CN107808346A - A kind of appraisal procedure and apparatus for evaluating of potential target object - Google Patents
A kind of appraisal procedure and apparatus for evaluating of potential target object Download PDFInfo
- Publication number
- CN107808346A CN107808346A CN201610814375.0A CN201610814375A CN107808346A CN 107808346 A CN107808346 A CN 107808346A CN 201610814375 A CN201610814375 A CN 201610814375A CN 107808346 A CN107808346 A CN 107808346A
- Authority
- CN
- China
- Prior art keywords
- mutual
- rfm
- user
- action behavior
- interactive
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000002452 interceptive effect Effects 0.000 claims abstract description 137
- 238000012360 testing method Methods 0.000 claims description 47
- 238000011156 evaluation Methods 0.000 claims description 19
- 230000003993 interaction Effects 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 13
- 230000000694 effects Effects 0.000 claims description 4
- 230000006399 behavior Effects 0.000 description 194
- 230000009471 action Effects 0.000 description 13
- 238000003860 storage Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 5
- 244000046052 Phaseolus vulgaris Species 0.000 description 4
- 235000010627 Phaseolus vulgaris Nutrition 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 210000003813 thumb Anatomy 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000014509 gene expression Effects 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000007306 turnover Effects 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (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
The invention discloses a kind of appraisal procedure and apparatus for evaluating of potential target object, for solving the problems, such as accurately timely not evaluating by social media valuable latent objective in the prior art, according to mutual-action behavior group information of the user in social media and interactive grade RFM assessment models, user's classification is carried out to user, enterprise is allowd to obtain accurately and timely valuable latent visitor by social media channel, so as to improve the sales service of enterprise level.The appraisal procedure of potential target object includes：Obtain mutual-action behavior group information of the user in social media；According to interactive grade RFM assessment models and mutual-action behavior group information, the interactive degree of the RFM codes of user, RFM representations user and social media is obtained；User's classification is carried out to user according to RFM codes.
Description
Technical field
The present invention relates to computer realm, and in particular to a kind of appraisal procedure and assessment dress based on potential target object
Put.
Background technology
Latent visitor, full name is potential customers, and English generally uses leads (clue), is often referred to have enterprise product or service
Demand and the client to be developed for possessing purchasing power, it is the object that enterprise market selling group makes great efforts to obtain and convert.
Two kinds of technical schemes that existing enterprise carries out client segmentation are：1、RFM(Recency Frequency Money)
Model, the customer value that traditional RFM models are used for individual consumer's goods product by enterprise is assessed, by client segmentation.Practices well
It is as follows：Client is sorted respectively according to parameter R, F and M, for parameter R, the time buying is more near, sorts more forward, for parameter
F, purchase number more at most sort about forward, and for parameter M, the purchase amount of money is more high, sorts about forward；Then by R, F, M row
Sequence result is respectively divided into five deciles, and R5 represents time buying nearest client, and R1 represents time buying farthest client, and F5 represents purchase
The most client of number, the minimum client of F1 generation table purchase number are bought, M5 represents purchase amount of money highest client, and M1 represents purchase
The minimum client of the amount of money；So each client can obtain the typical value of tri- latitudes of R, F, M, it is assumed that three generations of certain client
Tabular value is respectively R4, F3, M5, then the RFM values of this client are 435, represent that this client buys, buys the frequency one recently
As but the higher client of spending amount, enterprise can be according to RFM values to client segmentation, and take different marketing methods respectively；
2nd, attributive classification method, according to SCRM (socialization customer relation management) according to client location, sex, social networks type, year
The simple classification such as age and nearest behavior.
It is not to be designed for latent visitor's classification, more but RFM category of model methods are the consumer behaviors based on client
It is inclined to a kind of general client's assessment models；And attributive classification method is essentially all using static data as basis of classification, it is applicable
In preliminary screening, and can not dynamic regulation client segmentation, therefore, enterprise can not evaluate valuable latent visitor, and can not and
When regulation client segmentation, the sales service for influenceing enterprise is horizontal.
The content of the invention
The invention provides a kind of appraisal procedure and apparatus for evaluating of potential target object, for solving in the prior art
The problem of valuable latent objective, timely accurately can not be evaluated by social media, according to user social media interaction
Behavior group information and interactive grade RFM assessment models, user's classification is carried out to user so that enterprise can by social media channel
To obtain accurately and timely valuable latent visitor, so as to improve the sales service of enterprise level.
First aspect present invention provides a kind of appraisal procedure of potential target object, including：
Obtain user social media mutual-action behavior group information, the mutual-action behavior group information be the user with it is described
The interactive information of social media；
According to interactive grade RFM assessment models and the mutual-action behavior group information, the RFM codes of the user are obtained, it is described
User described in RFM representations and the interactive degree of the social media；
User's classification is carried out to the user according to the RFM codes.
Second aspect of the present invention provides a kind of apparatus for evaluating of potential target object, including：
Acquisition module, for obtaining mutual-action behavior group information of the user in social media, the mutual-action behavior group information is
The user information interactive with the social media；
Processing module, for according to interactive grade RFM assessment models and the mutual-action behavior group information, obtaining the user
RFM codes, the interactive degree of user described in the RFM representations and the social media；
Sort module, for carrying out user's classification to the user according to the RFM codes.
As can be seen from the above technical solutions, the embodiment of the present invention has advantages below：
Unlike prior art 1, user classification be according to user social media mutual-action behavior group information and mutually
Obtained from dynamic grade RFM assessment models, enterprise is according to user's classification it is known that the potential value of each user；With it is existing
Unlike technology 2, because user's classification needs to use mutual-action behavior group information of the user in social media, and mutual-action behavior group
Information is to be obtained according to user using the situation of social media, and in practice, user is dynamic using the situation of social media
, thus the user that obtains of dynamic classifies, and therefore, enterprise can accurately and timely be selected by social media channel
Valuable latent visitor, so as to improve the sales service of enterprise level.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, make required in being described below to embodiment
Accompanying drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for
For those skilled in the art, on the premise of not paying creative work, it can also be obtained according to these accompanying drawings other attached
Figure.
Fig. 1 is a structural representation of system in the embodiment of the present invention；
Fig. 2 is one embodiment schematic diagram of the appraisal procedure of potential target object in the embodiment of the present invention；
Fig. 3 is action trail schematic diagram of the user in wechat public number in the embodiment of the present invention；
Fig. 4 is action trail schematic diagram of the user on QQ group in the embodiment of the present invention；
Fig. 5 is action trail schematic diagram of the user on QQ in the embodiment of the present invention；
Fig. 6 is the schematic diagram of mutual-action behavior group of the user in wechat public number in the embodiment of the present invention；
Fig. 7 is a structural representation of the apparatus for evaluating of potential target object in the embodiment of the present invention；
Fig. 8 is another structural representation of the apparatus for evaluating of potential target object in the embodiment of the present invention；
Fig. 9 is another structural representation of the apparatus for evaluating of potential target object in the embodiment of the present invention；
Figure 10 is an entity apparatus structural representation of the apparatus for evaluating of potential target object in the embodiment of the present invention.
Embodiment
The invention provides a kind of appraisal procedure and apparatus for evaluating of potential target object, for solve in the prior art without
Method is accurate and timely evaluates the problem of valuable latent objective by social media, according to user social media interactive row
For group information and interactive grade RFM assessment models, user's classification is carried out to user so that enterprise can be with by social media channel
Valuable latent visitor is obtained accurately and timely, so as to improve the sales service of enterprise level.
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, the every other implementation that those skilled in the art are obtained under the premise of creative work is not made
Example, belongs to the scope of protection of the invention.
Today of high speed development in internet, many people have got used to carrying out exchange of information by internet and divided
To enjoy, media provide Domestic News also by internet for the numerous common people, and enterprise provides commerce services by internet for the common people, by
Facility in internet, in time and the features such as wide coverage so that it is more and more to be active in the netizen of internet, and enterprise is
It is convenient by giving these netizens to provide, after the function that netizen is needed carries out centralization, it is proposed that the concept of social media, make
It is referred to as user with the netizen of the social media.Broadly, social media is often referred on internet the content life based on customer relationship
Production and switching plane.Social media mainly includes social network sites and social APP, typical case such as QQ, wechat, microblogging and ends of the earth forum
Deng.
Some enterprises may have a kind of social media, but the enterprise also has other kinds of service concurrently, is lifted with Tengxun
Example explanation, QQ is the software of a social media type of Tengxun, and present Tengxun is newly proposed a highly professional soft again
Part, for example, stock class.In view of the experience property of user, when new software is promoted, it is impossible to enter to all QQ users
Row is promoted, because the uninterested user of stock can dislike such a promotion message, but in the huge number of users of QQ user, meeting
It is interested in stock to have certain customers, and it is expected as new software users, and this certain customers is exactly latent for enterprise
Visitor, it is the potential target object for needing to promote.And how to be distinguished by social media channel from vast user group latent
In destination object, and this part potential target object is converted into the user of new software, it is horizontal to be that sales service improves in enterprise
It is crucial.Therefore, the classification to the user of the social media using enterprise is extremely necessary.
In existing user classification method, traditional RFM models are a kind of, but can be made based on consumer behavior, enterprise
Value assessment is carried out to user with the consumer record of RFM model long-term follow users, value, the identification for defining different user are latent
Customer loss risk, speculate the unusual fluctuation situation of customer consumption or assess existing customer loss risk, so as to formulate product
Promote or keep the measures such as client.But consumer behavior is not necessarily present using the user of social media, it is based on disappearing so using
Traditional RFM models of expense behavior are not extensive enough and accurate to user's classification.
SCRM can obtain some open attributes of user, such as sex, age, educational background, place rapidly based on social media
Ground and social networks type etc., so as to which user quickly is carried out into simple classification, help enterprise's locking potential target object.Enterprise
Sale can be set by the user of point good class or contact staff receives the priority of user, preferential reception more meets potential mesh
Mark the user of characteristics of objects.For example, education and training organization can be by the user in the area of suitable age bracket and service covering
As reception ordering rule.But this classification that simply the open attribute simply based on user is carried out, it is to be made with static data
For basis of classification, data when social media is actually used, do not use in classification, can not dynamically adjust use
Family is classified, for example, some user, when just social media is begun to use, what hobby was filled in is film, but is made
When with social media, it is not related to film, but the function of the music type used, so can not dynamically it adjust
User classifies, and it is wrong necessarily to cause the potential target object that enterprise obtains.
Mentioned above in order to solve the problems, such as, the present invention has carried out some changes to RFM models so that is obtained after improvement
Interactive grade RFM assessment models, social media can be based on user's classification is carried out to user.Technical scheme is applied to
Communication system, as shown in figure 1, communication system includes terminal and server, terminal is connected with server communication, and terminal can have more
Individual, each terminal sends the data of itself to server, after server receives the data that each terminal is sent, to all
Data carry out analysis classification, and the data analyzed after sorting out are preserved, can be from server when being needed so as to subsequent terminal
Directly obtain.
In actual applications, the software of social media is installed, general terminal is mobile phone and computer etc. with upper in terminal
The electronic equipment of net function, user records the operation of user when using social media, for example, sending out wechat public number
The article of cloth such as is forwarded or thumbed up at the behavior, referred to as mutual-action behavior of the user in social media.Terminal will can be collected into
The data of the operation of user are sent to server, and server obtains the data that multiple terminals are sent, so as to preserve all users
Data in social media.
Referring to Fig. 2, the embodiment of the present invention provides a kind of appraisal procedure of potential target object, including：
201st, mutual-action behavior group information of the user in social media is obtained；
In the present embodiment, user can perform a series of mutual-action behavior when using social media in terminal, and user
Mutual-action behavior is the action trail for having sequencing, if Fig. 3 is user's possible action trail in wechat public number, is such as schemed
4 be user's possible action trail on QQ group, if Fig. 5 is user's possible action trail on QQ, in Fig. 3, Fig. 4 and Fig. 5
Rounded Box represents the mutual-action behavior of user, and square box represents the page or carrier of the mutual-action behavior of carrying user, for example, wechat is public
On this page of crowd's picture and text message, the mutual-action behavior that user may be carried out includes：Share, collect, thumbing up, commenting on, reading original
Text and author etc..Mutual-action behavior group is one group of orderly mutual-action behavior.So that user in Fig. 3 is in wechat public number as an example, use
Family first pass through " check author, search addition do not pay close attention to, business card is shared, public number contact point and sweep Quick Response Code " in one
Mutual-action behavior enters the wechat public number data card page, includes " checking history in the mutual-action behavior that the data card page can be carried out
Message, concern, cancel concern, message, unlatching/closing geographical position be set, shares business card and calls ", if interactive row
To be " checking history message ", then picture and text message page is entered；If mutual-action behavior is not carried out, or the mutual-action behavior performed
It is not " checking history message ", then " sweeps Quick Response Code, search enters and had focused on, into public number, message by performing mutual-action behavior
One in list and concern public number list " enters the public number chat window page, can be with the public number chat window page
The mutual-action behavior of progress includes " forwarding/share picture and text, menu and keyword to reply ", if mutual-action behavior is " menu or key
Word is replied ", then enter reception components page；If mutual-action behavior is not carried out, or what is performed is not that " menu or keyword return
It is multiple ", then by performing one in mutual-action behavior " checking/forward the picture and text shared, the picture checked picture and text message, check collection "
It is individual to enter picture and text message page, include in the mutual-action behavior that picture and text message page can perform " share, collect, comment on, thumb up,
Original read and author ", user can enter reception components page by performing mutual-action behavior " clicking on reception component ",
The mutual-action behavior that reception components page can perform includes " on-line consulting, calling, adding group, plusing good friend and the concern public
Number ".
