CN107392627A - A kind of user's relationship cycle relation recognition method based on contacts frequency and contacts index - Google Patents

A kind of user's relationship cycle relation recognition method based on contacts frequency and contacts index Download PDF

Info

Publication number
CN107392627A
CN107392627A CN201710585955.1A CN201710585955A CN107392627A CN 107392627 A CN107392627 A CN 107392627A CN 201710585955 A CN201710585955 A CN 201710585955A CN 107392627 A CN107392627 A CN 107392627A
Authority
CN
China
Prior art keywords
user
contacts
index
similarity
coefficient
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201710585955.1A
Other languages
Chinese (zh)
Other versions
CN107392627B (en
Inventor
胡东
牛桂东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Tandao Information Technology Corp Ltd
Original Assignee
Nanjing Tandao Information Technology Corp Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing Tandao Information Technology Corp Ltd filed Critical Nanjing Tandao Information Technology Corp Ltd
Publication of CN107392627A publication Critical patent/CN107392627A/en
Application granted granted Critical
Publication of CN107392627B publication Critical patent/CN107392627B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/01—Customer relationship services
    • 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)
  • General Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Primary Health Care (AREA)
  • Tourism & Hospitality (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Human Resources & Organizations (AREA)
  • General Health & Medical Sciences (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a kind of user's relationship cycle relation recognition method based on contacts frequency and contacts index, belong to the public praise of social circle of operator and family's set meal recommends the technical field of marketing, this method comprises the following steps：Calculate contacts frequency, the contacts index between two two users；The relationship cycle model according to corresponding to contacts frequency, contacts index and contact period preference establish user, the relationship cycle model include building ring model, life range model and comprehensive circle model；User social contact network model is established based on relationship cycle information corresponding to user；Based on user life range model construction family relationship model；The present invention efficiently solves that existing single customers' hairdo marketing specific aim is poor, and cost return rate is relatively low, and the problem of easily cause customer complaint, by sorting out to user social contact cyberrelationship, family relationship identification, marketing success rate significantly improves the inventive method.

