CN107958423A - User's social relationships analysis method and storage medium, server-side - Google Patents
User's social relationships analysis method and storage medium, server-side Download PDFInfo
- Publication number
- CN107958423A CN107958423A CN201711385718.7A CN201711385718A CN107958423A CN 107958423 A CN107958423 A CN 107958423A CN 201711385718 A CN201711385718 A CN 201711385718A CN 107958423 A CN107958423 A CN 107958423A
- Authority
- CN
- China
- Prior art keywords
- targeted customer
- level
- user
- social relationships
- contact person
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 31
- 238000004891 communication Methods 0.000 claims abstract description 60
- 238000005065 mining Methods 0.000 claims abstract description 23
- 230000006399 behavior Effects 0.000 claims description 50
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000000034 method Methods 0.000 claims description 13
- 239000000284 extract Substances 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 6
- 238000013075 data extraction Methods 0.000 abstract 1
- 230000006872 improvement Effects 0.000 description 9
- 238000009412 basement excavation Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000010295 mobile communication Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 230000007812 deficiency Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of user's social relationships analysis method to include step：According to the level-one contact person of communication behavior data extraction targeted customer of the targeted customer in preset time period；Targeted customer's second level contact is extracted according to communication behavior data of the targeted customer level-one contact person in preset time period；To be all common contacts of the second level contact as targeted customer and targeted customer level-one contact person of targeted customer level-one contact person；The social relationships information of user couple is calculated according to the communication behavior data between the All Contacts in preset time period；User after calculating inputs into social relationships information mining model trained in advance, so as to obtain the social relationships of the level-one contact person of targeted customer and targeted customer.The invention also discloses a kind of user's social relationships analytical equipment, storage medium and server-side.User's social relationships analysis method of the embodiment of the present invention realizes the accurate judgement of user's social relationships by excavating the scheme of targeted customer's profound level correspondence.
Description
Technical field
The present invention relates to mobile communication technology field, more particularly to a kind of user's social relationships analysis method and storage to be situated between
Matter, server-side.
Background technology
With the development of mobile communication technology and the popularization of mobile phone, the population coverage of mobile communications network is very high, moves
Dynamic communication data (including call ticket, short message ticket and location data etc.) has objectively responded the behavior of user from many aspects
Feature and social relationships feature.
The social relationships of excavation crowd are subject to the extensive concern of multiple industry fields, excavate social relationships and orient battalion to user
Sell, advertisement accurately pushes, it is significant to open up new user group etc.；Offender and aggrieved can be grasped in security fields
The family of person, friend, Peer Relationships, important information is provided for safety-related department.
Existing technical solution has weighed " power " of customer relationship mostly, not can determine that " classification " of customer relationship but,
Or can only determine a certain specific user's relation, or decision rule needs manually to set by experience.
Existing user's social relationships method for digging mainly has three kinds, and first method is by telecommunications speech path network number
According to, first divide two high users of weight into a relationship cycle, then calculate with user in circle have the other users of contacts with should
The degree of membership of relationship cycle, the high user of degree of membership are included in relationship cycle, and iteration obtains complete multiple relationship cycles；But this method is not
It can determine that the attribute of a relation of each relationship cycle, the member of household, colleague or a variety of social relationships of friend may be included in relationship cycle,
Social relationships between user are indefinite.Second method is that the user that call occurred is formed user couple first, then logical to it
Words data are associated rule process, and two users of correlation threshold will be met as the user for having family relationship；This method is only
It can identify the family relationship of user, and correlation threshold must be manually set so as to cause accuracy and science deficiency.The third
Method is the working time and non-working time of first definite user, the score between user is calculated according to the duration of call, by normalizing
Coefficient of relationship is obtained after change processing, social relationships are judged by coefficient of relationship；The party needs to set user's score rule by experience
And it is also required to close by the society for having had experience to lay down a regulation to judge between user after the coefficient of relationship of user is obtained
System, so that the applicability deficiency and efficiency of this method are low.
The content of the invention
The purpose of the embodiment of the present invention is to provide a kind of user's social relationships analysis method and device, storage medium, service
End, can effectively solve the problems, such as that customer relationship existing in the prior art is indefinite, determination methods not science and inefficiency,
Realization reduces manual intervention when judging user's social relationships, improves reliability and accuracy that user's social relationships are excavated.
To achieve the above object, an embodiment of the present invention provides a kind of user's social relationships analysis method, including step：
The level-one contact person of the targeted customer is extracted according to communication behavior data of the targeted customer in preset time period；
Wherein, the level-one contacts the artificially contact person with the targeted customer there are communication behavior；
Communication behavior data of the level-one contact person of the targeted customer in preset time period are obtained, and according to the mesh
Mark the second level contact that communication behavior data of the level-one contact person of user in preset time period extract the targeted customer；Its
In, the artificial contact person with the level-one contact person there are communication behavior of the second degree contacts；
When the second level contact of the targeted customer is the level-one contact person of the targeted customer at the same time, then by the mesh
Mark common contacts of the second level contact of user as the targeted customer and the level-one contact person of the targeted customer；
The targeted customer is extracted with the level-one group of contacts of the targeted customer into user couple, and according in preset time
The communication behavior data of the targeted customer, the level-one contact person of the targeted customer and the common contacts calculate in section
Obtain the social relationships information of the user couple；
The social relationships mining model that the social relationships information input for the user couple being calculated is trained in advance, from
And obtain the social relationships of the targeted customer and the level-one contact person of the targeted customer.
Compared with prior art, user's social relationships analysis method disclosed by the invention, by according to targeted customer pre-
If the communication behavior data in the period extract the level-one contact person of the targeted customer；Obtain the cascade of the targeted customer
It is communication behavior data of the people in preset time period, and according to the level-one contact person of the targeted customer in preset time period
Communication behavior data extract the second level contact of the targeted customer；When the second level contact of the targeted customer is at the same time institute
When stating the level-one contact person of targeted customer, then using the second level contact of the targeted customer as the targeted customer and the mesh
Mark the common contacts of the level-one contact person of user；Extract the level-one group of contacts of the targeted customer and the targeted customer into
User couple, and according to the targeted customer, the level-one contact person of the targeted customer and described common in preset time period
The social relationships information of the user couple is calculated in the communication behavior data of contact person；By the user's couple being calculated
Social relationships information input social relationships mining model trained in advance, so as to obtain the targeted customer and the targeted customer
The method of social relationships of level-one contact person reduce manual intervention when judging user's social relationships, deeper analysis
The contact relationship of user, improves reliability, accuracy and the applicability of user's social relationships excavation.
As the improvement of such scheme, the communication behavior data include communication data and position data；Wherein, it is described logical
Letter data includes the active communication behavior of user and the data of passive communication behavior.
As the improvement of such scheme, the social relationships information is included in the targeted customer and institute in preset time period
State call note data, cohesion, contact circle and the positional information between the level-one white silk contact person of targeted customer.
As the improvement of such scheme, the call note data is specifically included in the targeted customer in preset time period
Talk times, call number of days between the level-one white silk contact person of the targeted customer, the duration of call, call most period,
The single duration of call, short message number, short message period, short message content average length.
As the improvement of such scheme, the cohesion specifically includes call cohesion and short message cohesion；Wherein, it is described
Call cohesion calculation be：
Converse cohesion=[(AB talk times/B calls total degree) * ln (the level-one contact persons of all number number/A of A)
+ (AB talk times/A calls total degree) * ln (second level contacts of all number number/A of B)]/2
The calculation of the short message cohesion is：
Short message cohesion=[(AB short messages number/B short messages total degree) * ln (the level-one contact persons of all number number/A of A)
+ (AB short messages number/A calls total degree) * ln (second level contacts of all number number/A of B)]/2
Wherein, A is the targeted customer, and B is the level-one contact person of any targeted customer.
As the improvement of such scheme, the contact circle specifically includes the targeted customer and is cascaded with the targeted customer one
It is that the common connection number of people and the targeted customer and targeted customer's level-one contact person contact circle similarity；Wherein,
The calculation for contacting circle similarity of the targeted customer and targeted customer's level-one contact person is：
AB contact circle similarity=[(AB common connections number) * ln (the contact number of the contact number * B of A)]/(connection of A
It is the contact number-AB common connections number of number+B)
Wherein, A is the targeted customer, and B is the level-one contact person of any targeted customer.
As the improvement of such scheme, the positional information specifically includes the targeted customer, the one of the targeted customer
Level contact person and the common contacts day part positional distance；Wherein, the calculation of the positional distance is：
Distance d=R*accs [sin (π/180* latitudes 1) * sin (π/180* latitudes 2)+cos (π/180* latitudes 1) * cos
(π/180* latitudes 2) * cos (π/180* (latitude 1- latitudes 2))]
Wherein, R represents earth radius, latitude 1 and latitude 2 represent respectively same user in the position latitude of different periods or
Position latitude of the different user in the same period.
As the improvement of such scheme, by the social relationships information input training in advance for the user couple being calculated
Social relationships mining model, so that before obtaining the social relationships of the level-one contact person of the targeted customer and the targeted customer
Further include：
Obtain training sample of the associated person information of targeted customer's known relation as the social relationships mining model
This；
The sample is input to the social relationships mining model and obtains the social relationships mining model output
Sample results；
The social relationships mining model is optimized with the associated person information according to the sample results.
The embodiment of the present invention additionally provides a kind of storage medium, and the storage medium includes the program of storage, wherein, in institute
State user's social relationships analysis method described in equipment execution any of the above-described where controlling the storage medium during program operation.
The embodiment of the present invention additionally provides a kind of server-side, including one or more processors；Memory；And
One or more programs, wherein one or more of programs are stored in the memory, and are configured
To be performed by one or more of processors, described program includes being used to perform user's social relationships described in any of the above-described
Analysis method.
Brief description of the drawings
Fig. 1 is a kind of flow diagram for user's social relationships analysis method that the embodiment of the present invention 1 provides.
Fig. 2 is a kind of structure diagram for user's social relationships analytical equipment that the embodiment of the present invention 2 provides.
Embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution 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, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other without creative efforts
Embodiment, belongs to the scope of protection of the invention.
It is a kind of flow diagram for user's social relationships analysis method that the embodiment of the present invention 1 provides referring to Fig. 1, bag
Include step S1~S5：
S1, the level-one for extracting according to communication behavior data of the targeted customer in preset time period the targeted customer contact
People；Wherein, the level-one contacts the artificially contact person with the targeted customer there are communication behavior；
Preferably, the communication behavior data include communication data and position data；Wherein, the communication data includes using
The active communication behavior at family and the data of passive communication behavior.
S2, obtain communication behavior data of the level-one contact person of the targeted customer in preset time period, and according to institute
State the second degree contacts that communication behavior data of the level-one contact person of targeted customer in preset time period extract the targeted customer
People；Wherein, the artificial contact person with the level-one contact person there are communication behavior of the second degree contacts；
When S3, the second level contact as the targeted customer while the level-one contact person for the targeted customer, then by institute
State common contacts of the second level contact of targeted customer as the targeted customer and the level-one contact person of the targeted customer；
S4, extract the targeted customer with the level-one group of contacts of the targeted customer into user couple, and according to default
The communication behavior data of the targeted customer, the level-one contact person of the targeted customer and the common contacts in period
The social relationships information of the user couple is calculated；
Preferably, the social relationships information is included in the targeted customer and the targeted customer in preset time period
Level-one practices call note data, cohesion, contact circle and the positional information between contact person.
Specifically, the call note data is specifically included in the targeted customer in preset time period and is used with the target
The level-one at family practice talk times between contact person, call number of days, the duration of call, call most period, the single duration of call,
Short message number, short message period, short message content average length.
Specifically, the cohesion specifically includes call cohesion and short message cohesion；Wherein, the call cohesion
Calculation is：
Converse cohesion=[(AB talk times/B calls total degree) * ln (the level-one contact persons of all number number/A of A)
+ (AB talk times/A calls total degree) * ln (second level contacts of all number number/A of B)]/2
The calculation of the short message cohesion is：
Short message cohesion=[(AB short messages number/B short messages total degree) * ln (the level-one contact persons of all number number/A of A)
+ (AB short messages number/A calls total degree) * ln (second level contacts of all number number/A of B)]/2
Wherein, A is the targeted customer, and B is the level-one contact person of any targeted customer.
Specifically, the contact circle specifically includes the common connection of the targeted customer and targeted customer's level-one contact person
It is number and the targeted customer and targeted customer's level-one contact person contacts circle similarity；Wherein, the targeted customer
With targeted customer's level-one contact person contact circle similarity calculation be：
AB contact circle similarity=[(AB common connections number) * ln (the contact number of the contact number * B of A)]/(connection of A
It is the contact number-AB common connections number of number+B)
Wherein, A is the targeted customer, and B is the level-one contact person of any targeted customer.
Specifically, the positional information specifically includes the targeted customer, the level-one contact person of the targeted customer and institute
State positional distance of the common contacts in day part；Wherein, the calculation of the positional distance is：
Distance d=R*accs [sin (π/180* latitudes 1) * sin (π/180* latitudes 2)+cos (π/180* latitudes 1) * cos
(π/180* latitudes 2) * cos (π/180* (latitude 1- latitudes 2))]
Wherein, R represents earth radius, latitude 1 and latitude 2 represent respectively same user in the position latitude of different periods or
Position latitude of the different user in the same period.
S5, the social relationships excavation mould for training the social relationships information input for the user couple being calculated in advance
Type, so as to obtain the social relationships of the targeted customer and the level-one contact person of the targeted customer.
Further, step is further included before the step S5：
The social relationships mining model that the social relationships information input for the user couple being calculated is trained in advance, from
And the targeted customer is obtained with being further included before the social relationships of the level-one contact person of the targeted customer：
Obtain training sample of the associated person information of targeted customer's known relation as the social relationships mining model
This；
The sample is input to the social relationships mining model and obtains the social relationships mining model output
The social relationships of sample are as sample results；
The sample results are compared with the associated person information of targeted customer's known relation, and according to the ratio
Result optimizes the social relationships mining model.
Further, the social relationships mining model establishes machine learning model using Naive Bayes Classifier,
Verification to the model will also be carried out after optimization training by, which completing, assesses, the verification evaluation index include accuracy rate, precision and
Recall rate.By taking the social relationships classification of targeted customer is " close relation household " as an example (assessment mode of other classifications is similar),
The social relationships classification of targeted customer is defined as positive example for the associated person information of " close relation household " first, rather than it is " close
The associated person information of contact household " is defined as negative example.The targeted customer recorded according to sample information and the contact person of targeted customer
Fiel can be related to that the social relationships classification of classification and model output carries out cross validation, four kinds of results can be obtained.Referring to
Table 1, table 1 export result for model and illustrate table.
Table 1
Wherein, real example represents that model output is positive example, and true classification is also the situation of positive example；False positive example represents that model is defeated
It is the situation of negative example to go out for positive example, true classification；Very negative example represents model output as negative example, and true classification is also the feelings of negative example
Condition；The negative example of vacation represents model output as negative example, and true classification is the situation of positive example.It can be counted as follows according to the result of table 1
Calculate formula：Accuracy rate=(real example+very negative example)/sample number；Precision=real example/(real example+vacation positive example)；Recall rate=true
Positive example/(example is born in real example+vacation).Processing is optimized to the social mining model according to the verification evaluation index.
Preferably, as the artificial multiple contact persons of the level-one contact of the targeted customer, with the targeted customer with it is described
The level-one contact of either objective user is one group of user couple, the basis targeted customer, mesh in preset time period
Mark the level-one contact person of user and the communication behavior data of the common contacts establish the communication number of the corresponding user couple
According to table.User A and B in the present embodiment represent that there are the user of communication behavior mutually.Referring to table 2, table 2 is the communication of user couple
Tables of data.The communication data table includes 23 fields, is divided into call short message class field, cohesion class field, connection according to function
System's circle class field, positional distance class field and tag field；Wherein, the 1st, No. 2 field represents record identification；3-11 fields
The talk times of user A and B, call number of days, the duration of call, call most period, the secondary duration of call, in short-term are represented respectively
Call accounting, short message number, short message number of days and short message average length；12nd, No. 13 field represent respectively call familiarity index and
Short message familiarity index；It is based respectively on and converses with each other data and each other note data, weighs the intimate degree between user A and B；The
14th, No. 15 fields represent common connection number, contact circle similarity respectively；Wherein contact circle similarity is with the contact of user A
People, the contact person of user B are calculated with reference to common contacts Number synthesis；16-22 fields represent respectively job site away from
With a distance from, place of abode, working hour positional distance coefficient, inhabitation slot position distance coefficient, weekend positional distance coefficient, altogether
With contact person place of working distance coefficient, common contacts residence distance coefficient；Wherein, position of every kind of distance all in accordance with user
User job in data, residence and the longitude and latitude where each hour subdistrict position, be calculated by range formula；The
No. 23 fields represent social relationships classifications, NB Algorithm modelling phase, it is necessary to the sample of known social relationships classification
This record information, the field represent the label of known sample record, include " intimate household ", " common household ", " colleague ", " friend
Friend " and " other " totally five classifications, need not the field when model judges user's social relationships.
Table 2
Specifically, the data of the positional distance class field of the user are according to the cell information table of user and persistent district
Table, is calculated job site, the user place of abode of user, and user is resident the most subdistrict position of duration each hour
(i.e. each most 24 subdistrict positions points of user)；Wherein, the cell calculation of longitude ＆ latitude in the cell information table obtains
User job, residence and the longitude and latitude where each hour subdistrict position.
Based on such scheme, by extracting the target according to communication behavior data of the targeted customer in preset time period
The level-one contact person of user；Communication behavior data of the level-one contact person of the targeted customer in preset time period are obtained, and
The two of the targeted customer is extracted according to communication behavior data of the level-one contact person of the targeted customer in preset time period
Level contact person；When the second level contact of the targeted customer is the level-one contact person of the targeted customer at the same time, then by described in
Common contacts of the second level contact of targeted customer as the targeted customer and the level-one contact person of the targeted customer；Carry
The level-one group of contacts of the targeted customer and the targeted customer are taken into user couple, and according to the mesh in preset time period
The use is calculated in mark user, the level-one contact person of the targeted customer and the communication behavior data of the common contacts
The social relationships information at family pair；By the social relationships information input for the user couple being calculated social relationships trained in advance
Mining model, so that the method for obtaining the social relationships of the level-one contact person of the targeted customer and the targeted customer utilizes Piao
Plain bayesian algorithm reduces manual intervention when judging user's social relationships, is contacted by the level-one for extracting the targeted customer
The common contacts of people, the second level contact of the targeted customer and the targeted customer and the level-one contact person realize
It is deeper analysis user contact relationship, improve user's social relationships excavation reliability, accuracy and be applicable in
Property.
Referring to Fig. 2, a kind of structure diagram of the user's social relationships analytical equipment provided for the embodiment of the present invention 2, institute
Stating user's social relationships analytical equipment 100 includes：
First contact person's extraction module 101, for being carried according to communication behavior data of the targeted customer in preset time period
Take the level-one contact person of the targeted customer；Wherein, artificially there are communication behavior with the targeted customer for the level-one contact
Contact person；
Second contact person's extraction module 102, for obtaining the level-one contact person of the targeted customer in preset time period
Communication behavior data, and extracted according to the communication behavior data of the level-one contact person of the targeted customer in preset time period
The second level contact of the targeted customer；Wherein, artificially there are communication behavior with the level-one contact person for the second degree contacts
Contact person；
Common contacts extraction module 103, is used for the target at the same time for the second level contact as the targeted customer
During the level-one contact person at family, then using the second level contact of the targeted customer as the targeted customer with the targeted customer's
The common contacts of level-one contact person；
First analysis module 104, for extract the level-one group of contacts of the targeted customer and the targeted customer into
Family pair, and according to the targeted customer, the level-one contact person of the targeted customer and the common connection in preset time period
It is that the social relationships information of the user couple is calculated in the communication behavior data of people；
Second analysis module 105, for the social relationships information input for the user couple being calculated to be trained in advance
Social relationships mining model, so as to obtain the social relationships of the level-one contact person of the targeted customer and the targeted customer.