If user enters the data card page by the execution " business card is shared " of sharing of good friend, " checked in the data card page
History message " enters picture and text message, " original read " backed off after random wechat, and user thinks that this wechat public number is worth closing afterwards
Note, enters the data card page further through " business card is shared ", and " concern " is performed in the data card page, performs " concern public number list "
Into public number chat window, " menu " is clicked in public number chat window, then click on " checking picture and text message " and disappear into picture and text
The page is ceased, is clicked on " original read ".From the mutual-action behavior of user, two mutual-action behavior groups, a mutual-action behavior can be obtained
The mutual-action behavior that group includes is followed successively by " business card is shared ", " checking history message " and " original read ", another mutual-action behavior group
Be the mutual-action behavior included be followed successively by " business card is shared ", " concern ", " concern public number list ", " menu ", " check that picture and text disappear
Breath " and " original read ".
Mutual-action behavior group information is the information of all mutual-action behaviors in mutual-action behavior group, i.e. user and social media is interactive
Information.Mutual-action behavior group information is typically recorded in terminal, or the webserver of social media, is needing to be used
When family is classified, the mutual-action behavior group information of the user can be extracted from terminal or server.Certainly, in actual reality
Shi Shi, it is contemplated that the accuracy of user's classification, it may be necessary to use multigroup mutual-action behavior group, then mutual-action behavior group information is actual
On be multiple mutual-action behavior groups information an intersection, such as mutual-action behavior group information includes above-mentioned two mutual-action behavior group
Information, i.e. mutual-action behavior " business card is shared ", " checking history message ", " original read ", " business card is shared ", " concern ", " concern public affairs
Crowd list ", " menu ", the information of " checking picture and text message " and " original read ".
202nd, according to interactive grade RFM assessment models and mutual-action behavior group information, the RFM codes of user are obtained；
In the present embodiment, interactive grade RFM assessment models are built in advance on the basis of RFM models, interaction etc.
The sequence rule of level RFM assessment models does not change, has simply redefined RFM parameters, and RFM parameters include parameter R, parameter
F and parameter M, parameter R represent user and social media the last interactive time, and parameter F represents that user exists with social media
Interactive number in preset time, parameter M represents user and interactive depth of the social media in preset time, according to interactive row
Parameter R, parameter F and parameter M information can be calculated for group information, then using interactive grade RFM assessment models and and
The information of RFM parameters, the RFM codes with regard to user can be obtained, the user of RFM representations and the interactive degree of social media.
203rd, user's classification is carried out to user according to RFM codes.
In the present embodiment, the interactive degree of user and social media is known according to RFM codes, general RFM codes are with numerical value
Form represents, then, user's classification is carried out to user according to RFM codes, divided according to the numerical value of RFM codes, specifically
The criteria for classifying be also what enterprise was set.
In the embodiment of the present invention, unlike prior art 1, user's classification is the interaction according to user in social media
Obtained from behavior group information and interactive grade RFM assessment models, enterprise is known that each user's according to user's classification
Potential value；Unlike prior art 2, because user's classification needs to use mutual-action behavior group letter of the user in social media
Breath, and mutual-action behavior group information is to be obtained according to user using the situation of social media, in practice, user uses social matchmaker
The situation of body is dynamic, and thus the user that obtains of dynamic classifies, and therefore, enterprise can be accurate by social media channel
And valuable latent visitor is timely selected, so as to improve the sales service of enterprise level.
Grade RFM assessment models interactive in the embodiment shown in Figure 2 are built in advance, below to interactive grade
RFM assessment models are being described in detail of how building, specific as follows：
Optionally, in some embodiments of the present invention, user is obtained before the mutual-action behavior group information of social media, also
Including：
Redefine the RFM parameters of RFM models, RFM parameters include parameter R, parameter F and parameter M, parameter R represent user with
Social media the last interactive time, parameter F represent user and interactive number of the social media in preset time, parameter
M represents user and interactive depth of the social media in preset time；
The interactive depth weight of mutual-action behavior corresponding to mutual-action behavior information is set, obtains weight table；
According to RFM parameters and weight table, interactive grade RFM assessment models are built.
In the present embodiment, existing RFM models are built in the buying behavior of user, and in social media, user
Seldom directly generation buying behavior is more a kind of and the social media of enterprise mutual-action behavior, as described above micro-
Believe the action trail on public number and QQ, similar to concern, plusing good friend and add group etc. to be typical mutual-action behavior.Therefore, existing
RFM models on the basis of, by behavioral characteristic of the user in social media, redefined the RFM parameters of RFM models, such as
Shown in table 1.
Table 1
In table 1, RFM parameters include parameter R, parameter F and parameter M, and parameter R represents that user is the last with social media
The time of interaction, parameter F represent user and interactive number of the social media in preset time, and parameter M represents user and social activity
When what is represented in interactive depth of the media in preset time, wherein preset time is carry out user's classification that enterprise is set, need
Mutual-action behavior group information of the user to be obtained within a period of time.For example, enterprise needs to release a music class production recent
Product, then with regard to needing to know in nearest one month or 1 year, which user frequently carries out music class in social media
Mutual-action behavior.
Because parameter R represents user and social media the last interactive time, then parameter R information is exactly interactive
The temporal information of the last mutual-action behavior in behavior group, can directly it know from mutual-action behavior group information, and parameter F is represented
Be user and interactive number of the social media in preset time, then parameter R information is exactly all in preset time
The number of mutual-action behavior in mutual-action behavior group, can be directly obtained from mutual-action behavior group information, but parameter M information be can not be direct
Obtain, parameter M represents user and interactive depth of the social media in preset time, and parameter M information is not one aobvious
Property value.By the action trail for observing the user in Fig. 2 embodiments, it is found that some mutual-action behaviors represent is user couple
The positive positive feedback of enterprise, such as pay attention in, plusing good friend and add group etc.；What some mutual-action behaviors represented is user to enterprise
Negative negative feedback, such as cancel concern, delete good friend and move back group etc.；Also some mutual-action behaviors are with respect to other mutual-action behaviors
The stronger positive positive feedback of client is shown, such as shares picture and text message, contrast and check picture and text message, open prompting message
And contrast reception message but do not prompt etc..With a value come weigh representated by these mutual-action behaviors front actively/negatively disappear
The degree of pole feedback, i.e., interactive depth.However, weighing various mutual-action behaviors with how much values, interactive grade RFM is very dependent on
Understanding of the assessment models structure person to the mutual-action behavior of user.Below by taking mutual-action behavior of the user using wechat public number as an example,
It is that each behavior assigns weight by tripartite division, weight therein can be self-defined according to the understanding of the mutual-action behavior to user.Weight
Table is as shown in table 2.
Table 2
Because popularization and popularity of the attention rate for enterprise of the wechat public number of enterprise are very important, then use
The concern at family is highly important, and the interactive depth weight assigned from the different mutual-action behaviors in table 2 can be seen that wechat public affairs
The mutual-action behavior " concern " of the carrying page " data card " of many numbers, it is 3 to impart interactive depth weight, and mutual-action behavior " taking pass " takes
Pass is negative to wechat public number, and interactive depth weight is -3.
Interactive grade RFM assessment models obtain the sequence rule that the modes of RFM codes still uses existing RFM models, i.e.,
R1-R5, F1-F5 and M1-M5 are respectively divided into parameter R, parameter F and parameter M, thus obtain the RFM codes of client.
In the embodiment of the present invention, it is described in detail to how to build interactive grade RFM assessment models, will be existing
RFM parameters in RFM models are newly defined as related to mutual-action behavior so that carrying out classification to user by social media turns into
May.
Optionally, in some embodiments of the present invention, mutual-action behavior group information of the user in social media is obtained, including：
Mutual-action behavior information of the user in the mutual-action behavior of social media is gathered, mutual-action behavior is user in social media
Behavior, mutual-action behavior information are to include behavior mark, user's mark, social media mark, time of the act, behavior type and upper one
Behavior identifies；
Mutual-action behavior group is obtained according to mutual-action behavior information, mutual-action behavior group is the effective interaction row of one group of ordered arrangement
For；
The mutual-action behavior information of each effective interaction behavior in mutual-action behavior group is obtained, obtains mutual-action behavior group information.
In the present embodiment, due to social media it is targeted be large number of user, then the mutual-action behavior collected
Information, it is as caused by various mutual-action behaviors of many users in social media, is the behavior of a user as shown in table 3
Track, the data card page is entered by the execution " business card is shared " of sharing of good friend, entered in the data card page " checking history message "
Enter picture and text message, " original read " backed off after random wechat, afterwards user think that this wechat public number merits attention, further through " name
Piece is shared " enter the data card page, " concern " is performed in the data card page, performs and " pays close attention to public number list " to chat into public number
Skylight opening, " menu " is clicked in public number chat window, then click on " checking picture and text message " and enter picture and text message page, clicked on
" original read ".Behavior mark, user's mark, social media mark are contained in the mutual-action behavior information of each mutual-action behavior
Knowledge, time of the act, behavior type, lastrow are mark, user society matchmaker's account number etc..Behavior mark in mutual-action behavior information
Identified with a upper behavior, will can connect out two groups of mutual-action behavior groups, and first group of mutual-action behavior group behavior mark represents
Be " 0a0a1,0a0b1 and 0a0f1 ", second group the behavior of mutual-action behavior group mark represent be " 0a0a1,0a0b2,0a0c1,
0a0d1,0a0e1 and 0a0f1 ", but be not that all mutual-action behaviors in action trail are all effective, such as, enterprise thinks
Obtain nearest one month user classification, then in action trail time of the act be one month before be exactly invalid interaction
Behavior.It is as described above that the action trail for combing user in timing statisticses section is identified by user, effectively it is connected on an action trail
In each effective interaction behavior, form two groups of mutual-action behavior groups, distance statistics time point nearest mutual-action behavior group it is last
The time of one mutual-action behavior as nearest interaction time, as shown in table 4.It is effectively mutual to obtain each in two groups of mutual-action behavior groups
The mutual-action behavior information of dynamic behavior, obtains mutual-action behavior group information.It should be noted that simply lifted above with a user
Example explanation, several users in actual applications be present, then just need to combine user's mark to obtain each user's
Mutual-action behavior group, and the quantity of mutual-action behavior group does not limit, and can be not less than zero integer.
Table 3
Table 4
| User identifies | Mutual-action behavior group number | Nearest interaction time |
| 2000001 | 2 | 2016/5/23 16:26:00 |
In the embodiment of the present invention, the acquisition to mutual-action behavior group information is described in detail, and the multi-user's that comforms is mutual
In dynamic behavior, the effective interaction behavior of user is selected, and mutual-action behavior group of connecting out, so as to obtain mutual-action behavior group letter
Breath, can significantly reduce workload, improve the precision of information.
Optionally, in some embodiments of the present invention, according to interactive grade RFM assessment models and mutual-action behavior group information,
The RFM codes of user are obtained, including：
Parameter R information and parameter F information are obtained according to mutual-action behavior group information；
According to mutual-action behavior group information and weight table, parameter M information is obtained；
According to the sequence rule of interactive grade RFM assessment models, the information of information, parameter F to parameter R and parameter M's
Information is handled, and obtains the RFM codes of user.
In the present embodiment, by taking two mutual-action behavior groups of the user shown in Fig. 6 in wechat public number as an example, two groups interactive
Behavior group information is as shown in table 3, wherein a total of 9 mutual-action behaviors, wherein recent mutual-action behavior " original read "
Time of the act is for " 201,6/5,/23 16:26:00 ", then parameter R information is exactly " 201,6/5,/23 16:26:00 ", parameter
F information is exactly " 9 times ", and parameter F information also needs to the behavior type according to mutual-action behavior, with reference to the weight of above-mentioned table 1
Table is calculated, and calculates the parameter M of two groups of mutual-action behavior groups information, the parameter M of first group of mutual-action behavior group information respectively
For interactive (checking history message)+2 (original read)=3 of depth value M=0 (business card is shared)+1；Second group of mutual-action behavior group
Parameter M information is that interactive depth value M=0 (business card is shared)+3 (concern)+0 (entering public number)+1 (menu)+1 (checks figure
Literary message)+2 (original read)=7；Total parameter M of user information is interactive first group of depth value M=M values+the second
M value=10 of group, according to the sequence rule of interactive grade RFM assessment models, R1- is respectively divided into parameter R, parameter F and parameter M
R5, F1-F5 and M1-M5, for example, R1 expressions is to have mutual-action behavior to user the year before before 2 years, R2 is represented the year before to 10
User has a mutual-action behavior before individual month, and R3 to user before 8 months has mutual-action behavior before representing 10 months, and R4 is represented 8 months to 5 months
Preceding user has a mutual-action behavior, R5 represent 5 months before till now time point user have mutual-action behavior, parameter R information is " 2016/
5/23 16:26:00 ", current point in time is September in 2016 3, then parameter R is R5；F1 represents that mutual-action behavior is no more than 2
Secondary, F2 represents mutual-action behavior at 3-6 times, and F3 represents mutual-action behavior at 7-10 times, and F4 represents mutual-action behavior at 11-12 times, F5 tables
Show mutual-action behavior more than 13 times, parameter F information is exactly 9 times, then parameter F is F3；Parameter M information is interactive depth
Value, interactive depth value are M1 less than 1, and the interactive depth value of M2 expressions is in 1-5, and M3 represents interactive depth value in 6-8, and M4 represents interactive
For depth value in 9-11, M5 represents interactive depth value more than 12, and parameter M information is that interactive depth value is 10, then parameter M is
M4.It follows that the RFM codes of the user are R5F3M4.And user and social media can intuitively be seen according to user RFM codes
Interactive degree, what RFM codes were that R1F1M1 represents is that long-term not interactive, the interactive number of user is low and interactive depth is low, and RFM codes are
What R5F5M5 was represented is that user has interactive, interactive number high in the recent period and interactive depth is big, it is seen that is to be in from R1F1M1 to R5F5M5
The incremental situation of the interactive degree at current family, then according to the difference of the RFM codes of user, can just carry out user's classification to user.
It should be noted that R1-R5, F1-F5 and M1-M5 are a kind of dividing mode for presenting and being incremented by relation, specific
In implementation, can also there are more divisions, such as R1-R10, F1-F10 and M1-M10 etc..
Optionally, in some embodiments of the present invention, according to RFM parameters and weight table, build interactive grade RFM and assess mould
After type, in addition to：
Test whether interactive grade RFM assessment models meet default evaluation criteria；
If it is not, then reconstruct interactive grade RFM assessment models.
In the embodiment of the present invention, after interactive grade RFM assessment models are built, it is also necessary to tested, because ginseng
Number R and parameter F can be directly obtained from mutual-action behavior group information, but parameter M is not really simply to add and subtract, it is also necessary to
An important element be exactly mutual-action behavior interactive depth weight, the interactive depth weight of each mutual-action behavior is desirable
Set, under enterprise's different demands, the interactive depth weight of mutual-action behavior all needs to change, therefore tests interactive grade
Whether RFM assessment models meet that default evaluation criteria is highly important, are somebody's turn to do if meeting that default evaluation criteria can uses
Interactive grade RFM assessment models, if being unsatisfactory for default evaluation criteria, need to rebuild interactive grade RFM assessments
Model.
Optionally, in some embodiments of the present invention, test whether interactive grade RFM assessment models meet that default assess is marked
Standard is as follows：
Obtain test mutual-action behavior group information of the test user in social media；
According to interactive grade RFM assessment models and test mutual-action behavior information, the test RFM codes of test user are obtained；
Judge to test whether RFM codes are within the scope of predetermined deviation value with RFM codes；
If so, then interactive grade RFM assessment models meet default evaluation criteria；
If it is not, then interactive grade RFM assessment models are unsatisfactory for default evaluation criteria.
In the embodiment of the present invention, by taking wechat public number as an example, in order to improve the accuracy of test, it is necessary to improve test volume,
The number for so testing user should be larger, is once mass-sended, feedback of the tracking and testing user to mass-sending content：Feedback-less,
Mass-sending content is checked, is checked and shares mass-sending content, is checked and thumbs up mass-sending content etc., obtains test RFM codes, and contrast step
The RFM codes obtained in 202, because user is bigger to the potential interest of enterprise to the more positive explanation user of mass-sending feedback, if step
The larger user of RFM codes sorts also more before examination in current mass-sending feedback in rapid 202, then illustrates that RFM code divisions class can be in certain journey
Potential interest of the user to enterprise is reacted on degree.Whether the deviation for judging to test RFM codes and RFM codes is in predetermined deviation value model
In enclosing, predetermined deviation value scope is error allowed band, if so, then it is believed that interactive grade RFM assessment models meet default comment
Standard is estimated, if it is not, then interactive grade RFM assessment models are unsatisfactory for default evaluation criteria, it is necessary to be reconstructed.
It should be noted that go out by testing RFM codes with RFM codes deviation to determine interactive grade RFM assessment models
Whether satisfaction is preset outside evaluation criteria, can also be by contrasting the conclusion of the business results of SCRM systems.The conclusion of the business result of SCRM systems
It is most directly to react contribution degree of the user to enterprise, therefore by according to the good user of RFM code divisions class, by the conclusion of the business in SCRM systems
Record is analyzed, if in user's classification the larger user of RFM codes strike a bargain the cycle is relatively short, turnover relatively
Greatly, conclusion of the business number is relatively more, then illustrates that interactive grade RFM assessment models are effective, on the contrary then invalid, it is necessary to reset weight
Table reconstructs interactive grade RFM assessment models.