Description

A kind of user's relationship cycle relation recognition method based on contacts frequency and contacts index
Technical field
The invention belongs to the technical field that marketing is recommended in the public praise of social circle of operator and family's set meal, one kind is specifically related to User's relationship cycle relation recognition method based on contacts frequency and contacts index.
Background technology
It has been difficult to attract user, cost of marketing return rate that Internet era traditional single client mass-sends marketing mode on a large scale It is relatively low, and be easy to cause customer complaint.
In the prior art, recommend due to being mainly directed towards single client and marketing, following defect be present in recommendation process of marketing：
First, single customers' hairdo marketing easily causes user to dislike, and triggers customer complaint.
2nd, customer group's feature is not known about, it is impossible to catch group interest point.
3rd, family composition is not known about, family's set meal can not specific aim recommendation.
4th, to the no suitable point of penetration of rete mirabile marketing.
In a word, existing single customers' hairdo marketing specific aim is poor, and cost return rate is relatively low, and easily causes user's throwing Tell.
The content of the invention
Goal of the invention：In order to overcome the deficiencies in the prior art, traditional single client mass-sends marketing mode on a large scale It is difficult to attract user, cost of marketing return rate is relatively low, and the defects of be easy to cause customer complaint, the present invention provides a kind of User's relationship cycle relation recognition method based on contacts frequency and contacts index.
Technical scheme：To achieve the above object, a kind of user based on contacts frequency and contacts index of the invention associates Relation recognition method is enclosed, this method comprises the following steps：This method comprises the following steps：Based on contacts frequency and associate index User's relationship cycle relation
A：Calculate contacts frequency, the contacts index between two two users；
B：The relationship cycle model according to corresponding to contacts frequency, contacts index and contact period preference establish user, the contacts
Enclosing model includes building ring model, life range model and comprehensive circle model；
C：User social contact network model is established based on relationship cycle information corresponding to user；
D：Based on user life range model construction family relationship model.
Further, the step B comprises the following steps：
The contacts frequency of calculating user, contacts frequency=(weight 1* user has call/short message/multimedia message number of days+weight this month 2* user has call/short message/multimedia message week number+weight 3* user to have call/short message/multimedia message ten days number this month this month)/days of the month, One ten days=10 day；The weight 1, weight 2 and weight 3 pass through self-defined setting；
Contacts frequency based on user, building ring contacts index, life range contacts index and the comprehensive circle for calculating user are handed over Toward index.
Further, the contacts frequency based on user, calculate user building ring contacts index, life range contacts index and Comprehensive circle contacts index, specifically includes following steps：
Voice ticket, the short multimedia message ticket of user is arranged, when calculating user job period, the period of life respectively and not differentiating between The talk times of section, the duration of call, short message/multimedia message talk times；
The call time of contacts frequency+coefficient 2* user job periods of building ring contacts index=coefficient 1* user of user Short message talk times+coefficient 5* user's work of the duration of call of number+coefficient 3* user job periods+coefficient 4* user job periods Make the multimedia message talk times of period；
The call time of the contacts frequency of life range contacts index=coefficient 1* user of user+coefficient 2* user life period The short message talk times of the duration of call of number+coefficient 3* user life period+coefficient 4* user life period+coefficient 5* user life The multimedia message talk times of period living；
Contacts frequency+coefficient 2* user of comprehensive circle contacts index=coefficient 1* user of user does not differentiate between the call of period The duration of call+coefficient 4* user that number+coefficient 3* user does not differentiate between the period does not differentiate between short message talk times+coefficient 5* of period User does not differentiate between the multimedia message talk times of period；
The coefficient 1, coefficient 2, coefficient 3 and coefficient 4 pass through self-defined setting；
Maximum in the building ring contacts index, life range contacts index and comprehensive circle contacts index at family is taken, then index Maximum is the relationship cycle belonging to the user.
Further, the step C comprises the following steps：
All users in relationship cycle corresponding to user are divided into a colony first；
Then the contacts similarity between user in the colony is calculated；
The contact tight ness rating between two two users is calculated, the contacts between user are determined by the contact tight ness rating between two two users Similarity；
The user that the low user attaching of similarity is weak contact will be associated in the colony, it is to be based on the group that it is low, which to associate similarity, The self-defined setting of the close angle value of contact in body between user；
The high user of similarity will be associated to be divided into a colony, contacts similarity height is based in the colony between user The self-defined setting of the close angle value of contact；
Ultimately form the colony that multiple internal contacts similarities are high, outside contacts similarity is low；
User is divided into by above-mentioned corresponding colony by the method for iteration for new user in relationship cycle corresponding to user In.
Further, the contact tight ness rating between described two two users is calculated by formula (one)：
Wherein, WjFor period weight, period weight includes user job period weight W1, life period weight W2Not area Weight W at times3, period weight passes through self-defined setting；miFor i-th week talk times；diFor i-th week weight, by self-defined Set, n is week period number.
Further, the step D comprises the following steps：
Based on user life range model, the contacts similarity and contacts period in the model of calculating user life range between user are inclined It is good；
According to the same Customer ID of user, with customer households set meal and with customer accounting code information architecture family relationship model.
Further, the contacts similarity in the model of user life range between user is calculated to be calculated by following steps：
Contacts similarity calculation module between user, associate the journey that Similarity Measure is stored with similarity calculation module Degree, when the contacts similarity calculation module is executed by processor, the contacts similarity between user can be obtained, specifically include with Lower step：
The processing of data symmetrization, returned data storehouse；
Sorted ticket according to calling subscribe, calculate the original position of each calling subscribe, and its ticket number, will sort Good ticket, the original position of user and its ticket number determines, for calculating similarity；
The contacts similarity calculation module write is called to calculate the contacts similarity between user, if rejecting similarity is relatively low It is single；
According to Similarity Measure is associated between user, find the high Home Network/rete mirabile user of contacts similarity and bring same family into In relation, family relationship model is built.
Beneficial effect：The present invention compared with the prior art, this have the advantage that：
The present invention arranges voice ticket, short multimedia message ticket information first, contact contacts frequency, contacts between calculating user two-by-two Index；Then building ring, life range, comprehensive circle model are established according to contacts frequency, contacts index and contact period preference；Secondly Based on common contacts between relationship cycle calculating user, and index of similarity between user is calculated, establish social networks；It is finally based on same Client, same family set meal, found a family key member with accounts information, according to member and user's Similarity Measure, finds high phase Same family relation client is included like Home Network/rete mirabile user of degree；The inventive method efficiently solves existing single customers' hairdo Specific aim of marketing is poor, and cost return rate is relatively low, and the problem of easily cause customer complaint, by being closed to user social contact network System, family relationship identification are sorted out, and marketing success rate significantly improves.
Brief description of the drawings
Fig. 1 is the inventive method flowage structure schematic diagram.
Embodiment
The present invention is further described below in conjunction with the accompanying drawings.
Embodiment one：
A kind of user's relationship cycle relation recognition method based on contacts frequency and contacts index that the present embodiment proposes, reference Fig. 1, this method comprise the following steps：User's relationship cycle relation based on contacts frequency and contacts index
A：Calculate contacts frequency, the contacts index between two two users；
B：The relationship cycle model according to corresponding to contacts frequency, contacts index and contact period preference establish user, the contacts Enclosing model includes building ring model, life range model and comprehensive circle model；
C：User social contact network model is established based on relationship cycle information corresponding to user；
D：Based on user life range model construction family relationship model.
Wherein calculate user contacts frequency, contacts frequency=(weight 1* user have this month call/short message/multimedia message number of days+ Weight 2* user has call/short message/multimedia message week number+weight 3* user to have call/short message/multimedia message ten days number this month this month)/this month day Number, ten days=10 day；The weight 1, weight 2 and weight 3 pass through self-defined setting；Next the contacts frequency based on user Degree, calculate the building ring contacts index, life range contacts index and comprehensive circle contacts index of user；Contacts frequency based on user Degree, the building ring contacts index, life range contacts index and comprehensive circle contacts index of user are calculated, specifically includes following steps：
Voice ticket, the short multimedia message ticket of user is arranged, when calculating user job period, the period of life respectively and not differentiating between The talk times of section, the duration of call, short message/multimedia message talk times；
The call time of contacts frequency+coefficient 2* user job periods of building ring contacts index=coefficient 1* user of user Short message talk times+coefficient 5* user's work of the duration of call of number+coefficient 3* user job periods+coefficient 4* user job periods Make the multimedia message talk times of period；
The call time of the contacts frequency of life range contacts index=coefficient 1* user of user+coefficient 2* user life period The short message talk times of the duration of call of number+coefficient 3* user life period+coefficient 4* user life period+coefficient 5* user life The multimedia message talk times of period living；
Contacts frequency+coefficient 2* user of comprehensive circle contacts index=coefficient 1* user of user does not differentiate between the call of period The duration of call+coefficient 4* user that number+coefficient 3* user does not differentiate between the period does not differentiate between short message talk times+coefficient 5* of period User does not differentiate between the multimedia message talk times of period；
The coefficient 1, coefficient 2, coefficient 3 and coefficient 4 pass through self-defined setting；
Maximum in the building ring contacts index, life range contacts index and comprehensive circle contacts index at family is taken, then index Maximum is the relationship cycle belonging to the user.
User social contact network model is established based on relationship cycle information corresponding to user,
Comprise the following steps：
All users in relationship cycle corresponding to user are divided into a colony first；
Then the contacts similarity between user in the colony is calculated；
The contact tight ness rating between two two users is calculated, the contacts between user are determined by the contact tight ness rating between two two users Similarity；Such as：If the contact tight ness rating and user B and user the C tight ness rating that contacts that user A and user B be present approach, deposit again Approached in user A and user D contact tight ness rating and user D and user the C tight ness rating that contacts, there is user A's and user E again Contact tight ness rating and user E and user the C tight ness rating that contacts approach, or user A and user C also contact with other same subscribers Tight ness rating is also close to then illustrating that user A and user C contacts similarity are higher, user A and user's C common contacts are more, specifically How many user contacts tight ness rating with user A and user C respectively and approached, and depending on viewing system situation, finally will then associate similarity User A and user C much higher, common contacts are assigned in a colony；
The user that the low user attaching of similarity is weak contact will be associated in the colony, it is to be based on the group that it is low, which to associate similarity, The self-defined setting of the close angle value of contact in body between user；
The high user of similarity will be associated to be divided into a colony, contacts similarity height is based in the colony between user The self-defined setting of the close angle value of contact；
Ultimately form the colony that multiple internal contacts similarities are high, outside contacts similarity is low；
User is divided into by above-mentioned corresponding colony by the method for iteration for new user in relationship cycle corresponding to user In；Contact tight ness rating between above-mentioned two two users is calculated by formula (one)：
Wherein, WjFor period weight, period weight includes user job period weight W1, life period weight W2Not area Weight W at times3, period weight passes through self-defined setting；miFor i-th week talk times；diFor i-th week weight, by self-defined Set, n is week period number.
Iteratively divide group, all users are divided into a colony, associate similarity between calculating user, delete Except the weak contact between user, the deletion of Weak link it also avoid all users and be included in same group, social after carrying out this step Network can be degenerated to the secondary network that several internal connections are more, external relation is few, and much isolated node, each network pair Answer a colony (group)；
Remaining users are divided into and contact intensive microcommunity, how to ensure that network internal contact is more intensive, preferentially will altogether Assigned to the more user of contact person in a colony, under this precondition, by the method for iteration according to message registration according to It is secondary that user is included in corresponding group.
Based on user life range model construction family relationship model, based on user life range model, user life range is calculated Contacts similarity in model between user and contacts period preference, according to the same Customer ID of user, with customer households set meal and same Customer accounting code information architecture family relationship model, calculate the contacts similarity in the model of user life range between user and pass through following step Suddenly it is calculated：
Contacts similarity calculation module between user, associate the journey that Similarity Measure is stored with similarity calculation module Degree, when the contacts similarity calculation module is executed by processor, the contacts similarity between user can be obtained, specifically include with Lower step：
The processing of data symmetrization, returned data storehouse；
Sorted ticket according to calling subscribe, calculate the original position of each calling subscribe, and its ticket number, will sort Good ticket, the original position of user and its ticket number determines, for calculating similarity；
The contacts similarity calculation module write is called to calculate the contacts similarity between user, if rejecting similarity is relatively low It is single；
According to Similarity Measure is associated between user, find the high Home Network/rete mirabile user of contacts similarity and bring same family into In relation, family relationship model is built.
Embodiment two：
Based on embodiment one, the present embodiment proposes a kind of based on the user's relationship cycle relation for associating frequency and contacts index Recognition methods, the basic thought of the present embodiment are：Voice ticket, short multimedia message ticket information are arranged, is contacted between calculating user two-by-two Associate frequency, contacts index；
Building ring, life range, comprehensive circle model are established according to contacts frequency, contacts index and contact period preference；
Based on common contacts between relationship cycle calculating user, and index of similarity between user is calculated, establish social networks；
Based on same client, same family set meal, found a family key member with accounts information, according to member and user's similarity Calculate, the Home Network/rete mirabile user for finding high similarity includes same family relation client.
Established based on relationship cycle information corresponding to user in user social contact network model, for building the social network of relationship cycle Network, the intensive group of contact is found out, thought deeply here from the angle of common contacts, when considering two users being included in a group, Wish that two users there are more common contacts, and user is added step-wise to correspondingly by main body by way of iteratively faster Colony in；
Based in the model construction family relationship model of user life range, for building family relationship identification model, connection is calculated It is the similarity of both sides, it is with more number similarities under client higher, and associate the period and concentrate on the life period, it is believed that with client The possibility of the same family is higher；
User social contact network model is established and based on user life range model construction based on relationship cycle information corresponding to user Contacts similarity in family relationship model between middle user can call the get_same.c programs write to calculate by following Similarity；
Call get_same.c programs：
Step1:Data symmetrization handles (symmetrize=weak/strong two ways), returned data storehouse；
Step2:Sorted ticket according to calling subscribe, calculate the original position (st) of each calling subscribe, and its ticket Number (cnt), the ticket to have sorted, st, cnt are returned into R, associate similarity between user for calculating, as shown in table 1：
Table 1
Step3:Call the get_same.c programs write to calculate similarity, it is (adjustable to reject the relatively low ticket of similarity Percentage), as shown in table 2；
Table 2
Grouping result is as shown in table 3 below：G is group number, and count (1) is group's number of users；
Table 3
Above-mentioned divides group to be initial key group, the isolated point that will be rejected by get_same.c similarities program, is assigned to The closest key group with isolated point：
Table 4
After tenant group, by connectedness, authoritative index (Page ranking value), out-degree, the in-degree for dividing group to calculate each user Etc., connectivity between user can be further indicated that, as associated the high rete mirabile user of similarity in table 5, finds contacts similarity High rete mirabile user is brought into relational model；
Table 5
Embodiment three：
Based on the method for embodiment two, to the number that day wing emotional affection net obtains with the inventive method in China Telecom of operator According to as follows：
Circle is marketed
Emotional affection net marketing success rate significantly improves：From in by the end of July, 2013, emotional affection net short message distribute leaflets amount to 6347971, battalion It is only 0.4% to sell success rate；
On July 29th, 2013, life range is built by social networks, picks out leader, distribute leaflets 9255, with client The customer volume for participating in emotional affection net is 522, success rate 5.64% of marketing, takes full advantage of leader in daily social work With；
August 12 in 2013, by way of reducing colony of conversing, find out the user for concentration of conversing two-by-two, distribute leaflets amount 203350, there are 2624 users to participate in, success rate 1.3%；
In mid-August, 2013, are carried out by emotional affection net recommendation, is amounted to for " life range ", " building ring ", " comprehensive circle " data respectively The family of distribute leaflets 66354,916 families are successfully handled, success rate 1.4% of marketing.In by the end of August, 2013, coordinates Huaian LAN to carry out fine Change marketing, choose 4249 users and carry out emotional affection net marketing, the family of successful user 184, marketing success rate is 4.33%, is further illustrated The marketing success rate of the inventive method significantly improves.
Described above is only the preferred embodiment of the present invention, it should be pointed out that：Come for those skilled in the art Say, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should be regarded as Protection scope of the present invention.

Claims (7)

  1. A kind of 1. user's relationship cycle relation recognition method based on contacts frequency and contacts index, it is characterised in that：This method bag Include following steps：User's relationship cycle relation based on contacts frequency and contacts index
    A：Calculate contacts frequency, the contacts index between two two users；
    B：The relationship cycle model according to corresponding to contacts frequency, contacts index and contact period preference establish user, the relationship cycle mould Type includes building ring model, life range model and comprehensive circle model；
    C：User social contact network model is established based on relationship cycle information corresponding to user；
    D：Based on user life range model construction family relationship model.
  2. 2. user's relationship cycle relation recognition method according to claim 1 based on contacts frequency and contacts index, it is special Sign is：The step B comprises the following steps：
    The contacts frequency of calculating user, contacts frequency=(weight 1* user has call/short message/multimedia message number of days+weight 2* to use this month Family this month, has call/short message/multimedia message week number+weight 3* user to have call/short message/multimedia message ten days number this month)/days of the month, a ten days =10 days；The weight 1, weight 2 and weight 3 pass through self-defined setting；
    Contacts frequency based on user, the building ring contacts index, life range contacts index and comprehensive circle contacts for calculating user refer to Number.
  3. 3. user's relationship cycle relation recognition method according to claim 2 based on contacts frequency and contacts index, it is special Sign is：
    Contacts frequency based on user, the building ring contacts index, life range contacts index and comprehensive circle contacts for calculating user refer to Number, specifically includes following steps：
    Voice ticket, the short multimedia message ticket of user is arranged, user job period, the period of life is calculated respectively and does not differentiate between the period Talk times, the duration of call, short message/multimedia message talk times；
    The talk times of contacts frequency+coefficient 2* user job periods of building ring contacts index=coefficient 1* user of user+ Short message talk times+coefficient 5* user jobs of the duration of call of coefficient 3* user job periods+coefficient 4* user job periods The multimedia message talk times of period；
    The talk times of the contacts frequency of life range contacts index=coefficient 1* user of user+coefficient 2* user life period+ The short message talk times of the duration of call of coefficient 3* user life period+coefficient 4* user life period+coefficient 5* user life The multimedia message talk times of period；
    Contacts frequency+coefficient 2* user of comprehensive circle contacts index=coefficient 1* user of user does not differentiate between the talk times of period The duration of call+coefficient 4* user that+coefficient 3* user does not differentiate between the period does not differentiate between short message talk times+coefficient 5* user of period The multimedia message talk times of period are not differentiated between；
    The coefficient 1, coefficient 2, coefficient 3 and coefficient 4 pass through self-defined setting；
    Maximum in the building ring contacts index, life range contacts index and comprehensive circle contacts index at family is taken, then index is maximum Be the user belonging to relationship cycle.
  4. 4. user's relationship cycle relation recognition method according to claim 1 based on contacts frequency and contacts index, it is special Sign is：The step C comprises the following steps：
    All users in relationship cycle corresponding to user are divided into a colony first；
    Then the contacts similarity between user in the colony is calculated；
    The contact tight ness rating between two two users is calculated, determines that the contacts between user are similar by the contact tight ness rating between two two users Degree；
    The user that the low user attaching of similarity is weak contact will be associated in the colony, it is low, which to associate similarity, is based in the colony The self-defined setting of the close angle value of contact between user；
    The high user of similarity will be associated to be divided into a colony, contacts similarity height is based on the connection between user in the colony Fasten the self-defined setting of density value；
    Ultimately form the colony that multiple internal contacts similarities are high, outside contacts similarity is low；
    User is divided into above-mentioned corresponding colony by the method for iteration for new user in relationship cycle corresponding to user.
  5. 5. user's relationship cycle relation recognition method according to claim 4 based on contacts frequency and contacts index, it is special Sign is：Contact tight ness rating between described two two users is calculated by formula (one)：
    Wherein, WjFor period weight, period weight includes user job period weight W1, life period weight W2During with not differentiating between Duan Quanchong W3, period weight passes through self-defined setting；miFor i-th week talk times；diFor i-th week weight, set by self-defined Put, n is week period number.
  6. 6. user's relationship cycle relation recognition method according to claim 1 based on contacts frequency and contacts index, it is special Sign is：The step D comprises the following steps：
    Based on user life range model, the contacts similarity and contacts period preference in the model of user life range between user are calculated；
    According to the same Customer ID of user, with customer households set meal and with customer accounting code information architecture family relationship model.
  7. 7. user's relationship cycle relation recognition method according to claim 6 based on contacts frequency and contacts index, it is special Sign is：The contacts similarity in the model of user life range between user is calculated to be calculated by following steps：
    Contacts similarity calculation module between user, the degree that Similarity Measure is stored with similarity calculation module is associated, when When the contacts similarity calculation module is executed by processor, the contacts similarity between user can be obtained, specifically includes following step Suddenly：
    The processing of data symmetrization, returned data storehouse；
    Sorted ticket according to calling subscribe, the original position of each calling subscribe, and its ticket number are calculated, by what is sorted Ticket, the original position of user and its ticket number determine, for calculating similarity；
    Call the contacts similarity calculation module write to calculate the contacts similarity between user, reject the relatively low ticket of similarity；
    According to Similarity Measure is associated between user, find the high Home Network/rete mirabile user of contacts similarity and bring same family relation into In, build family relationship model.
CN201710585955.1A 2016-07-18 2017-07-18 User interaction circle relation identification method based on interaction frequency and interaction index Active CN107392627B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2016105637015 2016-07-18
CN201610563701.5A CN106228371A (en) 2016-07-18 2016-07-18 A kind of social network analysis based on the ultra-large user associating frequency and associate index and family relation recognizer

Publications (2)

Publication Number Publication Date
CN107392627A true CN107392627A (en) 2017-11-24
CN107392627B CN107392627B (en) 2021-08-10

Family

ID=57519620

Family Applications (2)

Application Number Title Priority Date Filing Date
CN201610563701.5A Pending CN106228371A (en) 2016-07-18 2016-07-18 A kind of social network analysis based on the ultra-large user associating frequency and associate index and family relation recognizer
CN201710585955.1A Active CN107392627B (en) 2016-07-18 2017-07-18 User interaction circle relation identification method based on interaction frequency and interaction index

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN201610563701.5A Pending CN106228371A (en) 2016-07-18 2016-07-18 A kind of social network analysis based on the ultra-large user associating frequency and associate index and family relation recognizer

Country Status (1)

Country Link
CN (2) CN106228371A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108428155A (en) * 2018-03-20 2018-08-21 南京邮电大学 A kind of behavior processing analysis method based on service feature model
CN110337059A (en) * 2018-03-30 2019-10-15 中国联合网络通信集团有限公司 A kind of parser, server and the network system of subscriber household relationship

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN106875276A (en) * 2016-12-30 2017-06-20 广州市申迪计算机系统有限公司 A kind of communication user social circle generation method and system
CN108268477A (en) * 2016-12-30 2018-07-10 中国电信股份有限公司 A kind of customer relationship based on ticket determines method and apparatus
CN108632132A (en) * 2017-03-20 2018-10-09 大有秦鼎（北京）科技有限公司 The determination method and apparatus of relationship cycle
CN110019996A (en) * 2017-12-11 2019-07-16 中国移动通信集团广东有限公司 A kind of family relationship recognition methods and system
CN108108419B (en) * 2017-12-15 2021-01-26 百度在线网络技术（北京）有限公司 Information recommendation method, device, equipment and medium
CN107958423A (en) * 2017-12-20 2018-04-24 广东宜通世纪科技股份有限公司 User's social relationships analysis method and storage medium, server-side
CN108628946A (en) * 2018-03-30 2018-10-09 维沃移动通信有限公司 A kind of method for information display, device and mobile terminal
CN111324741B (en) * 2018-12-17 2023-08-18 中国移动通信集团山西有限公司 User relationship identification method, device, equipment and medium
CN109829485A (en) * 2019-01-08 2019-05-31 科大国创软件股份有限公司 A kind of user relationship mining method and system based on mobile data
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
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
CN111160977A (en) * 2019-12-31 2020-05-15 中国移动通信集团黑龙江有限公司 Method, device, equipment and medium for acquiring user relation interest characteristic graph
CN111866146B (en) * 2020-07-22 2022-02-11 中国联合网络通信集团有限公司 Household user identification method and device
CN112163885A (en) * 2020-09-29 2021-01-01 贵州力创科技发展有限公司 User relationship mining method and system based on mobile communication data
CN112506063B (en) * 2020-11-25 2024-05-07 中移（杭州）信息技术有限公司 Data analysis method, system, electronic device and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030110140A1 (en) * 2001-06-28 2003-06-12 Louis Morrison Method for facilitating pricing, sale and distribution of fuel to a customer
CN101540739A (en) * 2009-04-14 2009-09-23 腾讯科技（深圳）有限公司 User recommendation method and user recommendation system
CN101764866A (en) * 2009-12-30 2010-06-30 宇龙计算机通信科技(深圳)有限公司 Method and system for displaying contact person communication frequency, and mobile terminal
JP2012175213A (en) * 2011-02-18 2012-09-10 Nec Casio Mobile Communications Ltd Portable electronic device, control method thereof, and program
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

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030110140A1 (en) * 2001-06-28 2003-06-12 Louis Morrison Method for facilitating pricing, sale and distribution of fuel to a customer
CN101540739A (en) * 2009-04-14 2009-09-23 腾讯科技（深圳）有限公司 User recommendation method and user recommendation system
CN101764866A (en) * 2009-12-30 2010-06-30 宇龙计算机通信科技(深圳)有限公司 Method and system for displaying contact person communication frequency, and mobile terminal
JP2012175213A (en) * 2011-02-18 2012-09-10 Nec Casio Mobile Communications Ltd Portable electronic device, control method thereof, and program
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

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
TORE OPSAHL: "Closeness centrality in networks with disconnected components", 《HTTPS://TOREOPSAHL.COM/2010/03/20/CLOSENESS-CENTRALITY-IN-NETWORKS-WITH-DISCONNECTED-COMPONENTS/》 *
王铁牛: "移动用户社交网络发现与应用", 《中国优秀硕士学位论文全文数据库经济与管理科学辑》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108428155A (en) * 2018-03-20 2018-08-21 南京邮电大学 A kind of behavior processing analysis method based on service feature model
CN110337059A (en) * 2018-03-30 2019-10-15 中国联合网络通信集团有限公司 A kind of parser, server and the network system of subscriber household relationship
CN110337059B (en) * 2018-03-30 2021-03-12 中国联合网络通信集团有限公司 Analysis algorithm, server and network system for family relationship of user

Also Published As

Publication number Publication date
CN107392627B (en) 2021-08-10
CN106228371A (en) 2016-12-14

Similar Documents

Publication Publication Date Title
CN107392627A (en) A kind of user's relationship cycle relation recognition method based on contacts frequency and contacts index
CN110032637A (en) A kind of resume intelligent recommendation algorithm based on natural semantic analysis technology
CN110413748A (en) Based on big data robot service knowledge base entry recommender system
CN109299372A (en) A kind of wisdom curricula-variable recommended method based on collaborative filtering
CN106203935B (en) Technical capability evaluation and Postmatch method based on user-generated content and customer relationship
CN106776941A (en) A kind of method of the effective solutionist of recommendation based on mass-rent pattern
CN110046740A (en) Supplier's bid behavioural analysis prediction technique based on big data
CN106951471A (en) A kind of construction method of the label prediction of the development trend model based on SVM
CN109740046A (en) Aerial double choosings based on internet recruitment can platform
CN110147427A (en) Project case method for pushing and device
CN110020099A (en) A kind of the user's recommended method and device of video friend-making
CN109543041A (en) A kind of generation method and device of language model scores
CN105681287A (en) Screening rule based user service allocation screening method
CN105183806A (en) Method and system for identifying same user among different platforms
CN102750288B (en) A kind of internet content recommend method and device
Kotnik et al. ICT as Enabler of Exports
CN106649280A (en) Method for creating shared corpus
CN108133296B (en) Event attendance prediction method combining environmental data under social network based on events
CN110210884A (en) Determine the method, apparatus, computer equipment and storage medium of user characteristic data
CN107679907A (en) Advertisement audience generation method and system
Sen Complexities of social networks: A Physicist's perspective
CN110110084A (en) The recognition methods of high quality user-generated content
CN108595674A (en) Social networks recommend method
CN109451449A (en) A kind of mobile communication 4G set meal downshift prediction technique based on decision Tree algorithms
CN111078888A (en) Method for automatically classifying comment data of social network users

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant