CN110111158A - The Marketing Design method, apparatus and storage medium of life cycle or Development phase - Google Patents
The Marketing Design method, apparatus and storage medium of life cycle or Development phase Download PDFInfo
- Publication number
- CN110111158A CN110111158A CN201910406818.6A CN201910406818A CN110111158A CN 110111158 A CN110111158 A CN 110111158A CN 201910406818 A CN201910406818 A CN 201910406818A CN 110111158 A CN110111158 A CN 110111158A
- Authority
- CN
- China
- Prior art keywords
- user
- class cluster
- consumption
- rule
- marketing
- 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 82
- 238000011161 development Methods 0.000 title description 4
- 238000013461 design Methods 0.000 title description 2
- 230000008569 process Effects 0.000 claims description 39
- 238000012549 training Methods 0.000 claims description 37
- 238000006243 chemical reaction Methods 0.000 claims description 31
- 230000035800 maturation Effects 0.000 claims description 17
- 230000014759 maintenance of location Effects 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 abstract description 7
- 238000004458 analytical method Methods 0.000 abstract description 5
- 230000003542 behavioural effect Effects 0.000 abstract description 3
- 238000004422 calculation algorithm Methods 0.000 abstract description 3
- 238000005192 partition Methods 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 description 19
- 238000004364 calculation method Methods 0.000 description 6
- 230000012010 growth Effects 0.000 description 6
- 230000018109 developmental process Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000013145 classification model Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000002459 sustained effect Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000018199 S phase Effects 0.000 description 1
- 230000003698 anagen phase Effects 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000000007 visual effect Effects 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A kind of user property and marketing rule classifying method provided by the invention, applied to electronic device, the following steps are included: target user divides: obtaining the consumption number of times of each user of the first preset period, multiple purchase number is divided into new mature user not less than the minimum user for purchasing number again；User property is sorted out: obtaining each spending amount of each new mature user, according to the first default computation rule, chooses immediate class cluster as class cluster belonging to presently described new mature user；Marketing rule matching: according to the corresponding marketing rule of each class cluster, match with the new mature user.User property and marketing rule classifying method in the present invention go out multiple disaggregated classification clusters with similar behavioural characteristic by Time Series Clustering algorithm partition to mature user, while summarizing the feature of each class cluster；It is finally gradually focused for marked off refinement class cluster by matrix analysis and forms marketing strategy suggestion, and its consumption mode is predicted and intervened.
Description
Technical field
It predicts and intervenes the present invention relates to the consumption mode of big data excavation applications, in particular to a kind of couple consumer
Method, electronic device and computer readable storage medium.
Background technique
As economy and technology are grown rapidly, a kind of new economic situation has been expedited the emergence of --- network economy.Consumer passes through
More and more channel contacts, are browsed, are done shopping, so that the competition between retail trade is also more and more fierce, it is more and more
Traditional retail brand be sold on development line under this pressure with opportunity, by line, under line fusion, business, digitlization think
Dimension is merged, and is gradually completing digitlization transition.The saying of traditional retail industry and pure electric business is not present in the future, retail business
Development comes into the full channel epoch.However, in current retail domain, for the marketing strategy matching of user, there is also following
Problem:
The first, subscriber lifecycle cutting excessively simplify, have ignored between user there is also have the difference of behavior pattern compared with
Big situation can not accomplish targetedly to market to different user；
The second, in the majority to the inclined business type angle of the judgement of the life cycle of user, and the combing based on user data and from
Dynamicization divides group and is not belonging to the speciality of business personnel's artificial treatment, thus not accurate enough to the judgement of subscriber lifecycle.
Summary of the invention
The present invention be to solve the above-mentioned problems and carry out, and it is an object of the present invention to provide a kind of couple of consumer user property into
Method, electronic device and the computer readable storage medium that row is sorted out.The present invention is sold electric business according to user on certain line, under line
Life cycle model marks off mature user；And to mature user by Time Series Clustering algorithm partition go out it is multiple have similar behavior
The disaggregated classification cluster of feature, while summarizing the feature of each class cluster；Finally pass through matrix point for marked off refinement class cluster
Analysis gradually focuses and forms marketing strategy suggestion, and its consumption mode is predicted and intervened.
To achieve the above object, the present invention proposes a kind of user property and marketing rule classifying method, is applied to electronics and fills
It sets, has the feature that, comprising the following steps:
Target user divides: the consumption number of times of each user of the first preset period is obtained, by purchase number again not less than minimum
The user for purchasing number again is divided into new mature user；
User property is sorted out: each spending amount of each new mature user is obtained, according to the first default computation rule,
Immediate class cluster is chosen as class cluster belonging to presently described new mature user；
Marketing rule matching: according to the corresponding marketing rule of each class cluster, match with the new mature user.
User property and marketing rule classifying method proposed by the invention, also has the feature that, described first is pre-
If computation rule includes:
Wherein, akFor the kth time spending amount of new mature user a, cikFor the class cluster center of class cluster i, siFor in class cluster i
Overall consumption number,For the Euclidean distance of new maturation user a and class cluster i, dacFor new maturation user a's and class cluster i
Euclidean distance minimum value.
User property and marketing rule classifying method proposed by the invention, also has the feature that, the attribute is returned
Further include following training step before class step:
Training user divides: according to the consumption number of times of each user of default t raining period, corresponding consumption number is consumed every time,
And the second default computation rule, choose that retain minimum corresponding when conversion ratio tends towards stability to purchase number again be described minimum multiple
Number is purchased, multiple purchase number is divided into mature training user group not less than the minimum user for purchasing number again；
Training data obtains: obtaining the consumption number of times of each user and each spending amount in the mature training user group and makees
For training set data；
Class cluster divides group to handle: Time Series Clustering carried out to the trained manifold, obtains the class cluster belonging to each user, with
And the class cluster center of each class cluster.
User property and marketing rule classifying method proposed by the invention, also has the feature that, in the class cluster
The calculation formula of the heart is as follows:
Wherein, mkFor the user number of kth in class cluster time consumption, bkjFor active user b in the class clusterjKth time disappear
Take the amount of money, class cluster center ckFor user's average consumption amount of money of the kth time consumption of the class cluster.
User property and marketing rule classifying method proposed by the invention, also has the feature that, described second is pre-
If computation rule includes:
Wherein, n is consumption number of times, n >=2, mnAnd mn-1Respectively consume n times and n-1 user number, dnTo consume n
Retention conversion ratio when secondary.
User property and marketing rule classifying method proposed by the invention, also has the feature that, the trained number
It is further comprising the steps of according to obtaining step:
Training set boundary demarcation: according to each consumption number of times of user of the mature training user group, consumption pair every time
The consumption number and third answered preset computation rule, set each consumption number of times no more than highest and purchase frequency n again2Data make
For the training set data.
User property and marketing rule classifying method proposed by the invention, also has the feature that, the third is pre-
If computation rule:
Wherein, f is multiple purchase person-time accounting, mnFor the user number for consuming n times, mn1Frequency n is purchased again for consumption minimum1Secondary
User number, corresponding multiple purchase number minimum value purchases number as highest again when using f≤5%.
User property and marketing rule classifying method proposed by the invention, also has the feature that, further includes following
Step:
Class cluster feature is concluded: obtaining each class cluster average consumption number, each class cluster average single spending amount, each
The class cluster single user is averaged the overall consumption amount of money, each class cluster average consumption time interval and each class cluster center, obtains
Feature to each class cluster describes；
Marketing rule obtains: obtaining the be averaged overall consumption amount of money, mature user's mean residence of the mature user and converts
Rate, mature user's average consumption interval time be averaged the overall consumption amount of money, each class cluster use with each class cluster single user
Family number, each class cluster mean residence conversion ratio, each class cluster average consumption time interval comparison, and it is based on each class
The feature of cluster describes, and obtains the marketing rule corresponding with each class cluster.
In addition, to achieve the above object, the present invention also provides a kind of electronic devices, have the feature that, electronics dress
Setting includes: memory, processor, user property and marketing rule sorting process is stored on the memory, the user belongs to
Property and marketing rule sorting process perform the steps of when being executed by the processor
Target user divides: the consumption number of times of each user of the first preset period is obtained, by purchase number again not less than minimum
The user for purchasing number again is divided into new mature user；
User property is sorted out: each spending amount of each new mature user is obtained, according to the first default computation rule,
Immediate class cluster is chosen as class cluster belonging to presently described new mature user；
Marketing rule matching: according to the corresponding marketing rule of each class cluster, match with the new mature user.
In addition, to achieve the above object, the present invention also provides a kind of computer readable storage mediums, there is such spy
It levies, is stored with the user property and marketing rule sorting process, the user property on the computer readable storage medium
The step of above-mentioned user property and marketing rule classifying method are realized when being executed by processor with marketing rule sorting process.
Invention effect and effect
The method, apparatus and storage medium that a kind of user property and marketing rule according to the present invention are sorted out, provide
A kind of user belongs to class cluster disaggregated model and method, according to the consumer behavior developmental process of user, is decomposed based on its consumer behavior
Multiple behavior key points out, and the data of different consumer phases are extracted according to these behavior key points, disappear to mark off user
Take each stage of life cycle, and carry out class cluster according to the consumer behavior of mature user and performance and divide group, to obtain maturation
Sub- life cycle behavioural characteristic of the user group in each class cluster.User class cluster disaggregated model of the invention and method can incite somebody to action
Mature user with consumption feature general character is respectively divided in each certain kinds cluster.In addition, user property and battalion in the present invention
Then classifying method can also be matched one by one new mature user according to above-mentioned class cluster pin gauge, the monoid of Institute of Automation ownership,
Obtain the life cycle pullulation module of the new mature user's growth habit of more fitting；Then, according to each class of above-mentioned new ripe user
Consumption feature of the user in different growth stages in cluster pointedly sets marketing objectives, strategy, and is guided, by with
Track user's critical behavior data clues find specific aim and import, promotion user retention, increase mature user volume, reduce loss use
Access is guided in the iterative marketing that family ratio, Drain Causes are tracked.User's future not only can be predicted in method provided by the invention
Life cycle and consumption mode can also formulate the rule with formation marketing strategy suggestion according to subdivision group's monitoring index, so as to
Sustained period according to latest data situation, the subsequent marketing strategy suggestion of output.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
The structure shown according to these attached drawings obtains other attached drawings.
Fig. 1 is the schematic diagram of one embodiment of electronic device of the present invention；
Fig. 2 shows the Program modual graph of one embodiment of user property and marketing rule sorting process of the invention；
Fig. 3 is the subscriber lifecycle stage signal of one embodiment of user property and marketing rule sorting process of the invention
Figure；
Fig. 4 is the customer consumption data statistics chart of one embodiment of user property and marketing rule sorting process of the invention
One；
Fig. 5 is each stage observation signal of user of one embodiment of user property and marketing rule sorting process of the invention
Figure；
Fig. 6 is one embodiment of user property and marketing rule sorting process of the invention based on class cluster average consumption number
With the class cluster matrix distribution map of class cluster average single spending amount；
Fig. 7 is that the class cluster of one embodiment of user property and marketing rule sorting process of the invention divides schematic diagram；
Fig. 8 is the class cluster signature analysis table of one embodiment of user property and marketing rule sorting process of the invention；
Fig. 9 is that the quasi-group characteristic of one embodiment of user property and marketing rule sorting process of the invention simplifies conclusion table；
Figure 10 is being turned based on class cluster mean residence for one embodiment of user property and marketing rule sorting process of the invention
Rate-class cluster single user is averaged the class cluster matrix distribution map of the overall consumption amount of money；
Figure 11 is the subdivision monoid marketing strategy table of one embodiment of user property and marketing rule sorting process of the invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
The principle and features of the present invention will be described below with reference to the accompanying drawings, and the given examples are served only to explain the present invention, and
It is non-to be used to limit the scope of the invention.
The present invention provides a kind of electronic device 1.It is the signal of 1 preferred embodiment of electronic device of the present invention shown in referring to Fig.1
Figure.
In the present embodiment, which includes memory 11, processor 12, network interface 13 and communication bus.Its
In, communication bus is for realizing the connection communication between these components.
Network interface 13 may include standard wireline interface and wireless interface (such as WI-FI interface).
Memory 11 includes the readable storage medium storing program for executing of at least one type.The readable storage medium storing program for executing of at least one type
It can be the non-volatile memory medium of such as flash memory, hard disk, multimedia card, card-type memory.In some embodiments, described can
Reading storage medium can be the internal storage unit of the electronic device 1, such as the hard disk of the electronic device 1.In other realities
It applies in example, the readable storage medium storing program for executing is also possible to the External memory equipment of the electronic device 1, such as the electronic device 1
The plug-in type hard disk of upper outfit, intelligent memory card (Smart Media Card, SMC), secure digital (Secure Digital,
SD) block, flash card (Flash Card) etc..
In the present embodiment, the readable storage medium storing program for executing of the memory 11 is installed on the electronic device commonly used in storage
1 user property and marketing rule sorting process 10 etc..The memory 11 can be also used for temporarily storing exported or
The data that person will export.
Processor 12 can be in some embodiments a central processing unit (Central Processing Unit,
CPU), microprocessor or other data processing chips, program code or processing data for being stored in run memory 11, example
Such as execute user property and marketing rule sorting process 10.
Fig. 1 illustrates only the electronic device 1 with component 11-13 and user property and marketing rule sorting process 10,
It should be understood that be not required for implementing all components shown, the implementation that can be substituted is more or less component.
Optionally, the electronic device 1 can also include user interface, user interface may include display (Display),
Input unit such as keyboard (Keyboard), optional user interface can also include standard wireline interface and wireless interface.
Optionally, which can also include display, can be light-emitting diode display, liquid crystal in some embodiments
Display, touch-control liquid crystal display and Organic Light Emitting Diode (Organic Light-Emitting Diode, OLED)
Touch device etc..Display is for showing the information handled in an electronic and for showing visual user interface.
Fig. 2 shows the Program modual graph of one embodiment of user property and marketing rule sorting process of the invention.
In Installation practice shown in Fig. 1, belong to as in a kind of memory 11 of computer storage medium including user
Property and marketing rule sorting process 10, processor 12 execute the user property and marketing rule sorting process that store in memory 11
Comprise the following modules when 10: target user's division module 110, user property sort out 120, marketing rule matching 130.Institute of the present invention
The module of title is the series of computation machine program instruction section for referring to complete specific function.It is user in Fig. 1 referring to shown in Fig. 2
The Program modual graph of attribute and marketing rule sorting process.
In the present embodiment, user property and marketing rule sorting process 10 may include:
Target user's division module: obtaining the consumption number of times of each user of the first preset period, and purchase number again is not less than
The minimum user for purchasing number again is divided into new mature user.
In above-mentioned target user's partiting step, selected is the user group for needing to carry out category analysis, above-mentioned
There is no limit can be the user of current period, be also possible to the user prior to current period, can also be first preset period
The user in arbitrarily selected period.It is common, the division and consumption of consumption mode are carried out in order to the user to the newest period
The prediction of behavior chooses the user group of current period as target user in the present embodiment.The potential user group is current institute
There is the summation of user.
Fig. 3 is the subscriber lifecycle stage signal of one embodiment of user property and marketing rule sorting process of the invention
Figure.
Specifically, as shown in figure 3, the consumption life cycle of user can at least be divided into stage A (initial stage), stage B (growth
Phase), stage C (maturity period), stage D (decline phase) and stage E (being lost area).Initial stage, growth stage user and maturity period user
Difference with purchase number.It is above-mentioned minimum to purchase frequency n again1It is the data precalculated, as mature user
With the separation of other non-mature users, used for dividing mature user and other non-maturations in the user group of current period
Family.Purchase number refers to the number of user's Double Spending purchase again, will divide into not less than the minimum user for purchasing number again
Mature user.In the present embodiment, this minimum purchases frequency n again1It is 6, i.e. user of the purchase number not less than 6 times will be subdivided into
New maturation user.
User property classifying module: each spending amount of each new mature user is obtained, according to the first default calculating
Rule chooses immediate class cluster as class cluster belonging to presently described new mature user.
Above-mentioned class cluster existing all kinds of clusters classification moulds from above-mentioned user property and marketing rule classifying method program
It is chosen in type.According to the different consumption modes of each mature user, user property and marketing rule classifying method program kind have not
Same class cluster, it is however generally that, each class cluster will correspond to one with the user of certain consumption feature general character cluster to together
Typical consumption mode.Each class cluster has respective class cluster center.It is default according to first in above-mentioned new mature user
Each new mature user and all kinds of cluster centers are done comparing calculation, chosen and the immediate class of maturity period user by computation rule
Cluster.
Further, the first above-mentioned default computation rule is as follows:
Wherein, akFor the kth time spending amount of new mature user a, cikFor the class cluster center of class cluster i, siFor in class cluster i
Overall consumption number,For the Euclidean distance of new maturation user a and class cluster i, dacFor new maturation user a's and class cluster i
Euclidean distance minimum value.
In some embodiments, there are 5 different class clusters (i is followed successively by 1,2,3,4,5), therefore in user's disaggregated model
Active user a and class cluster center c1k、c2k、c3k、c4k、c5kEuclidean distance minimum value calculation formula are as follows:
The minimum value for choosing each Euclidean distance, as the matched class cluster of current user a institute.In some embodiments, a and class
Cluster center c1k、c2k、c3k、c4k、c5kEuclidean distance minimum value corresponding to class cluster center be c3k, then active user a and
Three classes cluster matches, and active user a sorts out to third class cluster.
Marketing rule matching module: according to the corresponding marketing rule of each class cluster, with new mature user's phase
Match.
In the present invention, each class cluster has a corresponding different marketing rule, and all kinds of corresponding marketing rules of cluster can be with
Be it is existing, be also possible to by obtaining in the training step of user property classification model.In some embodiments, pass through
Marketing rule corresponding with all kinds of clusters is obtained in the training of user property classification model, and is matched with new mature user.
Further, the present invention also provides a kind of training steps of above-mentioned class cluster disaggregated model:
Training user divides: according to the consumption number of times of each user of default t raining period, corresponding consumption number is consumed every time,
And the second default computation rule, choose that retain minimum corresponding when conversion ratio tends towards stability to purchase number again be described minimum multiple
Purchase frequency n1, not less than described minimum frequency n will be purchased again by purchase number again1User be divided into mature training user group.
Before carrying out user property division, need to the class cluster disaggregated model to user be trained.In above-mentioned training user
In partiting step, user data need to be chosen in advance and carries out analysis and model training, use of the user data from preset period
Family group, above-mentioned preset period are usually the user prior to current period, and default each user of t raining period is t raining period
The summation of all users.
As described above, as shown in figure 3, the consumption life cycle of user can be divided into initial stage, growth stage and maturity period.Above-mentioned
In acquired training user group, by observing the historical trading behavior of all users, statistics summarizes each user and disappears
All dates taken, and the All Activity occurred within the consumption date.User consumes all of the generation on date at one
One-time-consumption is can be regarded as in transaction, i.e., 1 purchase number is included in the total of this consumption in all amount of money that the consumption in a few days occurs
The amount of money.That is: the total amount of all consumption of the first consumption of user in a few days, the spending amount consumed as the 1st time；2nd disappears
The total amount for taking all consumption in a few days, the spending amount consumed as the 2nd time；I-th of consumption all consumption in a few days it is total
The amount of money, the spending amount as i-th consumption.
Fig. 4 is the customer consumption data statistics chart of one embodiment of user property and marketing rule sorting process of the invention
One.User is counted according to above-mentioned consumption number of times, obtains the statistics knot of user's number corresponding to each consumption number of times
Fruit, the partial content of the form of statistical result referring to fig. 4.
Fig. 5 is each stage observation signal of user of one embodiment of user property and marketing rule sorting process of the invention
Figure.
According to the statistical result of above-mentioned consumption number of times and the corresponding user number consumed, using consumption number of times as horizontal axis,
User number is the longitudinal axis, draws and obtains the number of users distribution map corresponding with consumption number of times of lower half portion in Fig. 5, purchase again
Number of users is reduced with the increase of consumption number of times, and purchase number is more again, and corresponding number of users is fewer.
Further, the retention conversion ratio purchased again every time is calculated according to the second default computation rule, chooses retention conversion ratio and becomes
Corresponding minimum purchases number as minimum again and purchases frequency n again when stablizing1, divide mature training user group.
The default computation rule of above-mentioned second includes:
Wherein, n is consumption number of times, n >=2, mnAnd mn-1Respectively consume n times and n-1 user number, dnTo consume n
Retention conversion ratio when secondary.As n=1, d is set1=100%.
According to the statistical result of number of users corresponding to each purchase number, calculate separately corresponding to each purchase number
User purchases conversion example again.In some embodiments, the number for 1,2,3,4,5,6,7 order occurring is respectively 46443,12479,
6034,3585,2427,1759,1311.According to the second above-mentioned default computation rule, the retention conversion ratio being calculated: d1
=100%, d2=26.9%, d3=48.4%, d3=59.5%, d4=67.7%, d5=72.5%, d6=74.5%.
According to above-mentioned calculated result, using consumption number of times as horizontal axis, retention conversion ratio is the longitudinal axis, draws and obtains in Fig. 5
The retention conversion profile figure corresponding with consumption number of times of half part.
In some embodiments, it as shown in figure 5, as n >=6, retains conversion ratio and tends towards stability, even if consumption number of times increase
Add, retains conversion ratio and remain unchanged substantially.Choosing the minimum Number of Orders corresponding with respect to stability region is to step into the stage of ripeness
The user demarcation point of (stage C in figure) is that is, minimum to purchase frequency n again1.In the present embodiment, n1=6.Ordered in t raining period
User of the number not less than 6 times will be subdivided into mature user pond.
Training data obtains: obtaining the consumption number of times of each user and each spending amount in the mature training user group and makees
For training set data.
In above-mentioned acquired mature user pond, all historical trading datas for choosing above-mentioned mature user form user
Sequence, as training set data.It is common, in user's sequence, for the ease of analyzing the data in sequence,
Boundary and the length of user's sequence need to be defined.
Specifically, also there is training set boundary in user property and marketing rule classifying method proposed by the invention
Partiting step: according to consuming corresponding user number every time in the mature training user group and third is preset computation rule and obtained
Highest purchases frequency n again2, the consumption number of times of each user in the training set data, which are set, no more than highest purchases frequency n again2。
It is as follows that above-mentioned third presets computation rule:
Wherein, f is multiple purchase person-time accounting, mnFor the user number for consuming n times, mn1Frequency n is purchased again for consumption minimum1Secondary
User number, corresponding multiple purchase number minimum value purchases frequency n as highest again when using f≤5%2。
Specifically, minimum purchasing according to the corresponding user number of each consumption and again frequency n1Corresponding user
Number calculates above-mentioned highest and purchases frequency n again2,.As shown in figure 5, in some embodiments, n1=6, corresponding user number at this time
It is 1759.As f≤5%, corresponding mnIt is 87.95, immediate n is 18 or 19.For ease of calculation with subsequent class cluster
The comprehensibility of Comparative result observation takes roughization processing to round up herein, and the highest of selection purchases frequency n again2For
10 or 5 multiple, therefore n at this time2=20, each Subscriber Unit consumption number of times, which are set, no more than highest purchases frequency n again2Data make
For the training set data.
Class cluster divides group to handle: Time Series Clustering carried out to the trained manifold, obtains the class cluster belonging to each user, with
And the class cluster center of each class cluster.
Specifically, according to the user data sequence in training set obtained by above-mentioned calculating, using DTW Time Series Clustering algorithm,
Carrying out class cluster to the related data of each user divides group to handle, and obtains the class cluster center of all kinds of clusters.
The calculation formula at class cluster described above center is as follows:
Wherein, mkFor the user number of kth in class cluster time consumption, bkjFor active user b in the class clusterjKth time disappear
Take the amount of money, class cluster center ckFor user's average consumption amount of money of the kth time consumption of the class cluster.
Fig. 7 is that the class cluster of one embodiment of user property and marketing rule sorting process of the invention divides schematic diagram.Such as Fig. 7
Shown, all maturation users are classified as different cluster classes.Each class cluster purchases frequency n by being not more than highest again2A column composition.Often
When a column represents the consumption of n-th corresponding to the column in sequence, the average consumption amount of money of all users in such cluster, often
A class cluster intuitively shows the consumption mode of active user group by column from the size and height trend of spending amount.It is different
The mature user of class cluster, there is apparent difference in consumer behavior mode and spending amount.
As shown in fig. 7, in some embodiments, all maturation users are classified as five different cluster classes, the respectively first kind
Cluster, the second class cluster, third class cluster, the 4th class cluster and the 5th class cluster.Each class cluster has 20 columns, to show such cluster
Consumer behavior and the propensity to consume.
In some embodiments, the specific formula for calculation of third class cluster is as follows:
When calculating the class cluster center of third class cluster, as shown in fig. 7, the user number m that kth time is consumed in third class clusterk
It is 412, bkjFor active user b in third class clusterjKth time spending amount, class cluster center ckFor the kth time consumption of third class cluster
User's average consumption amount of money.
User property and marketing rule classifying method proposed by the invention, also has the feature that, further includes following
Step:
Class cluster feature is concluded: obtaining each class cluster average consumption number, each class cluster average single spending amount, each
The class cluster single user is averaged the overall consumption amount of money, each class cluster average consumption time interval and each class cluster center, obtains
Feature to each class cluster describes.
Fig. 6 is one embodiment of user property y and marketing rule sorting process of the invention based on the average consumption time of class cluster
Several and class cluster average single spending amount class cluster matrix distribution map.
Specifically, in some embodiments, average consumption number and average single spending amount based on all kinds of clusters are (whole
Body Weighted Guidelines), obtain the class cluster matrix distribution map of Fig. 6.As shown in fig. 6, the number of user number is by representing class in class cluster
The size of the figure of cluster indicates.Accounting the second class cluster fewer in number, third class cluster, the 4th higher average single of class cluster disappear
Take the amount of money.In these three class clusters, third class cluster have highest average single spending amount and least consumption number of times, the 4th
Class cluster then has highest average consumption number as most frequent class cluster is consumed.
Fig. 8 is the class cluster signature analysis table of one embodiment of user property and marketing rule sorting process of the invention.
In addition, by statistics calculate each class cluster single user be averaged the overall consumption amount of money, each class cluster average consumption when
Between be spaced, in conjunction with the class cluster each in Fig. 6 and Fig. 5 center, obtain the feature description of each class cluster.
As shown in figure 8, in some embodiments, first kind cluster represents the user group of steady, slow heat type consumption；Second class cluster
Represent the high-value user group of rationality appreciation type；The high-value user of consumption-orientation, consumption mode table in third class cluster presenting set
Now it is to Price Sensitive, tends to store goods, every time consumption interval length；4th class cluster represent high frequency, high value can value-added user
Group；5th class cluster represents low frequency and is easy to run off user group.
Marketing rule obtains: obtaining the be averaged overall consumption amount of money, mature user's mean residence of the mature user and converts
Rate, mature user's average consumption interval time be averaged the overall consumption amount of money, each class cluster use with each class cluster single user
Family number, each class cluster mean residence conversion ratio, each class cluster average consumption time interval comparison, and it is based on each class
The feature of cluster describes, and obtains the marketing rule corresponding with each class cluster.
Fig. 9 is that the quasi-group characteristic of one embodiment of user property and marketing rule sorting process of the invention simplifies conclusion table.
Specifically, in the present invention, based on the mature number of users and each user's overall consumption amount of money in training set, obtain at
The average overall consumption amount of money of ripe user.As shown in figure 9, maturation user the average of the overall consumption amount of money and all kinds of clusters that be averaged always is disappeared
The expense amount of money compares, and will be above the be averaged class cluster of the overall consumption amount of money of mature user and is divided into top grade, will be less than maturation user and be averaged
The class cluster of the overall consumption amount of money is divided into low grade, will be divided into middle-grade with the class cluster that it is fair that maturation user is averaged the overall consumption amount of money.
In addition, the mean residence conversion ratio of the mature user, mature user's average consumption time interval can be obtained,
It is successively flat to all kinds of cluster mean residence conversion ratios, all kinds of cluster average user numbers, all kinds of clusters according to above-mentioned high, normal, basic stepping rule
Equal consumption time interval compares and stepping.
In addition, the average single spending amount of the mature user can be obtained, it is right according to above-mentioned high, normal, basic stepping rule
The average single spending amount of all kinds of clusters compares and stepping.
In some embodiments, as shown in figure 9, third class cluster is flat in class cluster average single spending amount, class cluster single user
Belong to high-grade classification on the equal overall consumption amount of money, class cluster mean residence conversion ratio, is higher than average level, in class cluster average consumption number
Belong to low-grade classification in class cluster average consumption time interval, is lower than average level.5th class cluster is consumed in class cluster average single
The amount of money belongs to middle-grade classification, is average level, is averaged the overall consumption amount of money, class cluster average consumption number, class cluster in class cluster single user
Belong to low-grade classification in average consumption time interval, class cluster mean residence conversion ratio, is lower than average level.
Figure 10 is being turned based on class cluster mean residence for one embodiment of user property and marketing rule sorting process of the invention
Rate-class cluster single user is averaged the class cluster matrix distribution map of the overall consumption amount of money.
As shown in Figure 10, the number of user number is indicated by representing the size of the figure of class cluster in class cluster.Have in figure
There is the cross figure for representing mature user's mean residence conversion ratio and mature user's overall consumption amount of money.With corresponding to Fig. 9,
The class cluster mean residence high conversion rate of three classes cluster is in mature user's mean residence conversion ratio.
Figure 11 is the subdivision monoid marketing strategy table of one embodiment of user property and marketing rule sorting process of the invention.
As shown in figure 11, according to Fig. 9, Figure 10, marketing Suggestions rule corresponding with class cluster as follows is formed:
First kind cluster: when class cluster average consumption total amount is low grade, and class cluster mean residence conversion ratio is high-grade, class cluster
Middle user is stable type user, and marketing strategy is emphasis holding, and promotes value in right amount；In addition, when the average consumption of class cluster is always golden
When volume is middle-grade and class cluster mean residence conversion ratio is top grade, also it is contemplated that using this marketing strategy.
Second class cluster: when class cluster average consumption total amount is top grade, and class cluster mean residence conversion ratio is low-grade, class cluster
Middle user is high-value user, then marketing strategy is that emphasis promotes retention, and provides good service.
Third class cluster: when class cluster average consumption total amount is top grade, and class cluster mean residence conversion ratio is top grade, and class cluster
When average consumption time interval is high-grade, user is high value and stable type user in class cluster, then marketing strategy is locking source
Channel, preferential to drain, emphasis is propagated；
4th class cluster: when class cluster average consumption total amount is top grade, and class cluster mean residence conversion ratio is low-grade, class cluster
Middle user is high-value user, then marketing strategy is that emphasis promotion is retained, waken up, loss is retrieved, emphasis tracks Drain Causes；
5th class cluster: when class cluster average consumption total amount is low-grade and class cluster mean residence conversion ratio is low-grade, and class cluster
When average consumption time interval is low grade, then marketing strategy is to promote value, general wake-up, be lost and retrieve.
In addition, the embodiment of the present invention also proposes a kind of computer readable storage medium, the computer readable storage medium
On be stored with user property and marketing rule sorting process, the user property classification realizes following behaviour when being executed by processor
Make:
Target user divides: the consumption number of times of each user of the first preset period is obtained, by purchase number again not less than minimum
Frequency n is purchased again1User be divided into new mature user；
User property is sorted out: each spending amount of each new mature user is obtained, according to the first default computation rule,
Choose the affiliated class cluster with each new mature immediate class cluster of user as presently described new mature user；
Marketing mode matching:, will be with each corresponding marketing rule of class cluster and the new maturation according to affiliated class cluster
User matches.
The specific embodiment of the computer readable storage medium of the present invention and above-mentioned user property and marketing rule are sorted out
The specific embodiment of method is roughly the same, and details are not described herein.
The action and effect of embodiment
The method, apparatus and storage medium that the user property according to involved in the present embodiment and marketing rule are sorted out, provide
A kind of user class cluster disaggregated model and method are decomposed according to the consumer behavior developmental process of user based on its consumer behavior
Multiple behavior key points out, and the data of different consumer phases are extracted according to these behavior key points, disappear to mark off user
Take each stage of life cycle, and carry out class cluster according to the consumer behavior of mature user and performance and divide group, to obtain maturation
Sub- life cycle behavioural characteristic of the user group in each class cluster.The class cluster disaggregated model and method of the present embodiment can will have
There is the mature user of consumption feature general character to be respectively divided in each certain kinds cluster.In addition, user property and battalion in the present embodiment
Then classifying method can also be matched one by one new mature user according to above-mentioned class cluster pin gauge, the monoid of Institute of Automation ownership,
Obtain the life cycle pullulation module of the new mature user's growth habit of more fitting；Then, according to each class of above-mentioned new ripe user
Consumption feature of the user in different growth stages in cluster pointedly sets marketing objectives, strategy, and is guided, by with
Track user's critical behavior data clues find specific aim and import, promotion user retention, increase mature user volume, reduce loss use
Access is guided in the iterative marketing that family ratio, Drain Causes are tracked.User's future not only can be predicted in method provided in this embodiment
Life cycle and consumption mode, can also according to subdivision group's monitoring index, formulate with formed marketing strategy suggestion rule, with
Just sustained period according to latest data situation, the subsequent marketing strategy suggestion of output.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, device, article or the method that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, device, article or method institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, device of element, article or method.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.Pass through above embodiment party
The description of formula, it is required general that those skilled in the art can be understood that above-described embodiment method can add by software
The mode of hardware platform is realized, naturally it is also possible to which by hardware, but in many cases, the former is more preferably embodiment.It is based on
Such understanding, substantially the part that contributes to existing technology can be with software product in other words for technical solution of the present invention
Form embody, which is stored in a storage medium (such as ROM/RAM, magnetic disk, light as described above
Disk) in, including some instructions use is so that a terminal device (can be mobile phone, computer, server or the network equipment
Deng) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of user property and marketing rule classifying method are applied to electronic device, which comprises the following steps:
Target user divides: obtaining the consumption number of times of each user of the first preset period, purchase number again is not less than minimum multiple purchase
The user of number is divided into new mature user；
User property is sorted out: each spending amount for obtaining each new mature user is chosen according to the first default computation rule
Immediate class cluster is as class cluster belonging to presently described new mature user；
Marketing rule matching: according to the corresponding marketing rule of each class cluster, match with the new mature user.
2. user property according to claim 1 and marketing rule classifying method, which is characterized in that described first is pre-designed
Calculating rule includes:
Wherein, akFor the kth time spending amount of new mature user a,
cikFor the class cluster center of class cluster i,
siFor the overall consumption number in class cluster i,
For the Euclidean distance of new maturation user a and class cluster i,
dacFor the Euclidean distance minimum value of new maturation user a and class cluster i.
3. user property as claimed in claim 1 or 2 and marketing rule classifying method, which is characterized in that the user property
Further include following training step before classifying step:
Training user divides: according to the consumption number of times of each user of default t raining period, corresponding consumption number is consumed every time, and
Second default computation rule, choosing minimum corresponding when retention conversion ratio tends towards stability and purchasing number again is the minimum multiple purchase time
Multiple purchase number is divided into mature training user group not less than the minimum user for purchasing number again by number；
Training data obtains: obtaining the consumption number of times of each user and each spending amount in the mature training user group and is used as instruction
Practice collection data；
Class cluster divides group to handle: carrying out Time Series Clustering to the trained manifold, obtains the class cluster belonging to each user, and each
The class cluster center of the class cluster.
4. user property as claimed in claim 3 and marketing rule classifying method, which is characterized in that the meter at class cluster center
It is as follows to calculate formula:
Wherein, mkFor the user number of kth in class cluster time consumption,
bkjFor active user b in the class clusterjKth time spending amount,
Class cluster center ckFor user's average consumption amount of money of the kth time consumption of the class cluster.
5. user property as claimed in claim 3 and marketing rule classifying method, which is characterized in that the described second default calculating
Rule includes:
Wherein, n is consumption number of times, n >=2,
mnAnd mn-1N times and n-1 user number are respectively consumed,
dnTo consume retention conversion ratio when n times.
6. user property as claimed in claim 3 and marketing rule classifying method, which is characterized in that the training data obtains
Step is further comprising the steps of:
Training set boundary demarcation: according to consuming corresponding user number every time in the mature training user group and third is pre-designed
Rule is calculated to obtain highest and purchase number again, set each user in the training set data consumption number of times purchase again no more than highest it is secondary
Number.
7. user property as claimed in claim 6 and marketing rule classifying method, which is characterized in that the third is default to be calculated
Rule:
Wherein, f is multiple purchase person-time accounting, mnFor the user number for consuming n times, mn1Frequency n is purchased again for consumption minimum1Secondary user
Number,
Corresponding multiple purchase number minimum value purchases number as highest again when using f≤5%.
8. user property as claimed in claim 3 and marketing rule classifying method, which is characterized in that the class cluster divides group's step
Afterwards, further comprising the steps of:
Class cluster feature is concluded: obtaining each class cluster average consumption number, each class cluster average single spending amount, each described
Class cluster single user is averaged the overall consumption amount of money, each class cluster average consumption time interval and each class cluster center, obtains each
The feature of the class cluster describes；
Marketing rule obtains: obtaining the mature user and be averaged the overall consumption amount of money, maturation user's mean residence conversion ratio, institute
Mature user's average consumption interval time is stated, is averaged the overall consumption amount of money, each class cluster user people with each class cluster single user
Several, each class cluster mean residence conversion ratio, each class cluster average consumption time interval comparison, and based on each class cluster
Feature description obtains the marketing rule corresponding with each class cluster.
9. a kind of electronic device, which is characterized in that the electronic device includes: memory, processor, is stored on the memory
It is real when user property and marketing rule sorting process, the user property and marketing rule sorting process are executed by the processor
Existing following steps:
Target user divides: obtaining the consumption number of times of each user of the first preset period, purchase number again is not less than minimum multiple purchase
The user of number is divided into new mature user；
User property is sorted out: each spending amount for obtaining each new mature user is chosen according to the first default computation rule
Immediate class cluster is as class cluster belonging to presently described new mature user；
Marketing rule matching: according to the corresponding marketing rule of each class cluster, match with the new mature user.
10. a kind of computer readable storage medium, which is characterized in that be stored with the use on the computer readable storage medium
It realizes when family attribute and marketing rule sorting process, the user property and marketing rule sorting process are executed by processor as weighed
Benefit require any one of 1 to 2 described in user property and the step of marketing rule classifying method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910406818.6A CN110111158A (en) | 2019-05-16 | 2019-05-16 | The Marketing Design method, apparatus and storage medium of life cycle or Development phase |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910406818.6A CN110111158A (en) | 2019-05-16 | 2019-05-16 | The Marketing Design method, apparatus and storage medium of life cycle or Development phase |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110111158A true CN110111158A (en) | 2019-08-09 |
Family
ID=67490350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910406818.6A Pending CN110111158A (en) | 2019-05-16 | 2019-05-16 | The Marketing Design method, apparatus and storage medium of life cycle or Development phase |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110111158A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751515A (en) * | 2019-10-16 | 2020-02-04 | 北京百分点信息科技有限公司 | Decision-making method and device based on user consumption behaviors, electronic equipment and storage medium |
| CN110827070A (en) * | 2019-10-30 | 2020-02-21 | 神州数码融信软件有限公司 | User growth calculation method and device based on dynamic expansion factor |
| CN113688326A (en) * | 2021-10-26 | 2021-11-23 | 腾讯科技（深圳）有限公司 | Recommendation method, device, equipment and computer readable storage medium |
| CN113781128A (en) * | 2021-10-15 | 2021-12-10 | 北京明略软件系统有限公司 | High-potential consumer identification method, system, electronic device, and medium |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20050024695A (en) * | 2003-09-01 | 2005-03-11 | 주식회사 엘지홈쇼핑 | A metod for determining marketing type according to the parameter of user attribute on electronic commerce |
| US20050102213A1 (en) * | 2003-11-07 | 2005-05-12 | Serkan Savasoglu | Systems and methods for accreting remarketable convertible securities |
| KR20110065970A (en) * | 2009-12-10 | 2011-06-16 | 동국대학교 산학협력단 | Apparatus and method for product recommendation |
| US8364522B1 (en) * | 2008-01-30 | 2013-01-29 | Intuit Inc. | Method and system for providing a small business coupon distribution system |
| US20150142521A1 (en) * | 2013-11-20 | 2015-05-21 | Sears Brands, Llc | Customer clustering using integer programming |
| 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 |
| CN106952145A (en) * | 2017-03-24 | 2017-07-14 | 广州茶途网络科技有限公司 | It is a kind of to try out user's choosing method and device based on the commodity that big data is analyzed |
| CN107103024A (en) * | 2016-12-23 | 2017-08-29 | 阿里巴巴集团控股有限公司 | Consumption data processing method, device and electronic equipment |
| CN108289121A (en) * | 2018-01-02 | 2018-07-17 | 阿里巴巴集团控股有限公司 | The method for pushing and device of marketing message |
| CN108492194A (en) * | 2018-03-06 | 2018-09-04 | 平安科技（深圳）有限公司 | Products Show method, apparatus and storage medium |
| CN108765052A (en) * | 2018-04-20 | 2018-11-06 | 网易无尾熊（杭州）科技有限公司 | Electric business recommendation/method for pushing and device, storage medium and computing device |
| CN109543987A (en) * | 2018-11-16 | 2019-03-29 | 广州知了科技有限公司 | The subjective and objective Double Data analytic approach of internet and mobile Internet product |
| CN109583651A (en) * | 2018-12-03 | 2019-04-05 | 焦点科技股份有限公司 | A kind of method and apparatus for insuring electric business platform user attrition prediction |
| CN109615426A (en) * | 2018-12-05 | 2019-04-12 | 重庆锐云科技有限公司 | A kind of marketing method based on Customer clustering, system |
-
2019
- 2019-05-16 CN CN201910406818.6A patent/CN110111158A/en active Pending
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20050024695A (en) * | 2003-09-01 | 2005-03-11 | 주식회사 엘지홈쇼핑 | A metod for determining marketing type according to the parameter of user attribute on electronic commerce |
| US20050102213A1 (en) * | 2003-11-07 | 2005-05-12 | Serkan Savasoglu | Systems and methods for accreting remarketable convertible securities |
| US8364522B1 (en) * | 2008-01-30 | 2013-01-29 | Intuit Inc. | Method and system for providing a small business coupon distribution system |
| KR20110065970A (en) * | 2009-12-10 | 2011-06-16 | 동국대학교 산학협력단 | Apparatus and method for product recommendation |
| US20150142521A1 (en) * | 2013-11-20 | 2015-05-21 | Sears Brands, Llc | Customer clustering using integer programming |
| 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 |
| CN107103024A (en) * | 2016-12-23 | 2017-08-29 | 阿里巴巴集团控股有限公司 | Consumption data processing method, device and electronic equipment |
| CN106952145A (en) * | 2017-03-24 | 2017-07-14 | 广州茶途网络科技有限公司 | It is a kind of to try out user's choosing method and device based on the commodity that big data is analyzed |
| CN108289121A (en) * | 2018-01-02 | 2018-07-17 | 阿里巴巴集团控股有限公司 | The method for pushing and device of marketing message |
| CN108492194A (en) * | 2018-03-06 | 2018-09-04 | 平安科技（深圳）有限公司 | Products Show method, apparatus and storage medium |
| CN108765052A (en) * | 2018-04-20 | 2018-11-06 | 网易无尾熊（杭州）科技有限公司 | Electric business recommendation/method for pushing and device, storage medium and computing device |
| CN109543987A (en) * | 2018-11-16 | 2019-03-29 | 广州知了科技有限公司 | The subjective and objective Double Data analytic approach of internet and mobile Internet product |
| CN109583651A (en) * | 2018-12-03 | 2019-04-05 | 焦点科技股份有限公司 | A kind of method and apparatus for insuring electric business platform user attrition prediction |
| CN109615426A (en) * | 2018-12-05 | 2019-04-12 | 重庆锐云科技有限公司 | A kind of marketing method based on Customer clustering, system |
Non-Patent Citations (4)
| Title |
|---|
| 吴自爱等: "《网上零售理论与实战》", 31 January 2014, 中国科学技术大学出版社, pages: 189 * |
| 杜孟新;洪云峰;周晓明;: "面向消费防伪领域的客户价值分析", 福建电脑, no. 02 * |
| 田雨聪;耿子月;谢安泰;袁伊珊;张淑芬;: "基于数据挖掘的消费者价值细分模型研究", 软件, no. 08 * |
| 郑继刚: "《数据挖掘及其应用研究》", 31 May 2014, 云南大学出版社, pages: 50 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751515A (en) * | 2019-10-16 | 2020-02-04 | 北京百分点信息科技有限公司 | Decision-making method and device based on user consumption behaviors, electronic equipment and storage medium |
| CN110827070A (en) * | 2019-10-30 | 2020-02-21 | 神州数码融信软件有限公司 | User growth calculation method and device based on dynamic expansion factor |
| CN113781128A (en) * | 2021-10-15 | 2021-12-10 | 北京明略软件系统有限公司 | High-potential consumer identification method, system, electronic device, and medium |
| CN113688326A (en) * | 2021-10-26 | 2021-11-23 | 腾讯科技（深圳）有限公司 | Recommendation method, device, equipment and computer readable storage medium |
| CN113688326B (en) * | 2021-10-26 | 2022-02-08 | 腾讯科技（深圳）有限公司 | Recommendation method, device, equipment and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110111158A (en) | The Marketing Design method, apparatus and storage medium of life cycle or Development phase | |
| US20090319931A1 (en) | Visual Intelligence Systems | |
| Choi et al. | Optimal two-stage ordering policy with Bayesian information updating | |
| CN109189904A (en) | Individuation search method and system | |
| CN107678845A (en) | Application program management-control method, device, storage medium and electronic equipment | |
| CN108090800A (en) | A kind of game item method for pushing and device based on player's consumption potentiality | |
| CN105229640B (en) | Commercial product code analysis system and commercial product code analysis method | |
| CN106651418A (en) | Method of recommending add-on item for special offer when spending enough by e-business | |
| WO2006102381A2 (en) | Constrained tree structure method and system | |
| CN105225129A (en) | Mobile O2O recommend method and system thereof | |
| CN115391669B (en) | Intelligent recommendation method and device and electronic equipment | |
| CN111489201A (en) | Method, device and storage medium for analyzing customer value | |
| CN110852785A (en) | User grading method, device and computer readable storage medium | |
| CN108389069A (en) | Top-tier customer recognition methods based on random forest and logistic regression and device | |
| CN109255715A (en) | Electronic device, Products Show method and computer readable storage medium | |
| CN109740642A (en) | Invoice category recognition methods, device, electronic equipment and readable storage medium storing program for executing | |
| CN109977316A (en) | A kind of parallel type article recommended method, device, equipment and storage medium | |
| CN108364191A (en) | Top-tier customer Optimum Identification Method and device based on random forest and logistic regression | |
| CN102279963A (en) | Two-level budget rationality inspection prompting and automatic optimization method, equipment thereof and system thereof | |
| CN111652278A (en) | User behavior detection method and device, electronic equipment and medium | |
| CN108399553A (en) | It is a kind of to consider geographical and circuit subordinate relation user characteristics label setting method | |
| CN110659948A (en) | Calculation method for matching degree of commodity sold and false invoice risk discovery method | |
| CN115018588A (en) | Product recommendation method and device, electronic equipment and readable storage medium | |
| CN109509016A (en) | Sale processing method, apparatus and computer readable storage medium | |
| CN113450141B (en) | Intelligent prediction method and device based on electricity sales quantity characteristics of large power customer group |
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 |