It is used to find potential target object it should be noted that the present invention is not only enterprise, can be also used for social flat
The bean vermicelli value assessment of platform, such as classify to the bean vermicelli of a certain user on microblogging, by the behavior such as forwarding, commenting on and thumb up
Bean vermicelli is divided into and enlivens the type such as bean vermicelli and " corpse " powder；Good friend's value assessment is can be also used for, such as in social platform
When good friend is large number of, according to the interactive situation of user and good friend, good friend is divided into each grade, in performance
In, provide and do not limit specifically for which kind of ground.
For ease of preferably implementing the above-mentioned correlation technique of the embodiment of the present invention, it is also provided below for coordinating the above method
Relevant apparatus.
Referring to Fig. 7, the embodiment of the present invention provides a kind of apparatus for evaluating 700 of potential target object, including：
Acquisition module 701, processing module 702 and sort module 703；
Acquisition module 701, for obtaining mutual-action behavior group information of the user in social media, mutual-action behavior group information is use
The family information interactive with social media；
Processing module 702, for the mutual-action behavior group according to interactive grade RFM assessment models and the acquisition of acquisition module 701
Information, obtain the interactive degree of the RFM codes of user, RFM representations user and social media；
Sort module 703, the RFM codes for being obtained according to processing module 702 carry out user's classification to user.
On the basis of apparatus for evaluating shown in Fig. 7, referring to Fig. 8, the apparatus for evaluating 700 of potential target object also includes：
Module 801 is built, for redefining the RFM parameters of RFM models, RFM parameters include parameter R, parameter F and parameter M,
Parameter R represents user and social media the last interactive time, and parameter F represents user with social media in preset time
Interactive number, parameter M represents user and interactive depth of the social media in preset time；
Module 801 is built, is additionally operable to set the interactive depth weight of mutual-action behavior corresponding to mutual-action behavior information, is weighed
Weight table；
Module 801 is built, is additionally operable to, according to RFM parameters and weight table, build interactive grade RFM assessment models.
Optionally, in some embodiments of the present invention,
Acquisition module 701, specifically for collection user in the mutual-action behavior information of the mutual-action behavior of social media, interactive row
For for user social media behavior, mutual-action behavior information be include behavior mark, user mark, social media mark, OK
It is mark for time, behavior type and lastrow；
Acquisition module 701, it is additionally operable to obtain mutual-action behavior group according to mutual-action behavior information, mutual-action behavior group is one group orderly
The effective interaction behavior of arrangement；
Acquisition module 701, it is additionally operable to obtain the mutual-action behavior information of each effective interaction behavior in mutual-action behavior group, obtains
To mutual-action behavior group information.
Optionally, in some embodiments of the present invention,
Processing module 702, specifically for obtaining parameter R information and parameter F information according to mutual-action behavior group information；
Processing module 702, it is additionally operable to, according to mutual-action behavior group information and weight table, obtain parameter M information；
Processing module 702, the sequence rule according to interactive grade RFM assessment models is additionally operable to, information, ginseng to parameter R
Number F information and parameter M information are handled, and obtain the RFM codes of user.
On the basis of apparatus for evaluating shown in Fig. 8, referring to Fig. 9, the apparatus for evaluating 700 of potential target object also includes：
Test module 901, for testing whether interactive grade RFM assessment models meet default evaluation criteria；
Module 901 is built, is additionally operable to be unsatisfactory for default comment when test module 901 tests out interactive grade RFM assessment models
When estimating standard, interactive grade RFM assessment models are reconstructed.
Optionally, in some embodiments of the present invention,
Test module 901, specifically for obtaining test mutual-action behavior group information of the test user in social media；
Test module 901, it is additionally operable to, according to interactive grade RFM assessment models and test mutual-action behavior group information, be surveyed
The test RFM codes at family on probation；
Test module 901, it is additionally operable within the scope of judging whether test RFM codes is in predetermined deviation value with RFM codes, if
It is that then interactive grade RFM assessment models meet default evaluation criteria, if it is not, then interactive grade RFM assessment models are unsatisfactory for presetting
Evaluation criteria.
To sum up, unlike prior art 1, user's classification is the mutual-action behavior group information according to user in social media
Obtained from interactive grade RFM assessment models, enterprise is according to user's classification it is known that the potential value of each user；With
Unlike prior art 2, because user's classification needs to use mutual-action behavior group information of the user in social media, and interactive row
It is to be obtained according to user using the situation of social media for group information, in practice, user is using the situation of social media
Dynamically, thus the user that obtains of dynamic classifies, and therefore, enterprise can accurately and timely be selected by social media channel
Valuable latent visitor is selected out, so as to improve the sales service of enterprise level.
Be described above the embodiment of the modular construction of the apparatus for evaluating of potential target object, below using apparatus for evaluating as
Exemplified by server, the entity apparatus of apparatus for evaluating is illustrated.
The apparatus for evaluating of potential target object is by taking server as an example, and as shown in Figure 10, the apparatus for evaluating can be because of configuration or property
Energy is different and produces bigger difference, can include one or more central processing units (central processing
Units, CPU) 1022 (for example, one or more processors) and memory 1032, one or more storage applications
The storage medium 1030 of program 1042 or data 1044 (such as one or more mass memory units).Wherein, memory
1032 and storage medium 1030 can be it is of short duration storage or persistently storage.One can be included by being stored in the program of storage medium 1030
Individual or more than one module (diagram does not mark), each module can include operating the series of instructions in server.More enter
One step, CPU1022 could be arranged to communicate with storage medium 1030, perform the system in storage medium 1030 on the server
Row command operating.
Apparatus for evaluating can also include one or more power supplys 10210, one or more radio network interfaces
1050, one or more input/output interfaces 1058, and/or, one or more operating systems 1041, such as
Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM etc..
Referring to Fig. 10, the embodiment of the present invention provides a kind of apparatus for evaluating of potential target object, including：
Radio network interface 1050, CPU1022 and memory 1032, radio network interface 1050, CPU1022 and memory
Interconnected between 1003 by bus, computer instruction is stored with memory 1032, CPU1022 is referred to by performing computer
Order, so as to realize following methods：
Mutual-action behavior group information of the user in social media is obtained, mutual-action behavior group information is that user and social media are interactive
Information；
According to interactive grade RFM assessment models and mutual-action behavior group information, the RFM codes of user, RFM representation users are obtained
With the interactive degree of social media；
User's classification is carried out to user according to RFM codes.
Claims (12)
- A kind of 1. appraisal procedure of potential target object, it is characterised in that including：Mutual-action behavior group information of the user in social media is obtained, the mutual-action behavior group information is the user and the social activity The information of Media；According to interactive grade RFM assessment models and the mutual-action behavior group information, the RFM codes of the user, the RFM codes are obtained Represent the interactive degree of the user and the social media；User's classification is carried out to the user according to the RFM codes.
- 2. the appraisal procedure of potential target object according to claim 1, it is characterised in that the acquisition user is in social activity Before the mutual-action behavior group information of media, in addition to：The RFM parameters of RFM models are redefined, the RFM parameters include parameter R, parameter F and parameter M, the parameter R represent use Family represents the user with the social media in preset time with social media the last interactive time, the parameter F Interior interactive number, the parameter M represent the user and interactive depth of the social media in preset time；The interactive depth weight of mutual-action behavior corresponding to the mutual-action behavior information is set, obtains weight table；According to the RFM parameters and the weight table, interactive grade RFM assessment models are built.
- 3. the appraisal procedure of potential target object according to claim 2, it is characterised in that the acquisition user is in social activity The mutual-action behavior group information of media, including：Mutual-action behavior information of the user in the mutual-action behavior of social media is gathered, the mutual-action behavior is the user in institute The behavior of social media is stated, the mutual-action behavior information is when including behavior mark, user's mark, social media mark, behavior Between, behavior type and lastrow be mark；Mutual-action behavior group is obtained according to the mutual-action behavior information, the mutual-action behavior group is the effective interaction of one group of ordered arrangement Behavior；The mutual-action behavior information of each effective interaction behavior in the mutual-action behavior group is obtained, obtains mutual-action behavior group information.
- 4. the appraisal procedure of potential target object according to claim 3, it is characterised in that described according to interactive grade RFM assessment models and the mutual-action behavior group information, the RFM codes of the user are obtained, including：The information of the parameter R and the information of the parameter F are obtained according to the mutual-action behavior group information；According to the mutual-action behavior group information and the weight table, the information of the parameter M is obtained；According to the sequence rule of the interactive grade RFM assessment models, the information of information, the parameter F to the parameter R and The information of the parameter M is handled, and obtains the RFM codes of the user.
- 5. the appraisal procedure of the potential target object according to any one of claim 2 to 4, it is characterised in that described According to the RFM parameters and the weight table, after building interactive grade RFM assessment models, in addition to：Test whether the interactive grade RFM assessment models meet default evaluation criteria；If it is not, then reconstruct the interactive grade RFM assessment models.
- 6. the latent objective appraisal procedure according to claim 5 based on social media, it is characterised in that the test is described mutually Whether dynamic grade RFM assessment models meet default evaluation criteria, including：Obtain test mutual-action behavior group information of the test user in social media；According to the interactive grade RFM assessment models and the test mutual-action behavior information, the test of the test user is obtained RFM codes；Judge whether the test RFM codes are within the scope of predetermined deviation value with the RFM codes；If so, then the interactive grade RFM assessment models meet default evaluation criteria；If it is not, then the interactive grade RFM assessment models are unsatisfactory for default evaluation criteria.
- A kind of 7. apparatus for evaluating of potential target object, it is characterised in that including：Acquisition module, for obtaining mutual-action behavior group information of the user in social media, the mutual-action behavior group information is described User's information interactive with the social media；Processing module, for according to interactive grade RFM assessment models and the mutual-action behavior group information, obtaining the user's RFM codes, user described in the RFM representations and the interactive degree of the social media；Sort module, for carrying out user's classification to the user according to the RFM codes.
- 8. the apparatus for evaluating of potential target object according to claim 7, it is characterised in that the apparatus for evaluating also wraps Include：Module is built, for redefining the RFM parameters of RFM models, the RFM parameters include parameter R, parameter F and parameter M, institute State parameter R and represent user and social media the last interactive time, the parameter F represents the user and the social matchmaker Interactive number of the body in preset time, the parameter M represent that the user and the social media are mutual in preset time Dynamic depth；The structure module, it is additionally operable to set the interactive depth weight of mutual-action behavior corresponding to the mutual-action behavior information, obtains Weight table；The structure module, it is additionally operable to, according to the RFM parameters and the weight table, build interactive grade RFM assessment models.
- 9. the apparatus for evaluating of potential target object according to claim 8, it is characterised in thatThe acquisition module, the mutual-action behavior information specifically for gathering mutual-action behavior of the user in social media are described Mutual-action behavior is behavior of the user in the social media, and the mutual-action behavior information is to include behavior mark, Yong Hubiao Knowledge, social media mark, time of the act, behavior type and lastrow are mark；The acquisition module, it is additionally operable to obtain mutual-action behavior group according to the mutual-action behavior information, the mutual-action behavior group is one The effective interaction behavior of group ordered arrangement；The acquisition module, it is additionally operable to obtain the mutual-action behavior information of each effective interaction behavior in the mutual-action behavior group, Obtain mutual-action behavior group information.
- 10. the apparatus for evaluating of potential target object according to claim 9, it is characterised in thatThe processing module, specifically for the information that the parameter R is obtained according to the mutual-action behavior group information and the parameter F Information；The processing module, it is additionally operable to, according to the mutual-action behavior group information and the weight table, obtain the letter of the parameter M Breath；The processing module, the sequence rule according to the interactive grade RFM assessment models is additionally operable to, to the letter of the parameter R The information of breath, the information of the parameter F and the parameter M is handled, and obtains the RFM codes of the user.
- 11. the apparatus for evaluating of the potential target object according to any one of claim 8 to 10, it is characterised in that described Apparatus for evaluating also includes：Test module, for testing whether the interactive grade RFM assessment models meet default evaluation criteria；The structure module, it is additionally operable to when the interactive grade RFM assessment models are unsatisfactory for the default evaluation criteria, reconstructs The interactive grade RFM assessment models.
