CN110533500A - User identification method, device, computer equipment and storage medium - Google Patents
User identification method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN110533500A CN110533500A CN201910696730.2A CN201910696730A CN110533500A CN 110533500 A CN110533500 A CN 110533500A CN 201910696730 A CN201910696730 A CN 201910696730A CN 110533500 A CN110533500 A CN 110533500A
- Authority
- CN
- China
- Prior art keywords
- user
- multiple users
- wool
- party
- wool party
- 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 45
- 210000002268 wool Anatomy 0.000 claims abstract description 169
- 230000000694 effects Effects 0.000 claims abstract description 54
- 238000004891 communication Methods 0.000 claims abstract description 31
- 230000006399 behavior Effects 0.000 claims description 81
- 238000010586 diagram Methods 0.000 claims description 31
- 238000004590 computer program Methods 0.000 claims description 17
- 230000003542 behavioural effect Effects 0.000 claims description 11
- 241001494479 Pecora Species 0.000 claims description 4
- 210000004209 hair Anatomy 0.000 claims description 4
- 230000000875 corresponding effect Effects 0.000 description 24
- 238000012544 monitoring process Methods 0.000 description 9
- 239000012141 concentrate Substances 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0609—Buyer or seller confidence or verification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Remote Sensing (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Probability & Statistics with Applications (AREA)
- Development Economics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This application involves intelligent decisions, and the identification of wool party is realized using clustering algorithm, specifically disclose a kind of recognition methods of wool party, device, equipment and storage medium.Wherein method includes: the marketing activity of monitors distribution, obtains the communication discriminating code for applying for multiple users of the marketing activity；Each user is positioned to obtain customer position information using the communication discriminating code based on mobile location-based service, user location relational graph is generated according to the customer position information；Judged in multiple users with the presence or absence of certain customers according to the user location relational graph for potential wool party；If in multiple users there are certain customers be potential wool party, obtain the historical behavior information of multiple users；Multiple users are clustered according to the historical behavior information to obtain user's monoid, and wool party user is determined from multiple users according to user's monoid.The method increase the recognition efficiency of wool party and accuracys rate.
Description
Technical field
This application involves Internet technical fields more particularly to a kind of wool party user identification method, device, computer to set
Standby and storage medium.
Background technique
With the rapid development of mobile Internet, enterprise marketing shifts under line on line, issues by mobile Internet
Marketing activity is to realize Brand Marketing.For example, common marketing activity include give red packet, give discount coupon, Presenting gifts or
Person head is mono- to exempt from list etc., is attracted clients by marketing activity to obtain better marketing effectiveness.Since the marketing award of businessman obtains
Means are cheap, occur being specifically chosen the marketing activity of Internet company with low cost even zero cost and exchange great number reward for
Wool party, cause loss to businessman, mostly use the mode of machine learning to excavate identification wool party at present, but due to internet
Data are general miscellaneous and diversity is big, cause that the recognition efficiency of wool party is lower and accuracy is poor, therefore how to improve wool party
The problem of recognition efficiency and accuracy rate become urgent need to resolve.
Summary of the invention
This application provides a kind of recognition methods of wool party, device, computer equipment and storage mediums, in marketing activity
When rapidly and accurately identify wool party.
In a first aspect, this application provides a kind of recognition methods of wool party, which comprises
The marketing activity of monitors distribution obtains the communication discriminating code for applying for multiple users of the marketing activity；
Each user is positioned using the communication discriminating code to obtain user location letter based on mobile location-based service
Breath generates user location relational graph according to the customer position information；
Judged in multiple users with the presence or absence of certain customers according to the user location relational graph for potential wool party；
If in multiple users there are certain customers be potential wool party, obtain the historical behavior information of multiple users；
Multiple users are clustered according to the historical behavior information to obtain user's monoid, and according to user's monoid
Wool party user is determined from multiple users.
Second aspect, present invention also provides a kind of wool party identification device, described device includes:
Acquiring unit is monitored, for the marketing activity of monitors distribution, obtains the multiple users' for applying for the marketing activity
Communication discriminating code；
Generation unit is positioned, for positioning using the communication discriminating code to each user based on mobile location-based service
To obtain customer position information, user location relational graph is generated according to the customer position information；
User's judging unit, for being judged in multiple users according to the user location relational graph with the presence or absence of certain customers
For potential wool party；
Information acquisition unit, if in multiple users there are certain customers be potential wool party, obtain multiple users'
Historical behavior information；
Determination unit is clustered, for being clustered to obtain user's monoid to multiple users according to the historical behavior information,
And wool party user is determined from multiple users according to user's monoid.
The third aspect, present invention also provides a kind of computer equipment, the computer equipment includes memory and processing
Device；The memory is for storing computer program；The processor, for executing the computer program and described in the execution
Such as above-mentioned wool party recognition methods is realized when computer program.
Fourth aspect, present invention also provides a kind of computer readable storage medium, the computer readable storage medium
It is stored with computer program, the computer program makes the processor realize that above-mentioned wool party such as knows when being executed by processor
Other method.
This application discloses a kind of recognition methods of wool party, device, computer equipment and storage mediums, are finding multiple use
When marketing activity is applied at family, by the communication discriminating code for obtaining multiple users；Mobile location-based service is used using communication discriminating code
Technology generates user location relational graph；Judge multiple users with the presence or absence of potential wool party further according to user location relational graph；If
There are potential wool parties, then obtain the historical behavior information of multiple users；According to the user behavior information to multiple users into
Row cluster obtains user's monoid to determine which user as wool party user.Relative to machine learning mode, this method can be quick
Accurately identify which user is wool party in multiple users of application marketing activity and which user is ordinary user, into
And improve the recognition efficiency and accuracy rate of wool party.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application, below will be to needed in embodiment description
Attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is some embodiments of the present application, general for this field
For logical technical staff, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of schematic flow diagram for wool party recognition methods that embodiments herein provides；
Fig. 2 is the effect diagram for the user location relational graph that embodiments herein provides；
Fig. 3 is the sub-step schematic flow diagram of the wool party recognition methods in Fig. 1；
Fig. 4 is the schematic flow diagram clustered based on K central point algorithm that embodiments herein provides；
Fig. 5 is the schematic flow diagram for another wool party recognition methods that embodiments herein provides；
Fig. 6 is a kind of schematic block diagram for wool party identification device that embodiments herein provides；
Fig. 7 is the schematic block diagram for another wool party identification device that embodiments herein provides；
Fig. 8 is a kind of structural representation block diagram for computer equipment that embodiments herein provides.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiment is some embodiments of the present application, instead of all the embodiments.Based on this Shen
Please in embodiment, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall in the protection scope of this application.
Flow chart shown in the drawings only illustrates, it is not necessary to including all content and operation/step, also not
It is that must be executed by described sequence.For example, some operation/steps can also decompose, combine or partially merge, therefore practical
The sequence of execution is possible to change according to the actual situation.
Embodiments herein provides a kind of recognition methods of wool party, device, computer equipment and storage medium.Its
In, which can be applied in server, by the marketing activity of real time monitoring businessman's publication, realize wool
The identification of party.Wherein, which can be independent server, or server cluster.
With reference to the accompanying drawing, it elaborates to some embodiments of the application.In the absence of conflict, following
Feature in embodiment and embodiment can be combined with each other.
Referring to Fig. 1, Fig. 1 is a kind of schematic flow diagram for wool party recognition methods that embodiments herein provides.It should
The recognition methods of wool party can be applied in server, for identifying to the user by terminal application marketing activity, with true
Recognize with the presence or absence of wool party.
As shown in Figure 1, the wool party recognition methods specifically includes step S101 to step S105.
The marketing activity of S101, monitors distribution obtain the communication discriminating code for applying for multiple users of the marketing activity.
Currently, businessman in order to promote product, can generally issue relevant to the popularization product in corresponding application program
Marketing activity, for example give red packet, discount coupon, head and singly exempt from single or first singles' folding etc..Publication marketing activity is specifically: described
The corresponding link of marketing activity is added on webpage in application program where the product.Wherein the application package includes local answer
With or Web application etc..
The marketing activity that monitoring businessman issues in application program refers to the marketing that monitoring businessman issues in application program
Whether the corresponding link of activity is opened and applies for the marketing activity.User applies for that the marketing activity needs to fill in some registration letters
Breath is filled in some registration informations such as cell-phone number, identification card number, email address and is led to participate in and obtain marketing activity
Address etc. is interrogated, wherein cell-phone number can be used as communication discriminating code.
The marketing activity is participated in due to after marketing activity is issued, might have wool party and multiple ordinary users, such as
What quickly distinguishes and identifies to wool party most important.
S102, each user is positioned using the communication discriminating code to obtain user position based on mobile location-based service
Confidence breath generates user location relational graph according to the customer position information.
Wherein, mobile location-based service is specially LBS positioning service, and LBS positioning service is based on location-based service
(Location Based Services, LBS) calculates terminal (user) institute by the base station signal difference of mobile communication
Position.
Specifically, where calculating the corresponding terminal of communication discriminating code that each user uses using LBS positioning service
Geographical location information, and carry out getting label ready in preset situational map according to the geographical location information that positioning obtains, to generate
User location relational graph.Preset situational map can be used one only include location information map, on the map only
Location information, without other rendering layers.
In one embodiment, user location relational graph, detailed process are generated according to the customer position information are as follows: obtain
Multiple users, are directed into the default map template, and described by default map template according to the customer position information
Line is carried out to generate user location relational graph to multiple users in default map template.
Specifically, default map template only can include the map of location information for one, according to customer position information
It determines corresponding coordinate points, these coordinate points is directed into the default map template, and in the default map template
To multiple coordinate points (user) carry out line to generate user location relational graph, can by each coordinate points with other coordinate points
Connection.
In one embodiment, line is carried out to multiple users in the default map template to generate user location pass
System's figure, specifically includes:
According to shortest path first, line is carried out to generate user location to multiple users in the default map template
Relational graph, the user location relational graph are a closed loop relational graph.
Specifically, select the corresponding coordinate points of any one user as starting point, root in the default map template
Find the coordinate points of the shortest path apart from the starting point according to shortest path first, and by the coordinate of the starting point and shortest path
Point carries out line, and the coordinate points of shortest path are continued to execute as starting point and described found according to shortest path first apart from this
The step of coordinate points of the shortest path of initial point, obtains user location until traversing all the points in the default map template
Relational graph.
It is starting point with user 1 referring particularly to Fig. 2, finds the shortest path apart from the starting point using shortest path first
The coordinate points of diameter are user 3, then carry out line to user 1 and user 3, then continue to be calculated with shortest path strength for starting point with user 3
Method finds the coordinate points of the shortest path apart from the user 3, circuits sequentially and executes until traversing all corresponding coordinates of user
Point obtains user location relational graph 10.It should be noted that user location relational graph 10 is a simple example.
S103, judged in multiple users with the presence or absence of certain customers according to the user location relational graph for potential wool
Party.
Since the quantity of wool party is more and it is organized to be, wool party all has corresponding property parameters, such as
Wool party is full-time application marketing activity, and it is constant generally to stay in some fixation position, it is possible thereby to according to user position
Relational graph judgement is set this time to apply for multiple users of the marketing activity whether there may be potential wool parties.For example, applying for the battalion
It sells movable to use with the presence or absence of multiple per family in same position region, which is such as in the corresponding cell in base station, such as
Fruit has multiple users in uniform location region, than if any 20 in the same cell, then can tentatively judging multiple per family
In user there are certain customers be potential wool party.
In one embodiment, potential wool party is judged according to the user location relational graph, using judging whether there is
The user of preset quantity corresponds to same position in the user location relational graph.As shown in figure 3, i.e. step S103 includes: son
Step S103a and S103b.
S103a, judge that the user in multiple users with the presence or absence of preset quantity is corresponding in the user location relational graph
Same position；The user of S103b, the if it exists preset quantity corresponds to same position in the user location relational graph, really
It is calmly potential wool party there are certain customers in multiple users.
Wherein, preset quantity can be set based on practical experience, for accurately judging exist in multiple users
Certain customers are potential wool party, and specific value is it is not limited here.Same position is corresponded in the user location relational graph,
The same position can be in the accuracy rating of LBS location-based service, such as in the distance range of 5M, naturally it is also possible to be defined as pre-
If in regional scope, such as the corresponding cell in base station.
In one embodiment, preset quantity according to preset ratio relational expression and can apply for the multiple of the marketing activity
The corresponding number of users of user is determining, preset ratio relational expression are as follows:
M=a*n (1)
Wherein, in formula (1), m is preset quantity, and a is proportionality coefficient, and n is the multiple users for applying for the marketing activity
Corresponding number of users.Proportionality coefficient can be according to the ratio-dependent of Historical Monitoring to wool party and total application user, certainly also
It can be with the ratio-dependent of the Historical Monitoring in marketing activity issuing time section to wool party and total application user, when being used for current
Between judge to whether there is in multiple users certain customers in section for potential wool party.For example, previous hour is issued in marketing activity
Proportionality coefficient, a hour in two hours proportionality coefficient or marketing activity issue intraday proportionality coefficient.Root
Preset quantity is determined using corresponding proportionality coefficient according to the corresponding period, it is possible thereby to which the identification for improving potential wool party is accurate
Rate.
If in S104, multiple users there are certain customers be potential wool party, obtain the historical behavior information of multiple users.
The historical behavior information of user includes the equipment usage behavior of user, information act of revision, login behavior and borrows
The information such as loan behavior.Such as equipment usage behavior includes being associated with the information such as multiple users using an equipment；Information modification row
Being includes whether modifying contact information or certificate address information after bank card binding；The login behavior preset time of referring to over is long
The frequency is logged in the period of degree, logs in place and login time etc.；Lend-borrow action refers to that the past period application is borrowed
The number that the money frequency and the amount of money, application pass through or refuses, whether refund in time after borrowing money successfully etc..
In one embodiment, the historical behavior information for obtaining multiple users, comprising: obtain the history of multiple users
Behavior record information, each historical behavior record information includes multiple behavioural characteristic data；The historical behavior is remembered
Behavioural characteristic data in record information carry out data cleansing and obtain historical behavior information.
It includes equipment usage behavior, information act of revision, login behavior and lend-borrow action etc. that historical behavior, which records information,
Information, these information are behavioural characteristic data；Behavioural characteristic data in historical behavior record information are located in advance
Reason, pretreatment are mainly data cleansing duplicate removal and missing values interpolation etc., and pretreated historical behavior record information, which is used as, to be used
The historical behavior information at family so as to the step of according to historical behavior information execution following clustering, and then improves wool party
Recognition accuracy.
S105, multiple users are clustered according to the historical behavior information to obtain user's monoid, and according to the use
Family monoid determines wool party user from multiple users.
After obtaining historical behavior information, using clustering algorithm, according to the historical behavior information to multiple users into
Row cluster obtains user's monoid, and determines that the corresponding user of user's monoid for having certain a kind of is wool party user.Wherein cluster
Partition clustering algorithm, such as K-means algorithm and CLARANS algorithm etc. can be used in algorithm, naturally it is also possible to use other types
Algorithm, such as fuzzy algorithmic approach and based on density algorithm etc..
In one embodiment, described that multiple users are clustered to obtain user class according to the historical behavior information
Group, comprising: be based on K central point algorithm, multiple users clustered according to the historical behavior information to obtain user's monoid.
In one embodiment, as shown in figure 4, carrying out cluster based on K central point algorithm specifically includes the following contents:
S105a, one group of user is chosen in multiple users as target's center's point set, according to the historical behavior information meter
Calculate the distance for each central point that each user concentrates to target's center's point；S105b, according to calculated distance by the use
Family point is sorted out in the shortest clustering cluster of the central point, and each user's point and other users in each clustering cluster are calculated
The sum of the distance of point；The shortest user's point of sum of the distance of S105c, determination and the other users point are as new central point structure
At new center point set；S105d, using the new center point set as target's center's point set, and return execution according to the history row
For information calculate each central point that each user concentrates to target's center's point apart from the step of, until the new center point set
User's monoid is obtained when identical as target's center's point set.
After determining target's center's point set, calculates target's center's point and concentrate all sample points except central point (described more
User in a user other than central point) arrive each central point distance.It specifically, can be with Euclidean distance.Specifically may be used
To calculate the Euclidean distance between user using the historical behavior information of user as characteristic.User's point is being calculated
To after the distance of each central point, the distance of same user point to different central points is compared to determine most short distance
From, and user's point is grouped into the corresponding clustering cluster of shortest distance central point.
For example, for non-central point D, being distinguished based on above-mentioned formula if center point set includes tri- central points of A, B and C
The distance between point D and central point A, B and C La, Lb and Lc are calculated, and three distances are compared, determines distance most
User's point D is referred in the corresponding clustering cluster of central point A by small value if being La apart from minimum value.Corresponding each non-central point
Identical mode is executed, to obtain multiple clustering clusters.
After classification obtains each clustering cluster, calculate in each clustering cluster each user's point at a distance from other users point it
With.For example, if in the corresponding clustering cluster of central point A including central point are as follows: A, D, E and F, then calculate separately A point and D, E and
The sum of the distance of F point is denoted as SUMa；The sum of the distance for calculating D point and A, E and F point is denoted as SUMd；Calculate E point and A, D and
The sum of the distance of F point is denoted as SUMe；The sum of the distance for calculating F point and A, D and E point is denoted as SUMf.
The determining and shortest user's point of other users point sum of the distance, each user's point constitute new center point set；By institute
New center point set is stated as target's center's point set, and returns to execution and each user is calculated to the mesh according to the historical behavior information
Mark central point concentrate each central point apart from the step of, until the new center point set it is identical as target's center's point set when
User's monoid is obtained, i.e., until all central points do not change, and terminates calculating, obtains final clustering cluster, this is final
Clustering cluster as user's monoid.
The recognition methods of above-mentioned wool party passes through the communication of the multiple users of acquisition when finding that multiple users apply for marketing activity
Identification code；User location relational graph is generated using mobile location-based service technology using communication discriminating code；It is closed further according to user location
System's figure judges multiple users with the presence or absence of potential wool party；Potential wool party if it exists, then obtain the historical behavior of multiple users
Information；Clustered to obtain user's monoid to determine which user as wool party to multiple users according to the user behavior information
User.Relative to machine learning mode, this method can rapidly and accurately identify application marketing activity multiple users in which
User is wool party and which user is ordinary user, and then improves the recognition efficiency and accuracy rate of wool party.
Referring to Fig. 4, Fig. 4 is the schematic flow diagram for another wool party recognition methods that embodiments herein provides.
As shown in figure 4, the wool party recognition methods specifically includes step S201 to S207.
The marketing activity of S201, monitors distribution obtain the communication discriminating code for applying for multiple users of the marketing activity.
The marketing activity that monitoring businessman issues in application program, and obtain user and apply for the hand filled in when the marketing activity
Machine number can be used as communication discriminating code.
S202, each user is positioned using the communication discriminating code to obtain user position based on mobile location-based service
Confidence breath generates user location relational graph according to the customer position information.
Specifically, where calculating the corresponding terminal of communication discriminating code that each user uses using LBS positioning service
Geographical location information, and carry out getting label ready in preset situational map according to the geographical location information that positioning obtains, to generate
User location relational graph.
S203, wool location diagram is obtained.
Wherein, wool location diagram is the relational graph of default setting, and the wool location diagram includes having identified
Multiple wool parties and multiple wool leading Party groups at wool network of personal connections.The wool location diagram can store in server pair
The database answered, obtain wool location diagram be specially read from database pre-stored wool be where relational graph,
For differentiating that whether there is certain customers in multiple users is potential wool party.
S204, it whether determines in multiple users comprising the user in the wool network of personal connections.
Specifically, the user location relational graph is compared with the wool location diagram, determines multiple users
In whether comprising be located at the wool network of personal connections in user.User comprising being located in the wool network of personal connections refers to: user
The position of multiple users and the wool party user in the wool network of personal connections in location diagram match, for example position is identical
Or position range is identical, such as in the corresponding cell in base station；Alternatively, multiple users in user location relational graph are located at
In wool network of personal connections.
If thening follow the steps S205 comprising the user being located in the wool network of personal connections in multiple users；If multiple users
In comprising be located at the wool network of personal connections in user, then return to step S201.
S205, determine in multiple users there are certain customers be potential wool party.
Specifically, if determining and existing in multiple users comprising the user being located in the wool network of personal connections in multiple users
Certain customers are potential wool party, and can determine that the multiple users this time applied in the marketing activity are with rapid preliminary includes wool
Party, it is possible to damage the interests of businessman.
S206, the historical behavior information for obtaining multiple users.
There are certain customers in determining multiple users after potential wool party, to obtain the historical behavior letter of multiple users
Breath.Wherein, historical behavior information includes equipment usage behavior, information act of revision, login behavior and the lend-borrow action of user
Etc. information.
S207, multiple users are clustered according to the historical behavior information to obtain user's monoid, and according to the use
Family monoid determines wool party user from multiple users.
After obtaining historical behavior information, using clustering algorithm, according to the historical behavior information to multiple users into
Row cluster obtains user's monoid, and determines that the corresponding user of user's monoid for having certain a kind of is wool party user.Wherein cluster
Partition clustering algorithm, such as K-means algorithm and CLARANS algorithm etc. can be used in algorithm, naturally it is also possible to use other types
Algorithm, such as fuzzy algorithmic approach and based on density algorithm etc..
The recognition methods of above-mentioned wool party passes through the communication of the multiple users of acquisition when finding that multiple users apply for marketing activity
Identification code；User location relational graph is generated using mobile location-based service technology using communication discriminating code；In conjunction with preset wool
Location diagram rapidly and accurately judges multiple users with the presence or absence of potential wool party；Potential wool party if it exists then obtains more
The historical behavior information of a user；Clustered to obtain user's monoid to multiple users according to the user behavior information with determination
Which user is wool party user.Relative to machine learning mode, this method can rapidly and accurately identify application marketing activity
Multiple users in which user be wool party and which user is ordinary user, and then improve the recognition efficiency of wool party
And accuracy rate.
Referring to Fig. 6, Fig. 6 is that embodiments herein provides a kind of schematic block diagram of wool party identification device, the sheep
Hair party's identification device is for executing wool party above-mentioned recognition methods.Wherein, which can be configured at service
Device.
As shown in fig. 6, the wool party identification device 400, comprising: monitoring acquiring unit 401, is used positioning generation unit 402
Family judging unit 403, information acquisition unit 404 and cluster determination unit 405.
Acquiring unit 401 is monitored, for the marketing activity of monitors distribution, obtains the multiple users for applying for the marketing activity
Communication discriminating code.
Generation unit 402 is positioned, for carrying out using the communication discriminating code to each user based on mobile location-based service
Positioning generates user location relational graph to obtain customer position information, according to the customer position information.
In one embodiment, positioning generation unit 402 is specifically used for: default map template is obtained, according to the user
Multiple users are directed into the default map template by location information, and in the default map template to multiple users into
Row line is to generate user location relational graph.
User's judging unit 403, for being judged in multiple users according to the user location relational graph with the presence or absence of part
User is potential wool party.
In one embodiment, user's judging unit 403, the use for judging to whether there is preset quantity in multiple users
Family corresponds to same position in the user location relational graph；The user of the preset quantity is closed in the user location if it exists
Be that same position is corresponded in figure, determine in multiple users there are certain customers be potential wool party.
Information acquisition unit 404, if in multiple users there are certain customers be potential wool party, obtain multiple users
Historical behavior information.
In one embodiment, information acquisition unit 404 are specifically used for: obtaining the historical behavior record letter of multiple users
Breath, each historical behavior record information includes multiple behavioural characteristic data；To in historical behavior record information
Behavioural characteristic data carry out data cleansing and obtain historical behavior information.
Determination unit 405 is clustered, for being clustered to obtain user class to multiple users according to the historical behavior information
Group, and wool party user is determined from multiple users according to user's monoid.
In one embodiment, determination unit 405 is clustered, for being based on K central point algorithm, is believed according to the historical behavior
Breath clusters multiple users to obtain user's monoid.
In one embodiment, determination unit 405 is clustered, is specifically used for: choosing one group of user's conduct in multiple users
Target's center's point set, according to the historical behavior information calculate each user's point to each central point distance, by each user's point
It is put into the shortest clustering cluster of the central point；
Calculate the sum of the distance of each user's point and other users point in each clustering cluster；The determining and other users
The shortest user's point of sum of the distance of point constitutes new center point set as new central point；Using the new center point set as target
Center point set, and return execute according to the historical behavior information calculate each user's point to each central point apart from the step of, directly
User's monoid is obtained when identical as target's center's point set to the new center point set.
Referring to Fig. 7, Fig. 7 is that embodiments herein provides a kind of schematic block diagram of wool party identification device, the sheep
Hair party's identification device is for executing wool party above-mentioned recognition methods.Wherein, which can be configured at service
Device.
As shown in fig. 7, wool party identification device 500, comprising: monitoring acquiring unit 501, positioning generation unit 502, relationship
Figure acquiring unit 503, user's determination unit 504, wool party determination unit 505, information acquisition unit 506 and cluster determination unit
507。
Acquiring unit 501 is monitored, for the marketing activity of monitors distribution, obtains the multiple users for applying for the marketing activity
Communication discriminating code.
Generation unit 502 is positioned, for carrying out using the communication discriminating code to each user based on mobile location-based service
Positioning generates user location relational graph to obtain customer position information, according to the customer position information.
Relational graph acquiring unit 503, for obtaining wool location diagram.
Wherein, the wool location diagram include the multiple wool parties identified and multiple wool leading Party groups at sheep
Hair network of personal connections.
User's determination unit 504, for the user location relational graph to be compared with the wool location diagram,
It whether determines in multiple users comprising the user in the wool network of personal connections.
Wherein, if calling wool party determination unit comprising the user being located in the wool network of personal connections in multiple users
505；If not returning to calling monitoring acquiring unit 501 comprising the user being located in the wool network of personal connections in multiple users.
Wool party determination unit 505, if for, comprising the user being located in the wool network of personal connections, being determined in multiple users
In multiple users there are certain customers be potential wool party.
Information acquisition unit 506, for obtaining the historical behavior information of multiple users.
Determination unit 507 is clustered, for being clustered to obtain user class to multiple users according to the historical behavior information
Group, and wool party user is determined from multiple users according to user's monoid.
It should be noted that it is apparent to those skilled in the art that, for convenience of description and succinctly,
The device of foregoing description and the specific work process of each unit, can refer to corresponding processes in the foregoing method embodiment, herein
It repeats no more.
Above-mentioned device can be implemented as a kind of form of computer program, which can be as shown in Fig. 8
Computer equipment on run.
Referring to Fig. 8, Fig. 8 is a kind of structural representation block diagram for computer equipment that embodiments herein provides.It should
Computer equipment can be server.
Refering to Fig. 8, which includes processor, memory and the network interface connected by system bus,
In, memory may include non-volatile memory medium and built-in storage.
Non-volatile memory medium can storage program area and computer program.The computer program includes program instruction,
The program instruction is performed, and processor may make to execute any one wool party recognition methods.
Processor supports the operation of entire computer equipment for providing calculating and control ability.
Built-in storage provides environment for the operation of the computer program in non-volatile memory medium, the computer program quilt
When processor executes, processor may make to execute any one wool party recognition methods.
The network interface such as sends the task dispatching of distribution for carrying out network communication.It will be understood by those skilled in the art that
Structure shown in Fig. 8, only the block diagram of part-structure relevant to application scheme, is not constituted to application scheme institute
The restriction for the computer equipment being applied thereon, specific computer equipment may include than more or fewer portions as shown in the figure
Part perhaps combines certain components or with different component layouts.
It should be understood that processor can be central processing unit (Central Processing Unit, CPU), it should
Processor can also be other general processors, digital signal processor (Digital Signal Processor, DSP), specially
With integrated circuit (Application Specific Integrated Circuit, ASIC), field programmable gate array
(Field-Programmable GateArray, FPGA) either other programmable logic device, discrete gate or transistor are patrolled
Collect device, discrete hardware components etc..Wherein, general processor can be microprocessor or the processor be also possible to it is any often
The processor etc. of rule.
Wherein, in one embodiment, the processor is for running computer program stored in memory, with reality
Existing following steps:
The marketing activity of monitors distribution obtains the communication discriminating code for applying for multiple users of the marketing activity；Based on shifting
Dynamic location-based service positions to obtain customer position information each user using the communication discriminating code, according to the user
Location information generates user location relational graph；Judge to use in multiple users with the presence or absence of part according to the user location relational graph
Family is potential wool party；If in multiple users there are certain customers be potential wool party, obtain multiple users historical behavior letter
Breath；Clustered to obtain user's monoid to multiple users according to the historical behavior information, and according to user's monoid from more
Wool party user is determined in a user.
In one embodiment, the processor is described according to customer position information generation user location pass in realization
When system's figure, for realizing:
Default map template is obtained, multiple users are directed by the default map template according to the customer position information
In, and line is carried out to generate user location relational graph to multiple users in the default map template.
In one embodiment, the processor described judges multiple users according to the user location relational graph realizing
In with the presence or absence of certain customers be potential wool party when, for realizing:
Judge that the user in multiple users with the presence or absence of preset quantity corresponds to identical bits in the user location relational graph
It sets；The user of the preset quantity corresponds to same position in the user location relational graph if it exists, determines in multiple users
It is potential wool party there are certain customers.
In one embodiment, the processor realize it is described according to described according to the user location relational graph
When judging to whether there is certain customers in multiple users as potential wool party, for realizing:
Wool location diagram is obtained, the wool location diagram includes the multiple wool parties identified and multiple
Wool leading Party group at wool network of personal connections；The user location relational graph is compared with the wool location diagram, is determined
Whether include the user in the wool network of personal connections in multiple users；If comprising being located at the wool relationship in multiple users
User in net, determine in multiple users there are certain customers be potential wool party.
In one embodiment, the processor is realizing the historical behavior information for obtaining multiple users, for real
It is existing:
The historical behavior record information of multiple users is obtained, each historical behavior record information includes multiple behaviors
Characteristic；Data cleansing is carried out to the behavioural characteristic data in historical behavior record information and obtains historical behavior information.
In one embodiment, the processor realize it is described according to the historical behavior information to multiple users carry out
When cluster obtains user's monoid, for realizing:
Based on K central point algorithm, multiple users are clustered according to the historical behavior information to obtain user's monoid.
In one embodiment, the processor is described based on K central point algorithm in realization, is believed according to the historical behavior
When breath clusters multiple users to obtain user's monoid, for realizing:
One group of user is chosen in multiple users as target's center's point set, and each use is calculated according to the historical behavior information
The distance for each central point that family is concentrated to target's center's point；According to calculated distance by user's point sort out to away from
From in the shortest clustering cluster of the central point, calculate in each clustering cluster each user's point at a distance from other users point it
With；The determining shortest user's point of sum of the distance with the other users point constitutes new center point set as new central point；It will
The new center point set returns to execution and calculates each user described according to the historical behavior information as target's center's point set
Each central point that target's center's point is concentrated apart from the step of, until the new center point set is identical as target's center's point set
When obtain user's monoid.
A kind of computer readable storage medium is also provided in embodiments herein, the computer readable storage medium is deposited
Computer program is contained, includes program instruction in the computer program, the processor executes described program instruction, realizes this
Apply for any one wool party recognition methods that embodiment provides.
Wherein, the computer readable storage medium can be the storage inside of computer equipment described in previous embodiment
Unit, such as the hard disk or memory of the computer equipment.The computer readable storage medium is also possible to the computer
The plug-in type hard disk being equipped on the External memory equipment of equipment, such as the computer equipment, intelligent memory card (Smart
Media Card, SMC), secure digital (Secure Digital, SD) card, flash card (Flash Card) etc..
The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, it is any
Those familiar with the art within the technical scope of the present application, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should all cover within the scope of protection of this application.Therefore, the protection scope of the application should be with right
It is required that protection scope subject to.
Claims (10)
1. a kind of recognition methods of wool party characterized by comprising
The marketing activity of monitors distribution obtains the communication discriminating code for applying for multiple users of the marketing activity；
Each user is positioned using the communication discriminating code to obtain customer position information, root based on mobile location-based service
User location relational graph is generated according to the customer position information；
Judged in multiple users with the presence or absence of certain customers according to the user location relational graph for potential wool party；
If in multiple users there are certain customers be potential wool party, obtain the historical behavior information of multiple users；
Clustered to obtain user's monoid to multiple users according to the historical behavior information, and according to user's monoid from more
Wool party user is determined in a user.
2. wool party according to claim 1 recognition methods, which is characterized in that described raw according to the customer position information
At user location relational graph, comprising:
Default map template is obtained, multiple users are directed into the default map template according to the customer position information,
And line is carried out to generate user location relational graph to multiple users in the default map template.
3. wool party according to claim 1 recognition methods, which is characterized in that described according to the user location relational graph
Judge in multiple users with the presence or absence of certain customers for potential wool party, comprising:
Judge that the user in multiple users with the presence or absence of preset quantity corresponds to same position in the user location relational graph；
The user of the preset quantity corresponds to same position in the user location relational graph if it exists, determines in multiple users
It is potential wool party there are certain customers.
4. wool party according to claim 1 recognition methods, which is characterized in that described according to the user location relational graph
Judge in multiple users with the presence or absence of certain customers for potential wool party, comprising:
Wool location diagram is obtained, the wool location diagram includes the multiple wool parties identified and multiple wools
Leading Party group at wool network of personal connections；
The user location relational graph is compared with the wool location diagram, whether is determined in multiple users comprising position
User in the wool network of personal connections；
If determining in multiple users there are certain customers to be latent comprising the user being located in the wool network of personal connections in multiple users
In wool party.
5. wool party according to claim 1 recognition methods, which is characterized in that the historical behavior for obtaining multiple users
Information, comprising:
The historical behavior record information of multiple users is obtained, each historical behavior record information includes multiple behavioural characteristics
Data；
Data cleansing is carried out to the behavioural characteristic data in historical behavior record information and obtains historical behavior information.
6. wool party according to any one of claims 1 to 5 recognition methods, which is characterized in that described according to the history
Behavioural information clusters multiple users to obtain user's monoid, comprising:
Based on K central point algorithm, multiple users are clustered according to the historical behavior information to obtain user's monoid.
7. wool party according to claim 6 recognition methods, which is characterized in that it is described to be based on K central point algorithm, according to institute
It states historical behavior information to cluster multiple users, obtains user's monoid, comprising:
One group of user is chosen in multiple users as target's center's point set, and each user is calculated according to the historical behavior information and is arrived
The distance for each central point that target's center's point is concentrated；
User's point is sorted out in the shortest clustering cluster of the central point according to calculated distance, calculates each institute
State the sum of the distance of each user's point and other users point in clustering cluster；
The determining shortest user's point of sum of the distance with the other users point constitutes new center point set as new central point；
Using the new center point set as target's center's point set, and returns to execution and each user is calculated according to the historical behavior information
Each central point concentrated to target's center's point apart from the step of, until the new center point set and target's center's point
User's monoid is obtained when collecting identical.
8. a kind of wool party identification device characterized by comprising
Acquiring unit is monitored, for the marketing activity of monitors distribution, obtains the communication for applying for multiple users of the marketing activity
Identification code；
Generation unit is positioned, for being positioned each user to obtain using the communication discriminating code based on mobile location-based service
To customer position information, user location relational graph is generated according to the customer position information；
User's judging unit, for judging that it is latent for whether there is certain customers in multiple users according to the user location relational graph
In wool party；
Information acquisition unit, if in multiple users there are certain customers be potential wool party, obtain the history of multiple users
Behavioural information；
Determination unit is clustered, obtains user's monoid, and root for being clustered according to the historical behavior information to multiple users
Wool party user is determined from multiple users according to user's monoid.
9. a kind of computer equipment, which is characterized in that the computer equipment includes memory and processor；
The memory is for storing computer program；
The processor, for executing the computer program and realization such as claim 1 when executing the computer program
To wool party recognition methods described in any one of 7.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer journey
Sequence, the computer program make the processor realize the sheep as described in any one of claims 1 to 7 when being executed by processor
The recognition methods of hair party.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910696730.2A CN110533500A (en) | 2019-07-30 | 2019-07-30 | User identification method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910696730.2A CN110533500A (en) | 2019-07-30 | 2019-07-30 | User identification method, device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110533500A true CN110533500A (en) | 2019-12-03 |
Family
ID=68661320
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910696730.2A Pending CN110533500A (en) | 2019-07-30 | 2019-07-30 | User identification method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110533500A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106301980A (en) * | 2015-05-28 | 2017-01-04 | 腾讯科技（深圳）有限公司 | A kind of brush amount tool detection method and apparatus |
| CN106557942A (en) * | 2015-09-30 | 2017-04-05 | 百度在线网络技术（北京）有限公司 | A kind of recognition methodss of customer relationship and device |
| CN107798597A (en) * | 2017-10-09 | 2018-03-13 | 上海二三四五金融科技有限公司 | A kind of dynamic excessive risk visitor group detection method and system |
| CN107958382A (en) * | 2017-12-06 | 2018-04-24 | 北京小度信息科技有限公司 | Abnormal behaviour recognition methods, device, electronic equipment and storage medium |
| US20180218369A1 (en) * | 2017-02-01 | 2018-08-02 | Google Inc. | Detecting fraudulent data |
| CN109102303A (en) * | 2018-06-28 | 2018-12-28 | 招联消费金融有限公司 | Risk checking method and relevant apparatus |
| CN109118119A (en) * | 2018-09-06 | 2019-01-01 | 多点生活(成都)科技有限公司 | Air control model generating method and device |
| CN109409918A (en) * | 2018-08-24 | 2019-03-01 | 深圳壹账通智能科技有限公司 | The recognition methods of wool party, device, equipment and storage medium based on user behavior |
| CN109685643A (en) * | 2018-12-13 | 2019-04-26 | 平安科技（深圳）有限公司 | Loan audit risk grade determines method, apparatus, equipment and storage medium |
-
2019
- 2019-07-30 CN CN201910696730.2A patent/CN110533500A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106301980A (en) * | 2015-05-28 | 2017-01-04 | 腾讯科技（深圳）有限公司 | A kind of brush amount tool detection method and apparatus |
| CN106557942A (en) * | 2015-09-30 | 2017-04-05 | 百度在线网络技术（北京）有限公司 | A kind of recognition methodss of customer relationship and device |
| US20180218369A1 (en) * | 2017-02-01 | 2018-08-02 | Google Inc. | Detecting fraudulent data |
| CN107798597A (en) * | 2017-10-09 | 2018-03-13 | 上海二三四五金融科技有限公司 | A kind of dynamic excessive risk visitor group detection method and system |
| CN107958382A (en) * | 2017-12-06 | 2018-04-24 | 北京小度信息科技有限公司 | Abnormal behaviour recognition methods, device, electronic equipment and storage medium |
| CN109102303A (en) * | 2018-06-28 | 2018-12-28 | 招联消费金融有限公司 | Risk checking method and relevant apparatus |
| CN109409918A (en) * | 2018-08-24 | 2019-03-01 | 深圳壹账通智能科技有限公司 | The recognition methods of wool party, device, equipment and storage medium based on user behavior |
| CN109118119A (en) * | 2018-09-06 | 2019-01-01 | 多点生活(成都)科技有限公司 | Air control model generating method and device |
| CN109685643A (en) * | 2018-12-13 | 2019-04-26 | 平安科技（深圳）有限公司 | Loan audit risk grade determines method, apparatus, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Snyder et al. | OR/MS models for supply chain disruptions: A review | |
| CN108295476A (en) | The method and apparatus for determining abnormal interactive account | |
| US8156172B2 (en) | Monitoring and reporting enterprise data using a message-based data exchange | |
| CN108492104A (en) | A kind of resource transfer monitoring method and device | |
| CN100334584C (en) | Information processing appts. information, processing method and program | |
| US20140180978A1 (en) | Instance weighted learning machine learning model | |
| CN106453357A (en) | Network ticket buying abnormal behavior recognition method and system and equipment | |
| CN107146089A (en) | The single recognition methods of one kind brush and device, electronic equipment | |
| CN105744005A (en) | Client positioning and analyzing method and server | |
| US10306068B1 (en) | Call center load balancing and routing management | |
| CN106506454A (en) | Fraud business recognition method and device | |
| CN109670797A (en) | Pay route selecting method, apparatus, equipment and storage medium | |
| CN107403334A (en) | Data processing method and device | |
| CN108304935A (en) | Machine learning model training method, device and computer equipment | |
| CN102713881A (en) | Syndication of multiple service instances | |
| CN106210163A (en) | IP address-based localization method and device | |
| CN106776651A (en) | The processing method and processing device of business dispute | |
| Ferrucci | Pro-active dynamic vehicle routing: real-time control and request-forecasting approaches to improve customer service | |
| CN108696597A (en) | The method and apparatus for pushing marketing message | |
| CN109657998A (en) | A kind of resource allocation methods, device, equipment and storage medium | |
| CN109241084A (en) | Querying method, terminal device and the medium of data | |
| CN108959048A (en) | The method for analyzing performance of modular environment, device and can storage medium | |
| US20140222538A1 (en) | Customer experience management for an organization | |
| Paurobally et al. | Automating negotiation for m-services | |
| CN108520455A (en) | A kind of method and apparatus of recommendation information |
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 |