CN106991577A - A kind of method and device for determining targeted customer - Google Patents
A kind of method and device for determining targeted customer Download PDFInfo
- Publication number
- CN106991577A CN106991577A CN201610039775.9A CN201610039775A CN106991577A CN 106991577 A CN106991577 A CN 106991577A CN 201610039775 A CN201610039775 A CN 201610039775A CN 106991577 A CN106991577 A CN 106991577A
- Authority
- CN
- China
- Prior art keywords
- user
- use information
- terminal
- history use
- information
- 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 32
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 29
- 238000003066 decision tree Methods 0.000 claims abstract description 28
- 238000004891 communication Methods 0.000 claims abstract description 27
- 230000008859 change Effects 0.000 claims abstract description 15
- 230000003542 behavioural effect Effects 0.000 claims description 55
- 238000012545 processing Methods 0.000 claims description 7
- 238000007418 data mining Methods 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 11
- 238000004458 analytical method Methods 0.000 description 9
- 230000006399 behavior Effects 0.000 description 6
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 5
- 238000004590 computer program Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 230000006978 adaptation Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000004215 lattice model Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 235000012054 meals Nutrition 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000004043 responsiveness 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
- G06Q30/0203—Market surveys; Market polls
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
本发明公开了一种确定目标用户的方法及装置，该方法包括：获取各终端的历史使用信息，历史使用信息包括用户使用终端的通讯信息和用户换机信息；根据预设的行为特征指标，利用聚类算法将用户使用终端的通讯信息进行聚类，得到每个行为特征指标对应的历史使用信息子集合；根据历史使用信息子集合中的用户使用终端的通讯信息，确定每个历史使用信息子集合的行为特征指标的取值；根据历史使用信息子集合中的用户换机信息，确定每个历史使用子集合对应的换机属性；并建立决策树模型；利用决策树模型分析终端的历史使用信息，从终端的使用用户中确定出有换机需求的目标用户，用以解决现有技术中存在目标客户确定方法不够精准，不利于准确营销的问题。
The present invention discloses a method and device for determining a target user. The method includes: obtaining historical usage information of each terminal, the historical usage information including communication information of the terminal used by the user and information of changing the user's phone; according to a preset behavior characteristic index, Use the clustering algorithm to cluster the communication information of the user's terminal to obtain a subset of historical usage information corresponding to each behavior characteristic index; determine each historical usage information according to the communication information of the user's terminal in the historical usage information subset The value of the behavior characteristic index of the subset; according to the user switch information in the historical use information subset, determine the switch attribute corresponding to each historical use subset; and establish a decision tree model; use the decision tree model to analyze the history of the terminal The information is used to determine the target users who need to change the phone from the users of the terminal, so as to solve the problem that the target customer determination method in the prior art is not accurate enough, which is not conducive to accurate marketing.
Description
Technical field
The present invention relates to communications network service field, more particularly to a kind of method and device for determining targeted customer.
Background technology
As the carrier of mobile service, mobile terminal be the exploitation in newly-increased market, stock market maintain and data and
One important point of penetration of information service development.Like mobile terminal to give a discount, send great number call charge allowance, shifting in past, operation commercial city
Dynamic 0 yuan of terminal is purchased, send material object.Under the background of current cost efficiency, this set marketing thinking is obviously improper, can neither effectively protect
There is client, high value customer can not be encouraged to migrate again.So, the mobile terminal marketing policy in 4G epoch is more accurate.Therefore
The target customer of mobile terminal marketing how is effectively found out, target customer is analyzed with machine preference, fortune is had become
The problem of seeking business's urgent need to resolve.
Existing mobile terminal targeted customer analysis method, mainly by business experience, extracts the passing business of user
The information such as usage behavior, business characteristic, the level of consumption judge whether user changes planes demand, and target mobile terminal consumption
Ability.Existing analysis method, relies on personal analysis experience and analysis preference, and the service attribute being related to is limited, determination methods
Lack of standardization, it is difficult to reach higher marketing accuracy rate, promotion effect is not ideal enough.
To sum up, existing targeted customer determines that method is not accurate enough, it is impossible to meet the purpose accurately marketed.
The content of the invention
The embodiment of the present invention provides a kind of method and device for determining targeted customer, to solve the presence of mesh in the prior art
Mark user determines that method is not accurate enough, the problem of being unfavorable for accurate marketing.
The inventive method includes a kind of method for determining targeted customer, and this method includes：The history for obtaining each terminal is used
Information, the communication information of the history use information including user's using terminal and user change planes information；According to default behavior
Characteristic index, is clustered the communication information of user's using terminal using clustering algorithm, is obtained each behavioural characteristic and is referred to
Mark corresponding history use information subclass；The communication letter of user's using terminal in the history use information subclass
Breath, it is determined that the value of the behavioural characteristic index of each history use information subclass；According to the history use information subclass
In user change planes information, it is determined that each history uses the corresponding attribute of changing planes of subclass；According to the default behavioural characteristic
Index, the value of the behavioural characteristic index of the history use information subclass and each history are corresponding using subclass
Attribute of changing planes sets up decision-tree model；Using the history use information of the decision-tree model analysing terminal, from the terminal
Use the targeted customer that the demand of changing planes is determined in user.
Based on same inventive concept, the embodiment of the present invention further provides a kind of device for determining targeted customer, should
Device includes：Acquiring unit, the history use information for obtaining each terminal, the history use information includes user and used eventually
The communication information at end and user change planes information；Cluster cell, for according to default behavioural characteristic index, being incited somebody to action using clustering algorithm
The communication information of user's using terminal is clustered, and obtains the corresponding history use information subset of each behavioural characteristic index
Close；Value unit is determined, for the communication information of user's using terminal in the history use information subclass, it is determined that
The value of the behavioural characteristic index of each history use information subclass；Template(-let) is determined, for being used according to the history
User during information subset is closed changes planes information, it is determined that each history uses the corresponding attribute of changing planes of subclass；Set up model unit,
For according to the default behavioural characteristic index, the value of the behavioural characteristic index of the history use information subclass and institute
State each history and set up decision-tree model using the corresponding attribute of changing planes of subclass；Target subscriber units are determined, for utilizing
The history use information of decision-tree model analysing terminal is stated, the mesh for the demand of changing planes is determined from the use user of the terminal
Mark user.
On the one hand the embodiment of the present invention obtains the history use information of each terminal, will be gone through using default behavioural characteristic index
History use information is clustered into the history use information subclass related to behavioural characteristic index；On the other hand, each history is utilized
The behavioural characteristic index of use information subclass, the value of behavioural characteristic index, attribute of changing planes set up decision-tree model, so sharp
The history use information of other users is analyzed with the decision-tree model built up, it is possible to draw the attribute of changing planes of the user, find
Really there is the targeted customer for the demand of changing planes.It can be seen that, because make use of the history use information of historic user to set up decision-tree model,
The model can be with Reusability, and the result accuracy rate that analysis is obtained is high, instead of existing artificial experience and distinguishes so that target
User determines that result is more accurate, also beneficial to the formulation of various marketing strategies.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, makes required in being described below to embodiment
Accompanying drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for this
For the those of ordinary skill in field, without having to pay creative labor, it can also be obtained according to these accompanying drawings
His accompanying drawing.
Fig. 1 provides a kind of method flow schematic diagram for determining targeted customer for the embodiment of the present invention；
Fig. 2 provides the schematic diagram that a kind of history use information subclass is quantified as three-dimensional array for the embodiment of the present invention；
Fig. 3 provides a kind of clustering method schematic flow sheet for the embodiment of the present invention；
Fig. 4 provides a kind of generation close quarters schematic diagram for the embodiment of the present invention；
Fig. 5 provides a kind of decision tree schematic diagram for the embodiment of the present invention；
Fig. 6 provides a kind of schematic device for determining targeted customer for the embodiment of the present invention.
Embodiment
In order that the object, technical solutions and advantages of the present invention are clearer, below in conjunction with accompanying drawing the present invention is made into
One step it is described in detail, it is clear that described embodiment is only embodiment of the invention a part of, rather than whole implementation
Example.Based on the embodiment in the present invention, what those of ordinary skill in the art were obtained under the premise of creative work is not made
All other embodiment, belongs to the scope of protection of the invention.
Shown in Figure 1, the embodiment of the present invention provides a kind of method flow schematic diagram for determining targeted customer, specifically real
Existing method includes：
Step S101, obtains the history use information of each terminal, and the history use information includes user's using terminal
Communication information and user change planes information.
Step S102, according to default behavioural characteristic index, using clustering algorithm by the communication of user's using terminal
Information is clustered, and obtains the corresponding history use information subclass of each behavioural characteristic index.
Step S103, the communication information of user's using terminal in the history use information subclass, it is determined that often
The value of the behavioural characteristic index of individual history use information subclass.
Step S104, the user in the history use information subclass changes planes information, it is determined that each history is used
The corresponding attribute of changing planes of subclass.
Step S105, according to the default behavioural characteristic index, the behavioural characteristic of the history use information subclass
The value of index and each history set up decision-tree model using the corresponding attribute of changing planes of subclass.
Step S106, using the history use information of the decision-tree model analysing terminal, is used from the use of the terminal
The targeted customer for the demand of changing planes is determined in family.
In step S101, the information of historic user is obtained from operator's platform, then finding these historic users makes
The history use information of terminal, the history use information of terminal is primarily referred to as user's telephone expenses bill information monthly and user
The terminal type used, e.g. so-called terminal type, 4G cell phone or 3G mobile, mobile phone brand, model, the color used
Deng the information such as sex, age, the color preference of user can be also obtained from the log-on message of user in addition.
Because the data volume that history use information is included is huge, and without certain regularity, analysis gets up sufficiently complex,
Therefore the embodiment of the present invention is classified original history use information using default behavioural characteristic index, wherein default
Behavioural characteristic index can be the call behavior of user, short message behavior, rule of being in, evolution rule, the moon telephone expenses amount of money, on
Net is monthly using indexs such as flows.Because what the corresponding data of different indexs reflected such user uses machine feature, knot
Closing different indexs can determine that the basic of such user uses machine feature.For example, the currently used 3G mobile of user's Mr. Wang, 12
The month telephone expenses amount of money is 120 yuan, wherein monthly monthly rent is 58 yuan, 30 yuan of cost of the phone call, moon flow monthly payment is 15 yuan, and short message is taken
With 17 yuan.It can substantially determine that the moon telephone expenses amount of money of the user is very high, monthly rent is higher, consuming capacity is strong, there is certain purchase
Power.
So-called attribute of changing planes refers to the feature of changing planes of such customer group, because can be utilized when setting up decision tree useful
The feature of changing planes at family.For example according to general experience, it is known that the user of frequent hand-off machine, when the mobile phone of new type meets the needs of the market
During sale, this kind of user can be main purchasing group, such as during annual iPhone issue, before using the user of iPhone
It may proceed to buy the iPhone of new model；Either some cellphone subscriber's surfing flow every month is larger, it was demonstrated that the user
Commonly using surfing Internet with cell phone, if the user it is currently used be 3G mobile, then in view of the faster advantage of 4G cell phone network speed, that
The type user, which is substantially all, can change the mobile phone of 4G types into.In view of these historical experiences, the history in historic user
Use information reflection behavioural characteristic index the characteristics of, be each user stamp whether the label of hand-off machine, that is, category of changing planes
Property, thus decision-tree model can be set up based on historic user.
Certainly, when obtaining the history use information of each terminal, in addition it is also necessary to using history of the method sampled to each terminal
Default value or exceptional value in use information are filtered, the history use information of each terminal after being handled.Because history
Default value either exceptional value in use information can not accurately give expression to the behavioural characteristic using user, use such as some moon
The bill at family is 0, and the value is likely due to operation system failure and caused, therefore can not be placed on the set of history use information
It is middle to be clustered.This have the effect that the accuracy that can effectively improve the targeted customer finally determined.
Further, for each behavioural characteristic index, by the communication information of user's using terminal with three-dimensional array
Mode expression；Three dimensions cluster is carried out to the data in the three-dimensional array according to k-d algorithms, k cluster is obtained, its
In, one history use information subclass of each cluster correspondence, k is integer.
Specifically, the embodiment of the present invention carries out the cluster of different information in history use information, k- using k-means algorithms
The method of means algorithms cluster probably includes following several steps：
Step 1: history use information is stored as into a data set D, cluster number k and k initial clustering are then given
Center；
Step 2: calculating each data to the distance of cluster centre, and each object is assigned to most according to distance minimum
Close cluster；
Step 3: recalculating the average of each cluster and determining new cluster centre；Calculation error sum-of-squares criterion letter
Number J；
Step 4: k cluster set of output.
It is one group and personality phase by default behavioural characteristic target setting for example, studied and sampled according to personality theory
The corresponding history use information subclass of each desired value, is quantified as three-dimensional array K (x, y, z), such as Fig. 2 institutes by the index of pass
Show, realized by setting up K-d trees to each user's index progress data tissue, realize the matching of data similar features.Feature
Matching is substantially one and similarity retrieval is carried out between high n dimensional vector n by space length function, and forms cluster set, its
Flow be illustrated in fig. 3 shown below including：
Step 201, study and sample according to personality theory, be that history use information determines behavioural characteristic index；
Step 202, the history use information subclass on each behavioural characteristic index is then obtained；
Step 203, history use information subclass is quantified as Multidimensional numerical K (x, y, z), history use information subclass
In each data be recorded in the space cube of Multidimensional numerical；
Step 204, Multidimensional numerical is clustered, obtains the close quarters in history use information subclass.
So, the corresponding history use information subclass of different personality reference indexs can just be quantified as three-dimensional respectively
Array, X characters reference index then corresponds to X three-dimensional array.
Further, each three-dimensional data is subjected to a point heap sort again, the index number of client is stored by building k-d tree
According to, then realize by the neighbor searching algorithm of k-d tree the cluster client partitions of data.For example combine seven adult's lattice models reason
By, customer action characteristic index, verified by sample investigation, it is determined that 7 class close quarterses, correspond to 7 class personality, what is be divided into is close
Cluster is as shown in Figure 4.
Such as, for the corresponding history use information subclass of call behavior this behavioural characteristic index, using above
Middle three-dimensional array quantifies, then x can represent the duration of call, y and represent talk times, z and represents opening of conversing every time in K (x, y, z)
Begin the moment, then be possible to include toll message user by the 7 intensive groups obtained after the communicating data cluster in K (x, y, z)
Group's information such as group, local connection customer group, international call customer group.
After above-mentioned history use information is finely divided, it is possible to according to the history use information subset after subdivision
Close corresponding value, behavioural characteristic index, attribute of changing planes and set up decision-tree model, for example, choose certain telecommunications company client in 2006
Payment data, choose 10000 samples, after data cleansing, there remains 8725 samples, and the call to these clients is remembered
Record, short message behavior, billing amount are analyzed, and are subdivided into following 5 customers using K-means algorithms, are shown in Table one：
Corresponding behavioural characteristic index in table one (possesses 4G cell phone, frequency of changing planes, the moon telephone expenses amount of money, online bag
Month flow, demand of changing planes etc.) and the corresponding behavioural characteristic index of each group value, the demand of changing planes sets up decision-tree model, such as
Shown in Fig. 5., wherein because customers I possesses 4G cell phone in table one, therefore this kind of user does not change the demand of 4G cell phone, remaining
Customers then further segment, and customers II frequency of changing planes is very high, so have the demand for changing 4G cell phone, the like, it is complete
Into the foundation of decision-tree model, as the user that newly arrived, two hands every year on average are found in the history use information of the user
Machine, then can learn that the user is changed planes demand, and target is defined as by the user according to the analysis result of decision-tree model
User.
After targeted customer is determined, it is desirable to have specific aim is that targeted customer determines the end message to be marketed, it is determined that
The method of end message is specially：According to frequent item set data mining algorithm, set up before changing planes with machine feature and after changing planes
The corresponding relation of end message；According to the history use information of the targeted customer, before determining the targeted customer using changing planes
Use machine feature；End message after determining that the targeted customer is corresponding according to the corresponding relation and changing planes, so as to described
Targeted customer pushes the corresponding end message of the targeted customer.
Wherein, the Frequent Itemsets Mining Algorithm can be realized using FP-growth algorithms or Apriori algorithm.FP-
Growth algorithms are a kind of new Frequent Itemsets Mining Algorithms proposed in 2000, and it must be produced departing from Apriori algorithm
The traditional approach of candidate, establishes the thought for not producing candidate based on FP-tree structures, opens correlation rule
New approaches, belong to prior art, do not repeat herein.
Determining the method for end message can be further described through by following example, for example, according to the history before changing planes
Use information, is first divided into 8 colonies, respectively more than 4000 yuan users of terminal price by the user before changing planes by terminal price section
Group, 3000-4000 members customer group, 2000-3000 members customer group, 1000-2000 members customer group, 500-1000 members customer group, 500
The following customer group of member, price unknown subscriber group, then price category after analyzing attribute before each customer group is changed planes, behavioural characteristic and changing planes
Between corresponding relation.The price for the terminal that targeted customer may buy is determined according to corresponding relation.Wherein corresponding relation
Determination mode is：For renewal user, the end after obtaining the history use information before renewal user changes planes and changing planes
Client information；According to the history use information before renewal user changes planes, it is determined that the use machine before the changing planes of renewal user is special
Levy；The corresponding relation for the end message with machine feature and after changing planes set up before changing planes.
For example, user A change planes before with the mobile phone price used in machine feature 1000-2000 member between, belong to 1000-
2000 yuan of customer groups, the mobile phone price after changing planes is still between 1000-2000, it is seen that this kind of user possesses such machine
Feature, therefore mobile phone of the price between 1000-2000 is promoted and gives the user A.
Similarly, operating system preference of the user to the terminal that uses, screen size can also be calculated in the same way
The corresponding relations such as preference, network formats preference, that is to say, that be equal to each user stamp brand of knowing clearly, price, operating system,
The adaptation label of screen size, network formats etc..So, terminal sale personnel can be according to existing end message on sale, from end
The different dimensions of attribute are held to meet desired client to screen.Realize that terminal on sale and the accurate of client are fitted according to actual conditions
Match somebody with somebody, improve marketing responsiveness.
Certain above-mentioned targeted customer for being to determine the demand of changing planes, if being found according to the behavioural characteristic of the customer group after cluster
There is no long-distance telephone expenses more in the bill of certain customers, then preferential set meal long-distance to such Customer design, stimulate that client beats more
Long-distance call.If it was found that user's use information subclass of some customer group reflect short message expense it is higher, then can be to short message
Take high lead referral superposition and use short message monthly package.
It can be seen that, the embodiment of the present invention can effectively find out the potential of terminal marketing on the basis of each customers are segmented
Client, and then ensure the result that terminal is recommended.I.e. the embodiment of the present invention is thin using client's character analysis based on k-means algorithms
Different client groups, set up client's personality label, and combine the potential customers that terminal marketing is identified based on decision tree C5.0 algorithms.
After the potential customers for identifying terminal marketing, the terminal adaptation degree of target customer is further determined, and then realizes that terminal is pushed away
Recommend, find out the use before being changed planes with machine feature and client after client changes planes in the embodiment of the present invention using Apriori association algorithms
Incidence relation between machine feature, the terminal type of Accurate Prediction adaptation.
Based on identical technical concept, the embodiment of the present invention also provides a kind of device for determining targeted customer, and the device can
Perform above method embodiment.Device provided in an embodiment of the present invention as shown in fig. 6, including：Acquiring unit 401, cluster cell
402nd, value unit 403 is determined, template(-let) 404 is determined, sets up model unit 405, determines target subscriber units 406, wherein：
Acquiring unit 401, the history use information for obtaining each terminal, the history use information is used including user
The communication information of terminal and user change planes information；
Cluster cell 402, for according to default behavioural characteristic index, using clustering algorithm by user's using terminal
Communication information clustered, obtain the corresponding history use information subclass of each behavioural characteristic index；
Value unit 403 is determined, the communication for user's using terminal in the history use information subclass
Information, it is determined that the value of the behavioural characteristic index of each history use information subclass；
Template(-let) 404 is determined, is changed planes information for the user in the history use information subclass, it is determined that often
Individual history uses the corresponding attribute of changing planes of subclass；
Model unit 405 is set up, for according to the default behavioural characteristic index, the history use information subclass
Behavioural characteristic index value and each history set up decision-tree model using the corresponding attribute of changing planes of subclass；
Target subscriber units 406 are determined, for the history use information using the decision-tree model analysing terminal, from institute
The targeted customer for the demand of changing planes is determined in the use user for stating terminal.
Wherein, the acquiring unit 401 is additionally operable to：Using the device of sampling to lacking in the history use information of each terminal
Province's value or exceptional value are filtered, the history use information of each terminal after being handled.
Further, the cluster cell 402 specifically for：For each behavioural characteristic index, the user is used
The communication information of terminal representing in the way of three-dimensional array；
Three dimensions cluster is carried out to the data in the three-dimensional array according to k-d algorithms, k cluster is obtained, wherein, often
One history use information subclass of individual cluster correspondence, k is integer.
Further, in addition to：End message unit 407 is determined, for according to frequent item set data mining algorithm, setting up
The corresponding relation of the end message with machine feature and after changing planes before changing planes；According to the history use information of the targeted customer,
Determine that the targeted customer uses machine feature using before changing planes；Determine that the targeted customer is corresponding according to the corresponding relation to change
End message after machine, so as to push the corresponding end message of the targeted customer to the targeted customer.
Wherein, it is described determination end message unit specifically for：For renewal user, renewal user changes described in acquisition
History use information before machine and the end message after changing planes；According to the history use information before renewal user changes planes,
It is determined that using machine feature before the changing planes of renewal user；Set up change planes before with machine feature with change planes after end message it is corresponding
Relation.
On the one hand the embodiment of the present invention obtains the history use information of each terminal, will be gone through using default behavioural characteristic index
History use information is clustered into the history use information subclass related to behavioural characteristic index；On the other hand, each history is utilized
The behavioural characteristic index of use information subclass, the value of behavioural characteristic index, attribute of changing planes set up decision-tree model, so sharp
The history use information of other users is analyzed with the decision-tree model built up, it is possible to draw the attribute of changing planes of the user, find
Really there is the targeted customer for the demand of changing planes.It can be seen that, because make use of the history use information of historic user to set up decision-tree model,
The model can be with Reusability, and the result accuracy rate that analysis is obtained is high, instead of existing artificial experience and distinguishes so that target
Client determines that result is more accurate, also beneficial to the formulation of various marketing strategies.
The present invention is the flow with reference to method according to embodiments of the present invention, equipment (system) and computer program product
Figure and/or block diagram are described.It should be understood that can be by every first-class in computer program instructions implementation process figure and/or block diagram
Journey and/or the flow in square frame and flow chart and/or block diagram and/or the combination of square frame.These computer programs can be provided
The processor of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices is instructed to produce
A raw machine so that produced by the instruction of computer or the computing device of other programmable data processing devices for real
The device for the function of being specified in present one flow of flow chart or one square frame of multiple flows and/or block diagram or multiple square frames.
These computer program instructions, which may be alternatively stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory, which is produced, to be included referring to
Make the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one square frame of block diagram or
The function of being specified in multiple square frames.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that in meter
Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented processing, thus in computer or
The instruction performed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in individual square frame or multiple square frames.
, but those skilled in the art once know basic creation although preferred embodiments of the present invention have been described
Property concept, then can make other change and modification to these embodiments.So, appended claims are intended to be construed to include excellent
Select embodiment and fall into having altered and changing for the scope of the invention.
Obviously, those skilled in the art can carry out the essence of various changes and modification without departing from the present invention to the present invention
God and scope.So, if these modifications and variations of the present invention belong to the scope of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to comprising including these changes and modification.
Claims (10)
1. a kind of method for determining targeted customer, it is characterised in that this method includes：
The history use information of each terminal is obtained, the history use information includes communication information and the user of user's using terminal
Change planes information；
According to default behavioural characteristic index, the communication information of user's using terminal is clustered using clustering algorithm,
Obtain the corresponding history use information subclass of each behavioural characteristic index；
The communication information of user's using terminal in the history use information subclass, it is determined that each history use information
The value of the behavioural characteristic index of subclass；
User in the history use information subclass changes planes information, it is determined that each history is changed using subclass is corresponding
Machine attribute；
According to the default behavioural characteristic index, the value of the behavioural characteristic index of the history use information subclass and institute
State each history and set up decision-tree model using the corresponding attribute of changing planes of subclass；
Using the history use information of the decision-tree model analysing terminal, determine and changed from the use user of the terminal
The targeted customer of machine demand.
2. the method as described in claim 1, it is characterised in that described according to default behavioural characteristic index, is calculated using cluster
Method is clustered the communication information of user's using terminal, obtains the corresponding history use information of each behavioural characteristic index
Subclass, including：
For each behavioural characteristic index, by communication information the representing in the way of three-dimensional array of user's using terminal；
Three dimensions cluster is carried out to the data in the three-dimensional array according to k-d clustering algorithms, k cluster is obtained, wherein, often
One history use information subclass of individual cluster correspondence, k is integer.
3. the method as described in claim 1, it is characterised in that the history use information of each terminal of acquisition, in addition to：
The default value or exceptional value in the history use information of each terminal are filtered using the method for sampling, obtained after processing
Each terminal history use information.
4. the method as described in claim 1, it is characterised in that needed if having determined and having changed planes from the use user of the terminal
After the targeted customer asked, in addition to：
According to frequent item set data mining algorithm, being closed with machine feature the corresponding of end message with after changing planes before changing planes is set up
System；
According to the history use information of the targeted customer, determine that the targeted customer uses machine feature using before changing planes；
End message after determining that the targeted customer is corresponding according to the corresponding relation and changing planes, so as to the targeted customer
Push the corresponding end message of the targeted customer.
5. method as claimed in claim 4, it is characterised in that according to frequent item set data mining algorithm, before foundation is changed planes
The corresponding relation of end message with machine feature and after changing planes, including：
For renewal user, the end message after obtaining the history use information before renewal user changes planes and changing planes；
According to the history use information before renewal user changes planes, it is determined that using machine feature before the changing planes of renewal user；
The corresponding relation for the end message with machine feature and after changing planes set up before changing planes.
6. a kind of device for determining targeted customer, it is characterised in that the device includes：
Acquiring unit, the history use information for obtaining each terminal, the history use information includes user's using terminal
Communication information and user change planes information；
Cluster cell, for according to default behavioural characteristic index, using clustering algorithm by the communication of user's using terminal
Information is clustered, and obtains the corresponding history use information subclass of each behavioural characteristic index；
Value unit is determined, for the communication information of user's using terminal in the history use information subclass, really
The value of the behavioural characteristic index of fixed each history use information subclass；
Template(-let) is determined, is changed planes information for the user in the history use information subclass, it is determined that each history
Use the corresponding attribute of changing planes of subclass；
Model unit is set up, for the behavior according to the default behavioural characteristic index, the history use information subclass
The value of characteristic index and each history set up decision-tree model using the corresponding attribute of changing planes of subclass；
Target subscriber units are determined, for the history use information using the decision-tree model analysing terminal, from the terminal
Use user in determine the targeted customer of the demand of changing planes.
7. device as claimed in claim 6, it is characterised in that the cluster cell specifically for：
For each behavioural characteristic index, by communication information the representing in the way of three-dimensional array of user's using terminal；
Three dimensions cluster is carried out to the data in the three-dimensional array according to k-d clustering algorithms, k cluster is obtained, wherein, often
One history use information subclass of individual cluster correspondence, k is integer.
8. device as claimed in claim 6, it is characterised in that the acquiring unit is additionally operable to：
The default value or exceptional value in the history use information of each terminal are filtered using the device of sampling, obtained after processing
Each terminal history use information.
9. device as claimed in claim 6, it is characterised in that also include：
Determine end message unit, for according to frequent item set data mining algorithm, set up before changing planes with machine feature with changing planes
The corresponding relation of end message afterwards；
According to the history use information of the targeted customer, determine that the targeted customer uses machine feature using before changing planes；
End message after determining that the targeted customer is corresponding according to the corresponding relation and changing planes, so as to the targeted customer
Push the corresponding end message of the targeted customer.
10. device as claimed in claim 9, it is characterised in that the determination end message unit is additionally operable to：
For renewal user, the end message after obtaining the history use information before renewal user changes planes and changing planes；
According to the history use information before renewal user changes planes, it is determined that using machine feature before the changing planes of renewal user；
The corresponding relation for the end message with machine feature and after changing planes set up before changing planes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610039775.9A CN106991577A (en) | 2016-01-21 | 2016-01-21 | A kind of method and device for determining targeted customer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610039775.9A CN106991577A (en) | 2016-01-21 | 2016-01-21 | A kind of method and device for determining targeted customer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106991577A true CN106991577A (en) | 2017-07-28 |
Family
ID=59414284
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610039775.9A Pending CN106991577A (en) | 2016-01-21 | 2016-01-21 | A kind of method and device for determining targeted customer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106991577A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451612A (en) * | 2017-07-31 | 2017-12-08 | 陕西识代运筹信息科技股份有限公司 | A kind for the treatment of method and apparatus that targeted customer is determined based on concern relation |
| CN107665185A (en) * | 2017-10-19 | 2018-02-06 | 上海大汉三通数据通信有限公司 | A kind of customer behavior analysis method and relevant apparatus based on short message service |
| CN108038709A (en) * | 2017-11-03 | 2018-05-15 | 平安科技（深圳）有限公司 | Client's sampling pilot marketing method, electronic device and computer-readable recording medium |
| CN110060093A (en) * | 2019-03-25 | 2019-07-26 | 广州瀚信通信科技股份有限公司 | A kind of terminal marketing method based on 4G high flow capacity client |
| CN110648153A (en) * | 2018-06-11 | 2020-01-03 | 中国移动通信集团浙江有限公司 | A method, device, electronic device and storage medium for predicting replacement tendency |
| TWI716057B (en) * | 2018-10-25 | 2021-01-11 | 開曼群島商創新先進技術有限公司 | Service recommendation method, device and equipment |
| CN113469475A (en) * | 2020-03-30 | 2021-10-01 | 北京骑胜科技有限公司 | Data processing method, data processing device, storage medium and electronic equipment |
| CN114066529A (en) * | 2021-11-26 | 2022-02-18 | 中国联合网络通信集团有限公司 | A terminal replacement prediction method, device and storage medium |
| CN114727274A (en) * | 2022-04-07 | 2022-07-08 | 中国联合网络通信集团有限公司 | User migration method, device, electronic device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030139963A1 (en) * | 2000-12-08 | 2003-07-24 | Chickering D. Maxwell | Decision theoretic approach to targeted solicitation by maximizing expected profit increases |
| US20100145773A1 (en) * | 2000-12-20 | 2010-06-10 | Paritosh Desai | System and Method for Generating Product Decisions |
| CN104794195A (en) * | 2015-04-17 | 2015-07-22 | 南京大学 | Data mining method for finding potential telecommunication users changing cell phones |
-
2016
- 2016-01-21 CN CN201610039775.9A patent/CN106991577A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030139963A1 (en) * | 2000-12-08 | 2003-07-24 | Chickering D. Maxwell | Decision theoretic approach to targeted solicitation by maximizing expected profit increases |
| US20100145773A1 (en) * | 2000-12-20 | 2010-06-10 | Paritosh Desai | System and Method for Generating Product Decisions |
| CN104794195A (en) * | 2015-04-17 | 2015-07-22 | 南京大学 | Data mining method for finding potential telecommunication users changing cell phones |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451612A (en) * | 2017-07-31 | 2017-12-08 | 陕西识代运筹信息科技股份有限公司 | A kind for the treatment of method and apparatus that targeted customer is determined based on concern relation |
| CN107665185A (en) * | 2017-10-19 | 2018-02-06 | 上海大汉三通数据通信有限公司 | A kind of customer behavior analysis method and relevant apparatus based on short message service |
| CN108038709A (en) * | 2017-11-03 | 2018-05-15 | 平安科技（深圳）有限公司 | Client's sampling pilot marketing method, electronic device and computer-readable recording medium |
| CN110648153A (en) * | 2018-06-11 | 2020-01-03 | 中国移动通信集团浙江有限公司 | A method, device, electronic device and storage medium for predicting replacement tendency |
| TWI716057B (en) * | 2018-10-25 | 2021-01-11 | 開曼群島商創新先進技術有限公司 | Service recommendation method, device and equipment |
| CN110060093A (en) * | 2019-03-25 | 2019-07-26 | 广州瀚信通信科技股份有限公司 | A kind of terminal marketing method based on 4G high flow capacity client |
| CN110060093B (en) * | 2019-03-25 | 2023-12-05 | 广州瀚信通信科技股份有限公司 | Terminal marketing method based on 4G high-flow clients |
| CN113469475A (en) * | 2020-03-30 | 2021-10-01 | 北京骑胜科技有限公司 | Data processing method, data processing device, storage medium and electronic equipment |
| CN113469475B (en) * | 2020-03-30 | 2024-05-14 | 北京骑胜科技有限公司 | Data processing method, device, storage medium and electronic equipment |
| CN114066529A (en) * | 2021-11-26 | 2022-02-18 | 中国联合网络通信集团有限公司 | A terminal replacement prediction method, device and storage medium |
| CN114727274A (en) * | 2022-04-07 | 2022-07-08 | 中国联合网络通信集团有限公司 | User migration method, device, electronic device and storage medium |
| CN114727274B (en) * | 2022-04-07 | 2023-07-21 | 中国联合网络通信集团有限公司 | User migration method, device, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106991577A (en) | A kind of method and device for determining targeted customer | |
| CN111612549B (en) | Construction method of platform operation service system | |
| CN110222267A (en) | A kind of gaming platform information-pushing method, system, storage medium and equipment | |
| CN106845731A (en) | A kind of potential renewal user based on multi-model fusion has found method | |
| CN105184315A (en) | Quality inspection treatment method and system | |
| CN108595461A (en) | Interest heuristic approach, storage medium, electronic equipment and system | |
| CN110766272A (en) | Power business collaborative classification method and system based on ID3 decision tree algorithm | |
| CN110766438B (en) | Method for analyzing user behavior of power grid user through artificial intelligence | |
| CN102402594A (en) | Rich media personalized recommendation method | |
| CN108764332A (en) | A kind of Channel Quality analysis method, computing device and storage medium | |
| Tong et al. | The research of customer loyalty improvement in telecom industry based on NPS data mining | |
| CN106408325A (en) | User consumption behavior prediction analysis method based on user payment information and system | |
| CN109948913A (en) | A comprehensive portrait system of multi-source power users based on double-layer xgboost algorithm | |
| CN112153636A (en) | Method for predicting number portability and roll-out of telecommunication industry user based on machine learning | |
| CN110019996A (en) | A kind of family relationship recognition methods and system | |
| CN116245399A (en) | Model training method and device, nonvolatile storage medium and electronic equipment | |
| CN106649380A (en) | Hot spot recommendation method and system based on tag | |
| CN111078859A (en) | Author recommendation method based on reference times | |
| CN110377828A (en) | Information recommendation method, device, server and storage medium | |
| CN107871055A (en) | A kind of data analysing method and device | |
| CN108710606A (en) | A kind of Task Progress monitoring method, computer readable storage medium and terminal device | |
| CN110457387B (en) | Method and related device applied to user tag determination in network | |
| CN114048294B (en) | Similar population extension model training method, similar population extension method and device | |
| CN102567425A (en) | Method and device for processing data | |
| Pan et al. | Iterative innovation design methods of internet products in the era of big data |
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: 20170728 |