- 12. the apparatus for evaluating of potential target object according to claim 11, it is characterised in thatThe test module, specifically for obtaining test mutual-action behavior group information of the test user in social media；The test module, it is additionally operable to, according to the interactive grade RFM assessment models and the test mutual-action behavior group information, obtain To the test RFM codes of the test user；The test module, be additionally operable to judge the test RFM codes and the RFM codes whether in predetermined deviation value scope it It is interior, if so, then the interactive grade RFM assessment models meet default evaluation criteria, if it is not, then the interactive grade RFM is assessed Model is unsatisfactory for default evaluation criteria.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610814375.0A CN107808346B (en) | 2016-09-08 | 2016-09-08 | Evaluation method and evaluation device for potential target object |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610814375.0A CN107808346B (en) | 2016-09-08 | 2016-09-08 | Evaluation method and evaluation device for potential target object |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107808346A true CN107808346A (en) | 2018-03-16 |
| CN107808346B CN107808346B (en) | 2021-08-24 |
Family
ID=61569636
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610814375.0A Active CN107808346B (en) | 2016-09-08 | 2016-09-08 | Evaluation method and evaluation device for potential target object |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107808346B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN109345075A (en) * | 2018-08-31 | 2019-02-15 | 深圳市轱辘汽车维修技术有限公司 | A kind of professional person, which authenticates, investigates method, apparatus and terminal device |
| CN110458298A (en) * | 2018-05-03 | 2019-11-15 | 上海博泰悦臻网络技术服务有限公司 | The analysis maintenance system and client's online management system and method for rolling stock and rolling stock accessories |
| CN111046270A (en) * | 2018-10-15 | 2020-04-21 | 深圳市加推科技有限公司 | Data processing method and device, and computer readable storage medium |
| CN111046269A (en) * | 2018-10-15 | 2020-04-21 | 深圳市加推科技有限公司 | Data acquisition method and device and computer-readable storage medium |
| CN111178976A (en) * | 2019-12-31 | 2020-05-19 | 北京纷扬科技有限责任公司 | BP neural network-based user behavior cultivation system and method |
| US12002114B2 (en) * | 2022-10-17 | 2024-06-04 | Douyin Vision Co., Ltd. | Method, apparatus, device and storage medium for content presentation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104951446A (en) * | 2014-03-25 | 2015-09-30 | 阿里巴巴集团控股有限公司 | Big data processing method and platform |
| CN105573995A (en) * | 2014-10-09 | 2016-05-11 | 中国银联股份有限公司 | Interest identification method, interest identification equipment and data analysis method |
| CN105590231A (en) * | 2014-10-24 | 2016-05-18 | 阿里巴巴集团控股有限公司 | User data processing method and apparatus |
| CN105608194A (en) * | 2015-12-24 | 2016-05-25 | 成都陌云科技有限公司 | Method for analyzing main characteristics in social media |
-
2016
- 2016-09-08 CN CN201610814375.0A patent/CN107808346B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104951446A (en) * | 2014-03-25 | 2015-09-30 | 阿里巴巴集团控股有限公司 | Big data processing method and platform |
| CN105573995A (en) * | 2014-10-09 | 2016-05-11 | 中国银联股份有限公司 | Interest identification method, interest identification equipment and data analysis method |
| CN105590231A (en) * | 2014-10-24 | 2016-05-18 | 阿里巴巴集团控股有限公司 | User data processing method and apparatus |
| CN105608194A (en) * | 2015-12-24 | 2016-05-25 | 成都陌云科技有限公司 | Method for analyzing main characteristics in social media |
Non-Patent Citations (1)
| Title |
|---|
| 赵萌: "基于评论行为的客户终身价值模型改进", 《中国优秀硕士学位论文全文数据库经济与管理科学辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN110458298A (en) * | 2018-05-03 | 2019-11-15 | 上海博泰悦臻网络技术服务有限公司 | The analysis maintenance system and client's online management system and method for rolling stock and rolling stock accessories |
| CN109345075A (en) * | 2018-08-31 | 2019-02-15 | 深圳市轱辘汽车维修技术有限公司 | A kind of professional person, which authenticates, investigates method, apparatus and terminal device |
| CN111046270A (en) * | 2018-10-15 | 2020-04-21 | 深圳市加推科技有限公司 | Data processing method and device, and computer readable storage medium |
| CN111046269A (en) * | 2018-10-15 | 2020-04-21 | 深圳市加推科技有限公司 | Data acquisition method and device and computer-readable storage medium |
| CN111178976A (en) * | 2019-12-31 | 2020-05-19 | 北京纷扬科技有限责任公司 | BP neural network-based user behavior cultivation system and method |
| US12002114B2 (en) * | 2022-10-17 | 2024-06-04 | Douyin Vision Co., Ltd. | Method, apparatus, device and storage medium for content presentation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107808346B (en) | 2021-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107808346A (en) | A kind of appraisal procedure and apparatus for evaluating of potential target object | |
| Dierkes et al. | Estimating the effect of word of mouth on churn and cross-buying in the mobile phone market with Markov logic networks | |
| Çavdar et al. | Airline customer lifetime value estimation using data analytics supported by social network information | |
| CN107066476A (en) | A kind of real-time recommendation method based on article similarity | |
| Goswami et al. | Sustainability assessment of smallholder farms in developing countries | |
| CN106251174A (en) | Information recommendation method and device | |
| CN104657425A (en) | Topic management type network public opinion evaluation management system and method | |
| CN106203773A (en) | A kind of method and device of automatic management employee | |
| CN102254028A (en) | Personalized commodity recommending method and system which integrate attributes and structural similarity | |
| CN113157752B (en) | Scientific and technological resource recommendation method and system based on user portrait and situation | |
| Lee et al. | A decision-making framework for evaluating enterprise resource planning systems in a high-tech industry | |
| CN112907305B (en) | Customer full-period management system based on big data analysis | |
| Hajizadeh et al. | Scenario-based capability evaluation of ecotourism development–an integrated approach based on WLC, and FUZZY–OWA methods | |
| CN108932669A (en) | A kind of abnormal account detection method based on supervised analytic hierarchy process (AHP) | |
| CN111723292A (en) | Recommendation method and system based on graph neural network, electronic device and storage medium | |
| CN111191099B (en) | User activity type identification method based on social media | |
| Moreri et al. | Volunteered geographic information quality assessment using trust and reputation modelling in land administration systems in developing countries | |
| CN108429776A (en) | Method for pushing, device, client, interactive device and the system of network object | |
| CN111429161A (en) | Feature extraction method, feature extraction device, storage medium, and electronic apparatus | |
| CN106294410A (en) | A kind of determination method of personalized information push time and determine system | |
| Zhou et al. | System and software processes in practice: Insights from chinese industry | |
| Cai et al. | An extension of social network group decision-making based on trustrank and personas | |
| CN104281599A (en) | Method and device for recommending information to user in social network | |
| Burak | E-Logistics: a bibliometric and visualization analysis | |
| CN108090794A (en) | A kind of merchandise items sort method and device based on dynamic sliding time window |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |