out data cleansing by customer value standard and customers' credit standard, the customer data that obtains that treated；Client
Subdivided data library, the customer data for being obtained according to Data Integration module construct the database based on dimension and theme；Client
Subdivided data excavates module, for being finely divided excavation based on the customer data in customer segmentation database, obtains corresponding client
Algorithms library, model library and the knowledge base of data；Customer segmentation application module, for being based on customer segmentation data-mining module foundation
Application target carries out client segmentation, obtains sorted customer data, has reached and prevented current big data field data, resource
The purpose that the client segmentation of business that wastes and lose contact with reality theorizes improves electric power application and level of customer service to realize
The technical effect of powerful support is provided, and then solves and is mentioned due to lacking in the prior art for power supply enterprise's customer segmentation demand
The technical issues of for comprehensively solve method.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes part of this application, this hair
Bright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.In the accompanying drawings:
Fig. 1 is the structural schematic diagram of the processing system of customer data according to an embodiment of the present invention；
Fig. 2 is the schematic diagram of traditional classification in the prior art；
Fig. 3 is the schematic diagram for using Cluster Classification in the prior art；
Fig. 4 is the schematic diagram of classification dimension in the processing system of customer data according to an embodiment of the present invention；
Fig. 5 is the structural schematic diagram of the processing system of another customer data according to an embodiment of the present invention；
Fig. 6 is the process signal of customer segmentation in the processing system of another customer data according to an embodiment of the present invention
Figure；
Fig. 7 is a kind of flow diagram of the processing method of customer data according to an embodiment of the present invention.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people
The model that the present invention protects all should belong in member's every other embodiment obtained without making creative work
It encloses.
It should be noted that description and claims of this specification and term " first " in above-mentioned attached drawing, "
Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way
Data be interchangeable under appropriate circumstances, so as to the embodiment of the present invention described herein can in addition to illustrating herein or
Sequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that cover
Cover it is non-exclusive include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to
Step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, product
Or other step or units that equipment is intrinsic.
Embodiment one
According to an aspect of an embodiment of the present invention, a kind of processing system of customer data is provided, Fig. 1 is according to this hair
The structural schematic diagram of the processing system of the customer data of bright embodiment, as shown in Figure 1, comprising:
Data Integration module 12, customer segmentation database 14, customer segmentation data-mining module 16 and customer segmentation application
Module 18, wherein Data Integration module 12, the historical customer data for will acquire pass through customer value standard and customers' credit
Standard carries out data cleansing, the customer data that obtains that treated；Customer segmentation database 14, for according to Data Integration module 12
Obtained database of the customer data building based on dimension and theme；Customer segmentation data-mining module 16, for being based on client
Customer data in subdivided data library 14 is finely divided excavation, obtains algorithms library, model library and the knowledge base of corresponding customer data；
Customer segmentation application module 18 is obtained for carrying out client segmentation according to application target based on customer segmentation data-mining module 16
To sorted customer data.
In embodiments of the present invention, by the way of big data technology, pass through Data Integration module, customer segmentation data
Library, customer segmentation data-mining module and customer segmentation application module, wherein Data Integration module, the client for will acquire
Historical data carries out data cleansing by customer value standard and customers' credit standard, the customer data that obtains that treated；Client
Subdivided data library, the customer data for being obtained according to Data Integration module construct the database based on dimension and theme；Client
Subdivided data excavates module, for being finely divided excavation based on the customer data in customer segmentation database, obtains corresponding client
Algorithms library, model library and the knowledge base of data；Customer segmentation application module, for being based on customer segmentation data-mining module foundation
Application target carries out client segmentation, obtains sorted customer data, has reached and prevented current big data field data, resource
The purpose that the client segmentation of business that wastes and lose contact with reality theorizes improves electric power application and level of customer service to realize
The technical effect of powerful support is provided, and then solves and is mentioned due to lacking in the prior art for power supply enterprise's customer segmentation demand
The technical issues of for comprehensively solve method.Optionally, Data Integration module, for being obtained needed for customer segmentation from electric power data library
Historical data, historical data derive from multiple operation systems；Wherein, the storage of history data P being drawn into systems between data
In library, and data merging, duplicate removal carried out according to customer segmentation dimension and index, deletes attribute or one of statistics reconstruct or a variety of
Operation.
Optionally, customer segmentation database 14, by being carried out based on customer value according to electricity, the electricity charge and its growth rate index
It calculates, obtains customer value score, and according to promptness rate four annual arrearage number, arrearage accounting, default electricity use number, payment passes
Key index carries out the calculating of customers' credit score, obtains customers' credit score；According to customer value score and customers' credit score, obtain
The database based on dimension and theme is constructed to customer data.
Optionally, customer segmentation data-mining module 16 is also used to the algorithms library in customer data, model library and knowledge base
In more new algorithm and/or model.
Optionally, customer segmentation application module 18 is used for according to different application target, from algorithms library, module library, knowledge base
It is middle to select corresponding algorithm and/or model, client segmentation is carried out, sorted customer data is obtained.
To sum up, the processing system of customer data provided by the embodiments of the present application is specific as follows:
As shown in Fig. 2, Fig. 2 is the schematic diagram of traditional classification in the prior art, wherein client segmentation traditional at present uses
Several dimensions: industry, agricultural, communications and transportation etc. are divided into according to electricity consumption industry, each major class can be divided into several groups again；According to
Electricity consumption classification is divided into big industry, business, resident, farming power etc.；According to client's importance be divided into first order load, two stage loads,
Three stage loads；It is divided into large user, responsible consumer, general user, ordinary user according to electricity consumption size.Using conventional electric power client
Classification method cannot provide targetedly marketing strategy and service for different type client, be unable to satisfy fining customer account management
Business demand needs further to segment customer type.
Meanwhile as shown in figure 3, Fig. 3 is the schematic diagram for using Cluster Classification in the prior art, wherein using big data
Technology carries out the trial of power customer classification, is classified using cluster, decision tree, neural network scheduling algorithm, classification results root
It is larger according to the mining algorithm and model difference of use.Since classification uses big data technology and mass historical data as complete or collected works' sample
This, the generally existing lower problem of classification effectiveness, and also this classification method core is to be based on mass historical data and algorithm model,
Fail sufficiently to combine power business demand, final classification result and real business error are larger, cannot be in real business extensively
It uses.
And the embodiment of the present application carries out customer value classification on the basis of real business demand first, meets current business
Client segmentation basal needs, and customers' credit model is constructed on the basis of this, credit scoring is carried out to all clients.Customer value point
Two dimensions are worth for current value and potentiality, again according to electricity consumption type, period electricity, the period electricity charge, history under each dimension
The indexs such as growth rate are classified, including key customer, big customer, Very Important Person, common customer, other clients.Credit Model
Using the indexs such as multiple indexs, including arrearage number, arrearage accounting, electricity consumption number lack of standardization, judicial dispute number, credit result
It is indicated using percentage value, client is divided by VIP client, credit client, top-tier customer, common customer, low letter according to credit score
With examination, dangerous client, as shown in figure 4, Fig. 4 be customer data according to an embodiment of the present invention processing system in classify dimension
Schematic diagram.
As shown in figure 5, Fig. 5 is the structural representation of the processing system of another customer data according to an embodiment of the present invention
The processing system of figure, customer data provided by the embodiments of the present application uses big data technology, is based on mass historical data, and data are come
Derived from Electric Power Marketing System, acquisition system, 95598 customer service systems, PMIS system etc., by Data Integration, building data warehouse,
Data mining, classification application process realize that client segmentation and application, two layers of the client segmentation needed by real business avoid
The poor efficiency and the more drawback of classification results invalid data of mass data processing, classification results can provide reality for power business
With reference to and application, prevented current big data field data, the wasting of resources and the client segmentation for the business that loses contact with reality and theorized and ask
Topic provides powerful support to improve electric power application and level of customer service.
As shown in figure 5, the processing system composition of customer data provided by the embodiments of the present application is as follows:
One, Data Integration module: data duplication or extraction, data mart modeling and conversion function are realized, it is therefore an objective to from electric power number
Historical data needed for obtaining customer segmentation according to library, data source is in multiple operation systems, as needed, is replicated using database
The mode that the ETL tool such as technology or Kettle extracts obtains.The data being drawn into are stored in this system intermediate database, according to
Customer segmentation dimension and index carry out data merging, duplicate removal, delete the operations such as attribute, statistics reconstruct, complete the preparation of basic data
Work.
Two, customer segmentation data warehouse: the data cleaned are carried out according to two dimensions of customer value and customers' credit
Subdivision constructs Data Mart, data warehouse based on dimension and theme.
Customer value algorithm:
Customer value calculating is carried out according to electricity, the electricity charge and its growth rate index, client's score is calculated, to different electricity consumption classes
Type client classifies, and type includes:
VIP client definition: M_PQ > M1and RPQ > R1OR LD1=1
Key customer's definition: M_PQ > M2and RPQ > R2OR LD1=2
Big customer's definition: M_PQ > M3and RPQ > R3；
Common customer definition: M_PQ > M4and RPQ > R4；
Other client definitions: M_PQ > M5and RPQ > R5；
M1:VIP client corresponds to charge level；
M2: key customer corresponds to charge level；
M3: big customer corresponds to charge level；
M4: common customer corresponds to charge level；
M5: other clients correspond to charge level；
R1:VIP client corresponds to electricity measuring growth rate；
R2: key customer corresponds to electricity measuring growth rate；
R3: big customer corresponds to electricity measuring growth rate；
R4: common customer corresponds to electricity measuring growth rate；
R5: other clients correspond to electricity measuring growth rate；
First order load: LD1；
Two stage loads: LD2；
Three stage loads: LD3；
Customer value algorithm is as follows:
Annual monthly electricity (the M_PQ)=Sum (each month of prior year electricity)/12 of client
Client's year electricity growth rate (the RPQ)=above client's valence of (last year total electricity-the year before last total electricity)/last year total electricity
The citing of value-based algorithm calculating process:
Certain commercial user Zhang San 2014,2 years 2015 each moon electricity, electricity charge data are shown in Table 1: table 1
VIP client corresponds to electricity: the Wan Du of M1 > 1,000,000；
Key customer corresponds to electricity: 1,000,000 degree < M2 < 10,000,000 degree；
Big customer corresponds to 10,000 degree < M3 < 1,000,000 degree of electricity；
Common customer corresponds to 500 degree < M4 < 10,000 degree of electricity；
Other clients correspond to electricity M5 < 500 degree electricity；
VIP client corresponds to 0.8 < R1 of electricity measuring growth rate；
Key customer corresponds to electricity measuring growth rate 0.5 < R2 < 0.8；
Big customer corresponds to electricity measuring growth rate 0.3 < R3 < 0.5；
Common customer corresponds to electricity measuring growth rate 0.05 < R4 < 0.3；
Other clients correspond to electricity measuring growth rate R5 < 0.05；
The enterprise customer belongs to three-level；
Calculating process is as follows:
In monthly, 2014 electricity (M_PQ)=11470/12=955.8
In monthly, 2015 electricity (M_PQ)=17039/12=1419.9
2015 annual electricity growth rate (RPQ)=(17039-11470)/11470=0.48553
According to the above calculated result and customer value definition rule, 10,000 degree < year monthly electricity (1.147 ten thousand degree) < 100
Ten thousand degree of and 0.3 < annual electricity growth rate (0.486) < 0.5, this user belong to big customer's type.
Customers' credit algorithm:
Client is carried out according to annual arrearage number, arrearage accounting, default electricity use number, payment four key indexes of promptness rate
CREDIT SCORE calculates, and classifies to different credit score users, with reference to external credit classification standard, be divided into AAA, AA, A,
Nine credit grades of BBB, BB, B, CCC, CC, C.Credit score algorithm is as follows:
Client's year arrearage number (Y_DT)=Sum (current year each moon arrearage number)
Client's year arrearage accounting (RY_DT)=| H11 |/| H12 | * 100；
(H11: the prior year arrearage amount of money, H12: the sum of prior year each moon electricity charge)
Client's year default electricity use number (Y_BL)=Sum (current year each moon default electricity use number)；
Client averagely pays the fees the period (Y_TM)=AVG year (single is paid the fees the date-single payment date in time limit)；
Credit score=100-Y_DT*20*W1-RY_DT*100*W2-Y_BL*50*W3-Y_TM*W4；
W1, W2, W3, W4 are index coefficient, and size is between 0 and 1.
Credit rating definition rule is as follows:
(1) AAA grade score range: >=95
(2) AA grade score range:<95and>=85
(3) A grade score range:<85and>=75
(4) BBB grade score range:<75and>=65
(5) BB grade score range:<60and>=50
(6) B grade score range:<50and>=40
(7) CCC grade score range:<40and>=30
(8) CC grade score range:<30and>=20
(9) C grade score range: < 20
Credit rating calculated example is as follows:
For example, using Zhang San user in example from above, 2015 annual arrearage numbers (Y_DT) 4 times；It breaks a contract in client's year use
Electric number (Y_BL) 1 time；Client averagely pays the fees the period (Y_TM) 7 days year；August and electricity charge arrearage in October are not paid.
W1=0.3, W2=0.3, W3=0.2, W4=0.2, as shown in table 1:
Client's year arrearage accounting (RY_DT)=(1243.34+1119.80)/15359.11*100=15.39 in 2015
Credit score=100-4*20*W1-15.39*W2-1*50*W3-7*W4=59.98
Because of 50 < credit score (59.98) < 60, then commercial user's credit grade is BB rank.
Three, customer segmentation data-mining module:
Including customer segmentation algorithms library, model library, knowledge base, algorithm newly developed, model can also be added to algorithm
Customer segmentation relevant knowledge is stored in library, model library, in knowledge base and applies result.
Four, customer segmentation application module:
According to different application target, corresponding algorithm, model are selected from algorithms library, module library, knowledge base, carry out client
Classification.
Customer segmentation model is as shown in table 2:
Table 2
According to user demand, it can establish multi -index system and carry out customer segmentation.
According to the commercial user all data in example above, customer value belongs to big customer's type, credit grade BB
Grade, this client of integrated application are classified divided method, which belongs to the client that the credit grade under big customer's type is BB grades
As shown in table 3.
Table 3
As shown in fig. 6, Fig. 6 is customer segmentation in the processing system of another customer data according to an embodiment of the present invention
Flow diagram, shown in customer segmentation overall procedure Fig. 6:
Process is made of five steps.
Step 1, source data.Analysis, obtain source historical data, data source systems include marketing system (client's essential information,
Electricity, electricity charge information), 95598 customer service systems (Customer Service Information), PMIS operation system (O&M information on services), acquisition system
It unites (electricity consumption essential information) etc..
Step 2, data pick-up, processing.According to classification demand, data are extracted from electric power application system historical data, into
Row data cleansing, processing, form available data sources.
Step 3, database is constructed.Based on processing, treated available data sources, according to the requirement of customer segmentation,
The processing such as dimensionality reduction, dimension transformation is carried out to data, building forms customer segmentation database.Customer value, customers' credit are bases
Dimension.
Step 4, data mining.On the basis of data warehouse, carry out further data mining, customer value by electricity,
The subdivision of the dimensions such as electricity price, electricity consumption history is excavated, and customers' credit is carried out according to arrearage accounting, arrearage number, violation electricity consumption number, contract
Market condition carries out dimension subdivision and excavates.
Step 5, client segmentation.Client segmentation is carried out according to customer value and customers' credit dimension respectively.
Step 6, customer segmentation.Analysis is compared to customer value and the sorted data of customers' credit dimension, according to
Customer value and customers' credit are carried out weight distribution by classification results and the practical business goodness of fit, according to weight distribution as a result, weight
It is new to carry out customer segmentation, interpretation of result is carried out later, adjusts weighted value again, until classification results and practical business situation are kissed
It closes, the customer segmentation result obtained is actual customer subdivision type.
Customer electricity situation changes frequent occurrence, and customer segmentation result is also required to dynamic and adjusts, can with big data technology
To excavate, analyze to the mass data under two dimensions of customer value and client's weight, provided really for Utilities Electric Co.
The drawbacks of customer type, customer type and actual conditions are not inconsistent before overcoming, lays for client's precision marketing and orientation service
Good basis.
The processing system of customer data provided by the embodiments of the present application uses big data technology, is based on mass historical data,
Data source passes through Data Integration, building number in Electric Power Marketing System, acquisition system, 95598 customer service systems, PMIS system etc.
Client segmentation and application are realized according to warehouse, data mining, classification application process, two layers of the client point needed by real business
Class avoids the poor efficiency and the more drawback of classification results invalid data of mass data processing, and classification results can be electric power industry
Business provides real reference and application, has prevented the client point of current big data field data, the wasting of resources and the business that loses contact with reality
Class theorizes problem, provides powerful support to improve electric power application and level of customer service.
The processing system of customer data provided by the embodiments of the present application passes through building customers' credit system, is with customers' credit
Dimension carries out customer segmentation.Traditional power customer classification method has been unable to meet power consumer according to electricity consumption projected demand
Propertyization and requirement of taking the initiative in offering a hand.
The processing system of customer data provided by the embodiments of the present application uses comprehensively according to electricity market client segmentation demand
Big data technology, in conjunction with power supply enterprise's reality business demand customizing model, is extracted based on power customer mass historical data
Data are classified first with the power customer property of value as analysis foundation, on this basis, construct customers' credit system,
Customer segmentation is carried out by dimension of customers' credit, customer segmentation result can be formulated with marketing strategy and provide foundation.
It should be noted that the application above-mentioned example is only to realize the processing system of customer data provided by the embodiments of the present application
Subject to system, specifically without limitation.
Embodiment two
According to another aspect of an embodiment of the present invention, a kind of processing method of customer data is additionally provided, Fig. 7 is according to this
A kind of flow diagram of the processing method of customer data of inventive embodiments, as shown in fig. 7, comprises:
Step S702, the historical customer data that will acquire are clear by customer value standard and customers' credit standard progress data
It washes, the customer data that obtains that treated；
Step S704 constructs the database based on dimension and theme according to obtained customer data；
Step S706 is finely divided excavation based on customer data, obtains corresponding to the algorithms library of customer data, model library and know
Know library；
Step S708 carries out client segmentation based on application target, obtains sorted customer data.
Optionally, the historical customer data that will acquire in step S702 by customer value standard and customers' credit standard into
Row data cleansing, obtaining that treated, customer data includes: historical data needed for obtaining customer segmentation from electric power data library, is gone through
History data source is in multiple operation systems；Wherein, the storage of history data P being drawn into is in systems in spatial database, and according to visitor
Family subdivision dimension and index progress data merging, duplicate removal delete one of attribute or statistics reconstruct or a variety of operations；According to electricity
Amount, the electricity charge and its growth rate index carry out customer value calculating, obtain customer value score；And according to annual arrearage number, owe
Take accounting, default electricity use number, payment four key indexes of promptness rate and carry out the calculating of customers' credit score, obtains customers' credit and obtain
Point；According to customer value score and customers' credit the score customer data that obtains that treated.
Further, optionally, according to obtained data of the customer data building based on dimension and theme in step S704
Library includes: to determine dimension according to customers' credit score and customer value score；Dimensionality reduction, dimension are carried out to historical data according to dimension
Conversion process obtains the database of theme.
Optionally, the processing method of customer data provided by the embodiments of the present application further include: update algorithms library, model library and
More new algorithm and/or model in knowledge base.
Optionally, step S708 is based on application target and carries out client segmentation, and obtaining sorted customer data includes: basis
Different application target selects corresponding algorithm and/or model from algorithms library, module library, knowledge base, carries out client segmentation, obtains
To sorted customer data.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
In the above embodiment of the invention, it all emphasizes particularly on different fields to the description of each embodiment, does not have in some embodiment
The part of detailed description, reference can be made to the related descriptions of other embodiments.
In several embodiments provided herein, it should be understood that disclosed technology contents can pass through others
Mode is realized.Wherein, the apparatus embodiments described above are merely exemplary, such as the division of the unit, Ke Yiwei
A kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or components can combine or
Person is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Between coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or communication link of unit or module
It connects, can be electrical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
On unit.It can some or all of the units may be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can for personal computer, server or network equipment etc.) execute each embodiment the method for the present invention whole or
Part steps.And storage medium above-mentioned includes: that USB flash disk, read-only memory (ROM, Read-OnlyMemory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic or disk etc. be various to can store program code
Medium.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (10)
1. a kind of processing system of customer data characterized by comprising
Data Integration module, customer segmentation database, customer segmentation data-mining module and customer segmentation application module, wherein
The Data Integration module, historical customer data for will acquire by customer value standard and customers' credit standard into
Row data cleansing, the customer data that obtains that treated；
The customer segmentation database, the customer data building for being obtained according to the Data Integration module are based on dimension
With the database of theme；
The customer segmentation data-mining module is thin for being carried out based on the customer data in the customer segmentation database
Divide and excavate, obtains the algorithms library, model library and the knowledge base that correspond to the customer data；
The customer segmentation application module, for carrying out client according to application target based on the customer segmentation data-mining module
Classification, obtains sorted customer data.
2. system according to claim 1, which is characterized in that the Data Integration module, for being obtained from electric power data library
Historical data needed for obtaining customer segmentation, the historical data derive from multiple operation systems；Wherein, the history being drawn into
Data are stored in system intermediate database, and data merging, duplicate removal are carried out according to customer segmentation dimension and index, delete attribute or
One of statistics reconstruct or a variety of operations.
3. system according to claim 1, which is characterized in that the customer segmentation database, for according to electricity, the electricity charge
And its growth rate index carries out customer value calculating, obtains customer value score, and according to annual arrearage number, arrearage accounting,
Default electricity use number, payment four key indexes of promptness rate carry out the calculating of customers' credit score, obtain customers' credit score；Foundation
The customer value score and the customers' credit score obtain the data of the customer data building based on dimension and theme
Library.
4. system according to claim 1, which is characterized in that the customer segmentation data-mining module is also used in institute
State more new algorithm and/or model in algorithms library, model library and the knowledge base of customer data.
5. system according to claim 1, which is characterized in that the customer segmentation application module, for being answered according to difference
With target, corresponding algorithm and/or model are selected from the algorithms library, module library, knowledge base, are carried out client segmentation, are obtained
The sorted customer data.
6. a kind of processing method of customer data characterized by comprising
The historical customer data that will acquire carries out data cleansing by customer value standard and customers' credit standard, after obtaining processing
Customer data；
The database based on dimension and theme is constructed according to the obtained customer data；
It is finely divided excavation based on the customer data, obtains the algorithms library, model library and the knowledge base that correspond to the customer data；
Client segmentation is carried out based on application target, obtains sorted customer data.
7. according to the method described in claim 6, it is characterized in that, the historical customer data that will acquire passes through customer value
Standard and customers' credit standard carry out data cleansing, and obtaining that treated, customer data includes:
Historical data needed for obtaining customer segmentation from electric power data library, the historical data derive from multiple operation systems；Its
In, the storage of history data P being drawn into is counted according to customer segmentation dimension and index in systems in spatial database
According to merging, duplicate removal, delete one of attribute or statistics reconstruct or a variety of operations；
Customer value calculating is carried out according to electricity, the electricity charge and its growth rate index, obtains customer value score；
And client's letter is carried out according to annual arrearage number, arrearage accounting, default electricity use number, payment four key indexes of promptness rate
It is calculated with score, obtains customers' credit score；
Treated the customer data is obtained according to the customer value score and the customers' credit score.
8. method according to claim 6 or 7, which is characterized in that the customer data that the basis obtains constructs base
Include: in the database of dimension and theme
The dimension is determined according to customers' credit score and customer value score；
Dimensionality reduction, dimension transformation processing are carried out to historical data according to the dimension, obtain the database of the theme.
9. according to the method described in claim 6, it is characterized in that, the method also includes:
Update more new algorithm and/or model in the algorithms library, model library and knowledge base.
10. according to the method described in claim 6, it is characterized in that, it is described based on application target carry out client segmentation, divided
Customer data after class includes:
According to different application target, corresponding algorithm and/or model are selected from the algorithms library, module library, knowledge base, are carried out
Client segmentation obtains the sorted customer data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910161595.1A CN109828969A (en) | 2019-03-04 | 2019-03-04 | The processing method and system of customer data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910161595.1A CN109828969A (en) | 2019-03-04 | 2019-03-04 | The processing method and system of customer data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109828969A true CN109828969A (en) | 2019-05-31 |
Family
ID=66865209
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910161595.1A Pending CN109828969A (en) | 2019-03-04 | 2019-03-04 | The processing method and system of customer data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109828969A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150161233A1 (en) * | 2013-12-11 | 2015-06-11 | The Board Of Trustees Of The Leland Stanford Junior University | Customer energy consumption segmentation using time-series data |
| CN106067137A (en) * | 2016-06-28 | 2016-11-02 | 中国地质大学（北京） | A kind of credit rating evaluation methodology based on intelligent grid monitoring system |
| CN106355518A (en) * | 2016-11-29 | 2017-01-25 | 国网山东省电力公司电力科学研究院 | Electricity fee payment customer screening method and system |
| CN106780140A (en) * | 2016-12-15 | 2017-05-31 | 国网浙江省电力公司 | Electric power credit assessment method based on big data |
| CN108280541A (en) * | 2018-01-11 | 2018-07-13 | 国网山东省电力公司 | Customer service strategies formulating method, device based on random forest and decision tree |
-
2019
- 2019-03-04 CN CN201910161595.1A patent/CN109828969A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150161233A1 (en) * | 2013-12-11 | 2015-06-11 | The Board Of Trustees Of The Leland Stanford Junior University | Customer energy consumption segmentation using time-series data |
| CN106067137A (en) * | 2016-06-28 | 2016-11-02 | 中国地质大学（北京） | A kind of credit rating evaluation methodology based on intelligent grid monitoring system |
| CN106355518A (en) * | 2016-11-29 | 2017-01-25 | 国网山东省电力公司电力科学研究院 | Electricity fee payment customer screening method and system |
| CN106780140A (en) * | 2016-12-15 | 2017-05-31 | 国网浙江省电力公司 | Electric power credit assessment method based on big data |
| CN108280541A (en) * | 2018-01-11 | 2018-07-13 | 国网山东省电力公司 | Customer service strategies formulating method, device based on random forest and decision tree |
Non-Patent Citations (2)
| Title |
|---|
| 卢海明 等: ""电力客户细分及增值服务系统研究与应用"", 《华北电力技术》 * |
| 殷博益: "《市场营销学 第2版》", 31 August 2012, 东南大学出版社 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109063945A (en) | A kind of 360 degree of customer portrait construction methods of sale of electricity company based on Value accounting system | |
| CN108280541A (en) | Customer service strategies formulating method, device based on random forest and decision tree | |
| CN105404944A (en) | Big data analysis method for warning of heavy-load and overload of electric power system | |
| CN108596467B (en) | Market operation simulation transaction simulation system suitable for electricity selling company | |
| CN111724039B (en) | Recommendation method for recommending customer service personnel to power users | |
| CN108388955A (en) | Customer service strategies formulating method, device based on random forest and logistic regression | |
| CN107578277B (en) | Rental house client positioning method for electric power marketing | |
| CN112308462A (en) | Power consumer classification method and device | |
| CN109636482B (en) | Data processing method and system based on similarity model | |
| CN108846555A (en) | A kind of efficient accurate enthesis of electric load big data missing values | |
| CN104346698A (en) | Catering member big data analysis and checking system based on cloud computing and data mining | |
| CN101320449A (en) | Power distribution network estimation method based on combination appraisement method | |
| CN110427418A (en) | Customer analysis grouping method based on customer energy value index system | |
| CN107977855B (en) | Method and device for managing user information | |
| Ripoll-Zarraga et al. | Exploring the reasons for efficiency in Spanish airports | |
| CN113205408A (en) | Customer manager capacity map generation method and device | |
| CN104915388B (en) | It is a kind of that method is recommended based on spectral clustering and the book labels of mass-rent technology | |
| CN109492055B (en) | A method of power grid big data warehouse is established based on CIM model | |
| CN109522934A (en) | A kind of power consumer clustering method based on clustering algorithm | |
| CN112580978B (en) | Power market member credit evaluation and credit label generation method | |
| Richardson et al. | An analysis of seasonal household waste generation | |
| CN109828969A (en) | The processing method and system of customer data | |
| CN107767278B (en) | Method and device for constructing community hierarchy | |
| CN106682205A (en) | Device and method for data processing | |
| CN113538011A (en) | Method for associating non-registered contact information with registered user in power system |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190531 |