CN110490629A - A kind of TP-RFM modeling method for corporate client value assessment - Google Patents
A kind of TP-RFM modeling method for corporate client value assessment Download PDFInfo
- Publication number
- CN110490629A CN110490629A CN201910172131.0A CN201910172131A CN110490629A CN 110490629 A CN110490629 A CN 110490629A CN 201910172131 A CN201910172131 A CN 201910172131A CN 110490629 A CN110490629 A CN 110490629A
- Authority
- CN
- China
- Prior art keywords
- transaction
- rfm
- index
- value assessment
- client
- 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
- 238000000034 method Methods 0.000 title claims abstract description 22
- 239000011159 matrix material Substances 0.000 claims description 7
- 238000013139 quantization Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000012821 model calculation Methods 0.000 abstract description 3
- 238000004458 analytical method Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Technology Law (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of TP-RFM modeling methods for corporate client value assessment, carry out curve fitting respectively to transaction cycle T, average transaction amount P, the last exchange hour interval R, transaction count F and transaction total amount M, obtain fitting function;Later, the weight coefficient of each index is calculated, and weighted sum obtains the TP-RFM model calculation formula of customer value assessment;Finally, the comprehensive score of each client can be calculated according to formula.The present invention models user using transaction data, is analyzed from different dimensions, accurately realizes the assessment of customer value, and enterprise is helped to formulate corresponding service strategy for the client in different value stages.
Description
Technical field
The present invention relates to customer value assessment technology field, specially a kind of TP-RFM for corporate client value assessment
Modeling method.
Background technique
Customer value assesses the client that enterprise can be helped to excavate high value, for the client in different phase value
Formulate corresponding service strategy.
Traditional RFM model is mainly used to describe the moneyness of client, the ranking without comprehensive value after clearly quantifying, and
It is confined to the last transaction tri- R, transaction count F and transaction amount M dimensions.What more existing pair of RFM model improved
Model goes analysis to assess from more dimensions, but has all lacked the analysis of the autocorrelation between dimension and dimension.
Summary of the invention
The purpose of the present invention is to provide a kind of TP-RFM modeling methods for corporate client value assessment, on solving
State the problem of proposing in background technique.
To achieve the above object, the invention provides the following technical scheme: a kind of TP- for corporate client value assessment
RFM modeling method, it is total to transaction cycle T, average transaction amount P, the last exchange hour interval R, transaction count F and transaction
Amount of money M carries out curve fitting respectively, and is weighted summation to each index to realize the assessment of customer value, including walk as follows
It is rapid:
A, it extracts between the transaction cycle T of each client in transaction data, average transaction amount P, the last exchange hour
Every R, transaction count F and transaction five indexs of total amount M;
B, to each index, some points is chosen and carry out quantization marking, according to the trend of these hash points, chosen respectively corresponding
Fitting function, parameter is found out based on the principle of least square, obtains fitting function T (f, t), P (f, p), R (r), F (f), M (m),
And then obtain score of all clients in each index;
C, to each index scoring criteria,xsAfter x criterion
Score, smaxFor the best result after standardization, x (f, t)maxFor maximum value of all clients in x index score;
D, the corresponding weight coefficient w of T, P, R, F, M index is calculated using analytic hierarchy process (AHP)T、wP、wR、wF、wM, finally obtain
It is s=∑ w that customer value, which assesses calculation formula,xxs(x=T, P, R, F, M).
Preferably, in the step A, transaction cycle T is the day of the time that client trades for the first time and current time interval
Number is poor, divided by transaction count.
Preferably, in the step B, when choosing fitting function, due to transaction cycle and average transaction value and transaction time
Number has close relationship, so selection binary function T (f, t), P (f, p) are fitted, wherein f is transaction count.
Preferably, in the step B, the number of fitting function parameter is not more than the number of the point of quantization marking.
Preferably, in the step D using analytic hierarchy process (AHP) calculate T, P, R, the weight coefficient of F, M, the specific steps of which are as follows:
A, to transaction cycle T, average transaction amount P, the last exchange hour interval R, transaction count F and the total gold of transaction
Volume M two compare, and establish judgment matrix;
B, weight coefficient vector is sought;
C, consistency check.
Preferably, in the step D, formula s=∑ w is pressed in the customer value assessmentxxs(x=T, P, R, F, M) weighting is asked
With obtain, wherein wxFor the weight of x index, xsFor the score after x criterion.
Compared with prior art, the beneficial effects of the present invention are: the present invention is to transaction cycle T, average transaction amount P, most
Nearly exchange hour interval R, transaction count F and transaction total amount M carry out curve fitting respectively, obtain fitting function;Later,
The weight coefficient of each index is calculated, and weighted sum obtains the TP-RFM model calculation formula of customer value assessment;Finally, according to public affairs
Formula can calculate the comprehensive score of each client.The present invention models user using transaction data, from different dimensions into
Row analysis, accurately realizes the assessment of customer value, and enterprise is helped to formulate corresponding service for the client in different value stages
Strategy.
Detailed description of the invention
Fig. 1 is flow chart of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Referring to Fig. 1, the present invention provides a kind of technical solution: a kind of TP-RFM modeling for corporate client value assessment
Method, comprising the following steps:
A, it extracts between the transaction cycle T of each client in transaction data, average transaction amount P, the last exchange hour
Every R, transaction count F and transaction M5 index of total amount;
B, each index is proceeded as follows: chooses n discrete point x (x1,x2,...,xn) carry out quantifying the y that gives a mark to obtain
(y1,y2,...,yn);It chooses fitting function y=f (x, C) --- (1), wherein C (c1,c2,…,cm) it is that m needs pass through calculating
Determining parameter, and m≤n;By (xi,yi) bring into formula (1), it obtains: yi=f (xi, C) --- (2), as m=n, m side of simultaneous
Journey solves parameter C (c1,c2,…,cm) value;It as m < n, is handled using least square method, quantization score value surrounds desired value f
(x, C) is swung, and is normal distribution, then yiProbability density beWherein σi 2For
Standard error considers that the quantization score of each client is independent from each other, therefore y (y1,y2,...,yn) likelihood functionLikelihood function L maximum is taken to estimate parameter C, so thatIt is minimized, formula (5) is least square method criterion, is had according to the requirement of formula (5)To obtain equation groupSolving equations (6), obtain the estimated value of parameterThe curvilinear equation being fitted
C, each index curvilinear equation according to obtained in step B calculates each index score value of all clients, using percentage
System is standardized,WhereinFor curvilinear equationIn domain most
Big value, s are the score after standardization;
D, each index weight coefficient is determined using analytic hierarchy process (AHP);
E, comprehensive score of summing to obtain is weighted to each indexWherein wiFor the weight coefficient of index i, siFor
Score after index i standardization, i=1, when 2 ..., 5 respectively represent transaction cycle T, averagely transaction amount P, the last time trades
Between be spaced R, transaction count F and transaction total amount M.
In the present invention, specific step is as follows by step D:
A, judgment matrix is established,Wherein
aij(i, j=1,2 ..., 5;) index i is indicated compared to the importance with index j, 1 is no less important, and 3 be slightly heavy
Want, 5 is obvious important, and 7 is strong important, 9 be it is extremely important, 2,4,6,9 i.e. in the intermediate value of above-mentioned two adjacent judgements, in addition,
B, the feature vector of matrix A is sought, and obtains weight coefficient vector after normalizing are as follows:
W=(w1,w2,w3,w4,w5)T, wherein wi(i=1,2 ..., 5) is the opposite weight coefficient of different indexs.
C, consistency check, the specific steps are as follows: calculate consistency ratioWherein CI is coincident indicator,λmaxFor the Maximum characteristic root of judgment matrix A, n is the order of judgment matrix A, and RI is random index,
Table look-up can obtain n be 5 when, RI=1.12;As consistency ratio CI < 0.1, by examining, weight coefficient vector w=(w is obtained1,
w2,w3,w4,w5)T, judgment matrix A is otherwise adjusted, b is repeated, c is until pass through consistency check.
In conclusion the present invention is to transaction cycle T, average transaction amount P, the last exchange hour interval R, transaction time
Number F and transaction total amount M carry out curve fitting respectively, obtain fitting function;Later, the weight coefficient of each index is calculated, and is added
Power sum customer value assessment TP-RFM model calculation formula;Finally, the synthesis of each client can be calculated according to formula
Score.The present invention models user using transaction data, is analyzed from different dimensions, accurately realizes customer value
Assessment, help enterprise to formulate corresponding service strategy for the client in different value stages.
It although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
A variety of variations, modification, replacement can be carried out to these embodiments without departing from the principles and spirit of the present invention by understanding
And modification, the scope of the present invention is defined by the appended.
Claims (6)
1. a kind of TP-RFM modeling method for corporate client value assessment, to transaction cycle T, averagely transaction amount P, recently
Exchange hour interval R, a transaction count F and transaction total amount M carry out curve fitting respectively, and are weighted and ask to each index
With the assessment to realize customer value, characterized by the following steps:
A, extract the transaction cycle T of each client in transaction data, average transaction amount P, the last exchange hour interval R,
Transaction count F and transaction five indexs of total amount M;
B, to each index, some points is chosen and carry out quantization marking, according to the trend of these hash points, chosen respectively corresponding quasi-
Function is closed, parameter is found out based on the principle of least square, obtains fitting function T (f, t), P (f, p), R (r), F (f), M (m), in turn
Obtain score of all clients in each index;
C, to each index scoring criteria,xsFor obtaining after x criterion
Point, smaxFor the best result after standardization, x (f, t)maxFor maximum value of all clients in x index score;
D, the corresponding weight coefficient w of T, P, R, F, M index is calculated using analytic hierarchy process (AHP)T、wP、wR、wF、wM, finally obtain client
Value assessment calculation formula is s=∑ wxxs(x=T, P, R, F, M).
2. a kind of TP-RFM modeling method for corporate client value assessment according to claim 1, it is characterised in that:
In the step A, transaction cycle T is that the time of transaction and the number of days of current time interval are poor for the first time by client, divided by transaction time
Number.
3. a kind of TP-RFM modeling method for corporate client value assessment according to claim 1, it is characterised in that:
In the step B, when choosing fitting function, since transaction cycle and average transaction value and transaction count have close relationship,
So selection binary function T (f, t), P (f, p) are fitted, wherein f is transaction count.
4. a kind of TP-RFM modeling method for corporate client value assessment according to claim 1, it is characterised in that:
In the step B, the number of fitting function parameter is not more than the number of the point of quantization marking.
5. a kind of TP-RFM modeling method for corporate client value assessment according to claim 1, it is characterised in that:
In the step D using analytic hierarchy process (AHP) calculate T, P, R, the weight coefficient of F, M, the specific steps of which are as follows:
A, to transaction cycle T, average transaction amount P, the last exchange hour interval R, transaction count F and transaction total amount M
Two two compare, and establish judgment matrix;
B, weight coefficient vector is sought;
C, consistency check.
6. a kind of TP-RFM modeling method for corporate client value assessment according to claim 1, it is characterised in that:
In the step D, formula s=∑ w is pressed in the customer value assessmentxxs(x=T, P, R, F, M) weighted sum obtains, wherein wxFor
The weight of x index, xsFor the score after x criterion.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910172131.0A CN110490629A (en) | 2019-03-07 | 2019-03-07 | A kind of TP-RFM modeling method for corporate client value assessment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910172131.0A CN110490629A (en) | 2019-03-07 | 2019-03-07 | A kind of TP-RFM modeling method for corporate client value assessment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110490629A true CN110490629A (en) | 2019-11-22 |
Family
ID=68545681
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910172131.0A Pending CN110490629A (en) | 2019-03-07 | 2019-03-07 | A kind of TP-RFM modeling method for corporate client value assessment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110490629A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112184046A (en) * | 2020-10-12 | 2021-01-05 | 上海移卓网络科技有限公司 | Advertisement service user value evaluation method, device, equipment and storage medium |
-
2019
- 2019-03-07 CN CN201910172131.0A patent/CN110490629A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112184046A (en) * | 2020-10-12 | 2021-01-05 | 上海移卓网络科技有限公司 | Advertisement service user value evaluation method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106952159B (en) | Real estate collateral risk control method, system and storage medium | |
CN107944740A (en) | Merit rating method based on block chain technology | |
CN108665159A (en) | A kind of methods of risk assessment, device, terminal device and storage medium | |
US20020099636A1 (en) | Computerized method, process and service for stock investment timing | |
Mayer-Foulkes | Convergence clubs in cross-country life expectancy dynamics | |
US20210109140A1 (en) | Method for identifying parameters of 10 kv static load model based on similar daily load curves | |
Kapsos | Estimating Growth Requirements for Reducing Working Poverty: Can the World Halve Working Poverty by 2015?. | |
Brzeziński et al. | Wealth inequality in Central and Eastern Europe: Evidence from household survey and rich lists’ data combined | |
JP2004500641A (en) | Method and system for automatically estimating credit score evaluation value | |
Auer et al. | Robust evidence on the similarity of Sharpe ratio and drawdown-based hedge fund performance rankings | |
CN110111024A (en) | Scientific and technological achievement market valuation method based on AHP model of fuzzy synthetic evaluation | |
CN104618924A (en) | Wireless ubiquitous network-based quality of experience index system and measuring method | |
Lowe et al. | The public and private marginal product of capital | |
CN112054943A (en) | Traffic prediction method for mobile network base station | |
Ueno et al. | Computerized adaptive testing based on decision tree | |
CN104679942A (en) | Construction land bearing efficiency measuring method based on data mining | |
CN105868906A (en) | Optimized method for analyzing maturity of regional development | |
Cerovic et al. | Growth and industrial policy during transition | |
CN114154672A (en) | Data mining method for customer churn prediction | |
CN110490629A (en) | A kind of TP-RFM modeling method for corporate client value assessment | |
CN103268391A (en) | Naive-Bayes-based adaptive lightning disaster risk estimation method | |
Lu et al. | Application of grey relational analysis for evaluating road traffic safety measures: advanced driver assistance systems against infrastructure redesign | |
Downes et al. | The macroeconomics of unemployment in the Treasury macroeconomic (TRYM) model | |
CN108711100A (en) | A kind of system of the P2P platform operation risk assessment based on neural network | |
CN113159634A (en) | Financial product management method and device and electronic equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191122 |