An embodiment of the present invention provides a kind of user's social relationships analytical equipment, passes through first contact person's extraction module 101
The level-one contact person of the targeted customer, the second contact are extracted according to communication behavior data of the targeted customer in preset time period
People's extraction module 102 obtains communication behavior data of the level-one contact person of the targeted customer in preset time period, and according to
Communication behavior data of the level-one contact person of the targeted customer in preset time period extract two cascades of the targeted customer
It is people, common contacts extraction module 103 extracts the mesh according to the level-one contact person of the targeted customer and second level contact
The common contacts of user and the level-one contact person of the targeted customer are marked, the first analysis module 104 extracts the targeted customer
Level-one group of contacts with the targeted customer is into user couple, and according to the targeted customer, the mesh in preset time period
Mark the level-one contact person of user and the society pass of the user couple is calculated in the communication behavior data of the common contacts
It is information, the social relationships information for the user couple being calculated is inputted society trained in advance by the second analysis module 105
Relation excavation model, so that the targeted customer and the method for the social relationships of the level-one contact person of the targeted customer are obtained,
Reduce manual intervention when judging user's social relationships using NB Algorithm, by extract the targeted customer one
The common contacts of level contact person, the second level contact of the targeted customer and the targeted customer and the level-one contact person
Realize it is deeper analysis user contact relationship, improve user's social relationships excavation reliability, accuracy with
And applicability.
Specifically, user A and B familiarity indexes include call cohesion and short message cohesion.
Cohesion calculation of conversing is as follows：
Converse cohesion=[(AB talk times/B calls total degree) * ln (the level-one contact persons of all number number/A of A)
+ (AB talk times/A calls total degree) * ln (second level contacts of all number number/A of B)]/2
Short message cohesion calculation is as follows：
Short message cohesion=[(AB short messages number/B short messages total degree) * ln (the level-one contact persons of all number number/A of A)
+ (AB short messages number/A calls total degree) * ln (second level contacts of all number number/A of B)]/2
Specifically, the calculation of the range formula is：
Distance d=R*accs [sin (π/180* latitudes 1) * sin (π/180* latitudes 2)+cos (π/180* latitudes 1) * cos
(π/180* latitudes 2) * cos (π/180* (latitude 1- latitudes 2))]
Wherein, R represents earth radius, latitude 1 and latitude 2 represent respectively same user in the position latitude of different periods or
Position latitude of the different user in the same period.
In a further preferred embodiment, it is not present when between the targeted customer and the level-one contact person of the targeted customer
During common contacts, communication behavior data of the common contacts in preset time period are sky, then the social relationships are dug
Dig communication behavior data of the model according to the level-one contact person of the targeted customer and the targeted customer in preset time period
Obtain the social relationships of the targeted customer.
The embodiment of the present invention additionally provides a kind of storage medium, it is characterised in that the storage medium includes the journey of storage
Sequence, wherein, equipment where the storage medium is controlled when described program is run performs the user society described in any of the above-described
Relationship analysis method.
The embodiment of the present invention additionally provides a kind of server-side, including one or more processors；Memory；And
One or more programs, wherein one or more of programs are stored in the memory, and are configured
To be performed by one or more of processors, described program includes being used to perform user's social relationships described in any of the above-described
Analysis method.
The above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also considered as
Protection scope of the present invention.
Claims (10)
- A kind of 1. user's social relationships analysis method, it is characterised in that the method includes the steps：The level-one contact person of the targeted customer is extracted according to communication behavior data of the targeted customer in preset time period；Its In, the level-one contacts the artificially contact person with the targeted customer there are communication behavior；Communication behavior data of the level-one contact person of the targeted customer in preset time period are obtained, and are used according to the target Communication behavior data of the level-one contact person at family in preset time period extract the second level contact of the targeted customer；Wherein, The artificial contact person with the level-one contact person there are communication behavior of the second degree contacts；When the second level contact of the targeted customer is the level-one contact person of the targeted customer at the same time, then the target is used Common contacts of the second level contact at family as the targeted customer and the level-one contact person of the targeted customer；The targeted customer is extracted with the level-one group of contacts of the targeted customer into user couple, and according in preset time period The communication behavior data of the targeted customer, the level-one contact person of the targeted customer and the common contacts are calculated The social relationships information of the user couple；By the social relationships information input for the user couple being calculated social relationships mining model trained in advance, so that To the social relationships of the targeted customer and the level-one contact person of the targeted customer.
- 2. user's social relationships analysis method as claimed in claim 1, it is characterised in that the communication behavior data include logical Letter data and position data；Wherein, the communication data includes the active communication behavior of user and the data of passive communication behavior.
- 3. user's social relationships analysis method as claimed in claim 1, it is characterised in that the social relationships information is included in The level-one of the targeted customer and the targeted customer practice call note data between contact person, intimate in preset time period Degree, contact circle and positional information.
- 4. user's social relationships analysis method as claimed in claim 3, it is characterised in that the call note data specifically wraps Include talk times, the call day in preset time period between the level-one of the targeted customer and the targeted customer white silk contact person Number, the duration of call, call most period, the single duration of call, short message number, short message period, short message content average length.
- 5. the user's social relationships analysis method stated such as claim 3, it is characterised in that the cohesion specifically includes call parent Density and short message cohesion；Wherein, the calculation of the call cohesion is：Converse cohesion=[(AB talk times/B calls total degree) * ln (the level-one contact persons of all number number/A of A)+(AB Talk times/A calls total degree) * ln (second level contacts of all number number/A of B)]/2The calculation of the short message cohesion is：Short message cohesion=[(AB short messages number/B short messages total degree) * ln (the level-one contact persons of all number number/A of A)+(AB Short message number/A calls total degree) * l n (second level contacts of all number number/A of B)]/2Wherein, A is the targeted customer, and B is the level-one contact person of any targeted customer.
- 6. the user's social relationships analysis method stated such as claim 3, it is characterised in that the contact circle specifically includes the mesh Mark the common connection number and the targeted customer and targeted customer's level-one of user and targeted customer's level-one contact person The contact circle similarity of contact person；Wherein, the targeted customer and targeted customer's level-one contact person contact circle similarity Calculation be：AB contact circle similarity=[(AB common connections number) * ln (the contact number of the contact number * B of A)]/(contact persons of A Contact number-AB common connections the number of number+B)Wherein, A is the targeted customer, and B is the level-one contact person of any targeted customer.
- 7. the user's social relationships analysis method stated such as claim 3, it is characterised in that the positional information specifically includes described The positional distance of targeted customer, the level-one contact person of the targeted customer and the common contacts in day part；Wherein, it is described The calculation of positional distance is：Distance d=R*accs [sin (π/180* latitudes 1) * sin (π/180* latitudes 2)+cos (π/180* latitudes 1) * cos (π/ 180* latitudes 2) * cos (π/180* (latitude 1- latitudes 2))]Wherein, R represents earth radius, and latitude 1 and latitude 2 represent position latitude or difference of the same user in different periods respectively Position latitude of the user in the same period.
- 8. the user's social relationships analysis method stated such as claim 1, it is characterised in that by the user's couple being calculated Social relationships information input social relationships mining model trained in advance, so as to obtain the targeted customer and the targeted customer Level-one contact person social relationships before further include：Obtain training sample of the associated person information of targeted customer's known relation as the social relationships mining model；The sample is input to the social relationships mining model and obtains the sample of the social relationships mining model output As a result；The social relationships mining model is optimized with the associated person information according to the sample results.
- A kind of 9. storage medium, it is characterised in that the storage medium includes the program of storage, wherein,Equipment perform claim described program controls the storage medium when running where requires 1~8 any one of them user Social relationships analysis method.
- A kind of 10. server-side, it is characterised in that including one or more processors, memory and one or more program, its In：One or more of programs are stored in the memory, and are configured as by one or more of processors Perform, described program includes being used for perform claim 1~8 any one of them user's social relationships analysis method of requirement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711385718.7A CN107958423A (en) | 2017-12-20 | 2017-12-20 | User's social relationships analysis method and storage medium, server-side |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711385718.7A CN107958423A (en) | 2017-12-20 | 2017-12-20 | User's social relationships analysis method and storage medium, server-side |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107958423A true CN107958423A (en) | 2018-04-24 |
Family
ID=61955582
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711385718.7A Pending CN107958423A (en) | 2017-12-20 | 2017-12-20 | User's social relationships analysis method and storage medium, server-side |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107958423A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109542950A (en) * | 2018-11-14 | 2019-03-29 | 中国联合网络通信集团有限公司 | Method for digging, device, terminal and the computer readable storage medium of social relationships |
| CN110677269A (en) * | 2018-07-03 | 2020-01-10 | 中国电信股份有限公司 | Method and device for determining communication user relationship and computer readable storage medium |
| CN110688407A (en) * | 2019-09-09 | 2020-01-14 | 创新奇智(南京)科技有限公司 | Social relationship mining method |
| CN110737702A (en) * | 2019-10-22 | 2020-01-31 | 北京明略软件系统有限公司 | Social relationship analysis method and device, computer equipment and readable storage medium |
| CN110851218A (en) * | 2019-10-23 | 2020-02-28 | 中国建设银行股份有限公司 | Personal interface operation function adding method and device based on personnel relationship |
| CN110971770A (en) * | 2019-11-27 | 2020-04-07 | 武汉虹旭信息技术有限责任公司 | Method and system for estimating social relationship sparse density based on ticket data analysis |
| CN111092764A (en) * | 2019-12-18 | 2020-05-01 | 电信科学技术第五研究所有限公司 | Real-time dynamic intimacy relationship analysis method and system |
| CN111510368A (en) * | 2019-01-31 | 2020-08-07 | 中国移动通信有限公司研究院 | Family group identification method, device, equipment and computer readable storage medium |
| CN113115200A (en) * | 2019-12-24 | 2021-07-13 | 中国移动通信集团浙江有限公司 | User relationship identification method and device and computing equipment |
| CN113127751A (en) * | 2019-12-30 | 2021-07-16 | 中移(成都)信息通信科技有限公司 | User portrait construction method, device and equipment and computer readable storage medium |
| CN114173006A (en) * | 2020-09-11 | 2022-03-11 | 中国联合网络通信集团有限公司 | Communication user off-network early warning method and server |
| CN115203480A (en) * | 2022-05-10 | 2022-10-18 | 中国人民解放军91977部队 | Target group formation membership mining method based on deep correlation analysis |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101854439A (en) * | 2010-02-04 | 2010-10-06 | 深圳市同洲电子股份有限公司 | Interpersonal relationship analysis method and system based on ticket |
| CN104244314A (en) * | 2014-07-30 | 2014-12-24 | 北京拓明科技有限公司 | Potential group client identification method based on Mc interface signaling |
| CN106228371A (en) * | 2016-07-18 | 2016-12-14 | 南京坦道信息科技有限公司 | A kind of social network analysis based on the ultra-large user associating frequency and associate index and family relation recognizer |
| CN106686198A (en) * | 2015-11-06 | 2017-05-17 | 中国移动通信集团公司 | Address book management method, intimacy information management method, address book management device, and intimacy information management device |
| CN106686261A (en) * | 2017-01-19 | 2017-05-17 | 腾讯科技（深圳）有限公司 | Information processing method and system |
| CN106899492A (en) * | 2017-02-17 | 2017-06-27 | 上海新炬网络技术有限公司 | A kind of method for digging of customer relationship chain of working together |
| CN106936971A (en) * | 2017-05-08 | 2017-07-07 | 沈阳航空航天大学 | A kind of incoming person's information presentation system and reminding method |
| CN107231494A (en) * | 2017-07-10 | 2017-10-03 | 北京亿赛通网络安全技术有限公司 | A kind of acquisition methods of user communication characteristic, storage medium and electronic equipment |
| CN107241481A (en) * | 2017-06-30 | 2017-10-10 | 广东欧珀移动通信有限公司 | Book management method, device, storage medium and electronic equipment |
| CN107277213A (en) * | 2017-07-27 | 2017-10-20 | 维沃移动通信有限公司 | Contact categories processing method, cloud device and mobile terminal |
-
2017
- 2017-12-20 CN CN201711385718.7A patent/CN107958423A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101854439A (en) * | 2010-02-04 | 2010-10-06 | 深圳市同洲电子股份有限公司 | Interpersonal relationship analysis method and system based on ticket |
| CN104244314A (en) * | 2014-07-30 | 2014-12-24 | 北京拓明科技有限公司 | Potential group client identification method based on Mc interface signaling |
| CN106686198A (en) * | 2015-11-06 | 2017-05-17 | 中国移动通信集团公司 | Address book management method, intimacy information management method, address book management device, and intimacy information management device |
| CN106228371A (en) * | 2016-07-18 | 2016-12-14 | 南京坦道信息科技有限公司 | A kind of social network analysis based on the ultra-large user associating frequency and associate index and family relation recognizer |
| CN106686261A (en) * | 2017-01-19 | 2017-05-17 | 腾讯科技（深圳）有限公司 | Information processing method and system |
| CN106899492A (en) * | 2017-02-17 | 2017-06-27 | 上海新炬网络技术有限公司 | A kind of method for digging of customer relationship chain of working together |
| CN106936971A (en) * | 2017-05-08 | 2017-07-07 | 沈阳航空航天大学 | A kind of incoming person's information presentation system and reminding method |
| CN107241481A (en) * | 2017-06-30 | 2017-10-10 | 广东欧珀移动通信有限公司 | Book management method, device, storage medium and electronic equipment |
| CN107231494A (en) * | 2017-07-10 | 2017-10-03 | 北京亿赛通网络安全技术有限公司 | A kind of acquisition methods of user communication characteristic, storage medium and electronic equipment |
| CN107277213A (en) * | 2017-07-27 | 2017-10-20 | 维沃移动通信有限公司 | Contact categories processing method, cloud device and mobile terminal |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110677269A (en) * | 2018-07-03 | 2020-01-10 | 中国电信股份有限公司 | Method and device for determining communication user relationship and computer readable storage medium |
| CN109542950A (en) * | 2018-11-14 | 2019-03-29 | 中国联合网络通信集团有限公司 | Method for digging, device, terminal and the computer readable storage medium of social relationships |
| CN111510368B (en) * | 2019-01-31 | 2023-01-03 | 中国移动通信有限公司研究院 | Family group identification method, device, equipment and computer readable storage medium |
| CN111510368A (en) * | 2019-01-31 | 2020-08-07 | 中国移动通信有限公司研究院 | Family group identification method, device, equipment and computer readable storage medium |
| CN110688407A (en) * | 2019-09-09 | 2020-01-14 | 创新奇智(南京)科技有限公司 | Social relationship mining method |
| CN110688407B (en) * | 2019-09-09 | 2022-05-17 | 创新奇智(南京)科技有限公司 | Social relationship mining method |
| CN110737702A (en) * | 2019-10-22 | 2020-01-31 | 北京明略软件系统有限公司 | Social relationship analysis method and device, computer equipment and readable storage medium |
| CN110851218A (en) * | 2019-10-23 | 2020-02-28 | 中国建设银行股份有限公司 | Personal interface operation function adding method and device based on personnel relationship |
| CN110851218B (en) * | 2019-10-23 | 2023-08-25 | 中国建设银行股份有限公司 | Personnel relationship-based method and device for adding personal interface operation functions |
| CN110971770A (en) * | 2019-11-27 | 2020-04-07 | 武汉虹旭信息技术有限责任公司 | Method and system for estimating social relationship sparse density based on ticket data analysis |
| CN111092764A (en) * | 2019-12-18 | 2020-05-01 | 电信科学技术第五研究所有限公司 | Real-time dynamic intimacy relationship analysis method and system |
| CN111092764B (en) * | 2019-12-18 | 2023-10-17 | 电信科学技术第五研究所有限公司 | Real-time dynamic affinity relation analysis method and system |
| CN113115200A (en) * | 2019-12-24 | 2021-07-13 | 中国移动通信集团浙江有限公司 | User relationship identification method and device and computing equipment |
| CN113115200B (en) * | 2019-12-24 | 2023-04-18 | 中国移动通信集团浙江有限公司 | User relationship identification method and device and computing equipment |
| CN113127751A (en) * | 2019-12-30 | 2021-07-16 | 中移(成都)信息通信科技有限公司 | User portrait construction method, device and equipment and computer readable storage medium |
| CN113127751B (en) * | 2019-12-30 | 2023-10-27 | 中移(成都)信息通信科技有限公司 | User portrait construction method, device, equipment and computer readable storage medium |
| CN114173006A (en) * | 2020-09-11 | 2022-03-11 | 中国联合网络通信集团有限公司 | Communication user off-network early warning method and server |
| CN115203480A (en) * | 2022-05-10 | 2022-10-18 | 中国人民解放军91977部队 | Target group formation membership mining method based on deep correlation analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107958423A (en) | User's social relationships analysis method and storage medium, server-side | |
| CN108924333A (en) | Fraudulent call recognition methods, device and system | |
| CN107222865B (en) | Communication swindle real-time detection method and system based on suspicious actions identification | |
| CN111159387B (en) | Recommendation method based on multi-dimensional alarm information text similarity analysis | |
| CN106372072A (en) | Location-based recognition method for user relations in mobile social network | |
| Frias-Martinez et al. | Towards large scale technology impact analyses: Automatic residential localization from mobile phone-call data | |
| Paraskevopoulos et al. | Identification and characterization of human behavior patterns from mobile phone data | |
| CN109451182A (en) | A kind of detection method and device of fraudulent call | |
| CN103854065A (en) | Customer loss prediction method and device | |
| CN105678457A (en) | Method for evaluating user behavior on the basis of position mining | |
| CN109800600A (en) | Ocean big data susceptibility assessment system and prevention method towards privacy requirements | |
| CN111754337A (en) | Method and system for identifying credit card maintenance contract group | |
| CN108989581B (en) | User risk identification method, device and system | |
| CN109190916A (en) | Method of opposing electricity-stealing based on big data analysis | |
| Xie et al. | Unsupervised user identity linkage via factoid embedding | |
| CN102083010A (en) | Method and equipment for screening user information | |
| CN110209729A (en) | The method and device of data transfer object identification | |
| CN112053222A (en) | Knowledge graph-based internet financial group fraud detection method | |
| CN109191167A (en) | A kind of method for digging and device of target user | |
| CN106897743B (en) | Mobile attendance anti-cheating big data detection method based on Bayesian model | |
| CN108090787A (en) | A kind of call bill data depth based on Apriori algorithm is excavated and the method for user's behavior prediction | |
| CN111105064A (en) | Method and device for determining suspected information of fraud event | |
| Chu et al. | Exploiting spatial-temporal behavior patterns for fraud detection in telecom networks | |
| Hilas et al. | A genetic programming approach to telecommunications fraud detection and classification | |
| CN108984773A (en) | Blacklist multidimensional information verification method and system and readable storage medium storing program for executing and equipment in the case of shortage of data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Fu Yu Inventor after: Lin Yuyang Inventor after: Liang Yonghua Inventor after: Zhang Yuan Inventor after: Wu Wen Inventor after: Huang Tao Inventor before: Fu Yu Inventor before: Lin Yuyang Inventor before: Liang Yonghua Inventor before: Zhang Yuan Inventor before: Wu Wen Inventor before: Huang Tao |
|
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 510665 East Floor 3, No. 14 and No. 16 Jianzhong Road, Tianhe District, Guangzhou City, Guangdong Province Applicant after: Yitong Century Science and Technology Co., Ltd. Address before: 510665 12/F, Building A, Guangzhou Information Port, 16 Keyun Road, Tianhe District, Guangzhou City, Guangdong Province Applicant before: Guangdong Eastone Technology Co., Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180424 |