CN107527240A - A kind of operator's industry product Praise effect identification system and method - Google Patents

A kind of operator's industry product Praise effect identification system and method Download PDF

Info

Publication number
CN107527240A
CN107527240A CN201710739922.8A CN201710739922A CN107527240A CN 107527240 A CN107527240 A CN 107527240A CN 201710739922 A CN201710739922 A CN 201710739922A CN 107527240 A CN107527240 A CN 107527240A
Authority
CN
China
Prior art keywords
user
data
module
identification
product
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
CN201710739922.8A
Other languages
Chinese (zh)
Other versions
CN107527240B (en
Inventor
肖松明
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 CN107527240A publication Critical patent/CN107527240A/en
Application granted granted Critical
Publication of CN107527240B publication Critical patent/CN107527240B/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
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data

Landscapes

  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a kind of operator's industry product Praise effect identification system, belongs to communication products marketing effectiveness identification technology field, including data acquisition module, data memory module, data processing module and data identification and analysis module；Data acquisition module is used to obtain related data for research object and research product；Data memory module is used for data storage；Data processing module is used to handle data storage；Data authentication analysis module is used to carry out identification and analysis to data, obtains effect identification result；Operator's industry product Praise effect identification method is based on the system, with reference to checking validity module, the intimate model of relationship cycle and space overlap degree model are built, science objective assessment is made to product Praise effect, marketing effectiveness qualification result is more accurate.

Description

A kind of operator's industry product Praise effect identification system and method
Technical field
The invention belongs to operator's field commodity ASCII stream, a kind of operator's industry product public praise is specifically related to Marketing effectiveness identification systems and method.
Background technology
The advertisement putting of telecommunications industry product is presented variation, including traditional tv media, paper media, the network media, Outdoor advertising etc., especially with the rise of mobile Internet, a kind of new, efficient battalion is become by oral marketing marketing Pin means.Initial media and it is undeveloped when, oral marketing mode is just taught orally, so far with the development of communications industry, mouth Upright stone tablet marketing breaches the mode taught orally face-to-face in the past, can be carried out between men by modes such as phone, short messages far The propagation information of distance.Marketed by oral marketing with a high credibility, because in general, oral marketing all occurs in friend Between the more close colony of the relations such as friend, relative, colleague, classmate, had built up before oral marketing process, between them A kind of relation steady in a long-term, for pure advertisement, promotion, public relations, businessman recommend etc., confidence level will be more It is high.
Telecommunications industry is mainly taken out at random when carrying out product Praise effect identification by way of artificial outgoing call at present Sample is investigated, and is taken time and effort；
And user be present because the consideration such as privacy and falseness is answered a question, the accuracy of impact evaluation；
Rely on microblogging, the simple statistics of the entry of social software such as Baidu's mhkc, but be difficult to assess its be forward-propagating also It is that negative sense is propagated；
Also it is difficult to assess the conversion rate of products in communication process simultaneously；
However, user experience data more can effectively identify marketing effectiveness in telecommunications industry, how these data are correctly utilized Identify marketing effectiveness, the problem of being extremely important in marketing effectiveness authentication method in the prior art, also occur a lot Kind of marketing methods Valuation Method, these methods form Valuation Method on the basis of user data, but these valencys Value appraisal procedure is not particularly suited for product marketing effect identification method；And value assessment result is ineffective.
Generally speaking, telecommunications industry not yet utilizes itself peculiar data at present, and maturation is formed on data basis The authentication method of marketing effectiveness.
Therefore, it is necessary to a kind of new Praise effect identification method.
The content of the invention
Goal of the invention：In order to overcome the deficiencies in the prior art, existing Value accounting system is not particularly suited for product Marketing identification, qualification result is inaccurate, and the present invention provides a kind of operator's industry product Praise effect identification system and side Method, science objective assessment can be made to product Praise effect according to user's relationship cycle and space-time trace information；And seek It is more accurate to sell effect identification result.
Technical scheme：To achieve the above object, operator's industry product Praise effect identification system of the invention, bag Include data acquisition module, data memory module, data processing module and data identification and analysis module；
Wherein, the data acquisition module is used to obtain related data, the packet for research object and research product The of that month communication bill of research object and signaling position data are included, wherein communication bill data include the duration of call, short message/coloured silk Creed number, communication time period, communication festivals or holidays type and communication type；
The data memory module is used for data storage；
The data processing module is used to handle data storage；
The data authentication analysis module is used to carry out identification and analysis to data, obtains effect identification result.
Further, the data authentication analysis module includes contact cohesion identification module, space overlap degree identification mould Block and checking validity module；
Wherein, the contact cohesion identification module, based on communication bill data, build user network relation, and to Family relationship cohesion is identified, by the accounting for judging use research product user in strong tie cohesion user group Size identifies the effect of oral marketing；
The space overlap degree identification module, based on signaling position data, each user is portrayed in different time different location Appearance track, the space overlap degree between user is identified, by judging in strong association user colony that use research produces The accounting size of product user identifies the effect of oral marketing；
The checking validity module, using Pearson came Chi-square Test method, in confirmatory experiment group user's intercommunion, with User's accounting of experimental group use research product and the user's accounting difference for compareing use research product in user's intercommunion circle It is whether notable.
Further, the data acquisition module is acquired using powercenterETL instruments to data.
Further, the data memory module uses the data warehouse that IBM DB2 are built.
Further, data are handled using QuestCentralForDB2 softwares.
Further, data authentication analysis is carried out using using PC.
Further, customer relationship is contacted using contact cohesion identification model in the contact cohesion identification module Cohesion is identified, shown in the contact cohesion identification model such as formula (one)：
Wherein, diFor festivals or holidays weight, XiFor type weight, WiFor period weight, piFor the duration of call or short message/multimedia message Bar number, n are number of communications.
Further, the space overlap degree identification module uses space overlap degree identification model to the space between user Degree of overlapping is identified, shown in the space overlap degree identification model such as formula (two)：
Wherein, djFor festivals or holidays weight, XjFor type weight, WjFor period weight, TjIt is same to be appeared in the identical period The duration of base station, n are the base station number that user occurs simultaneously.
A kind of operator's industry product Praise effect identification method, there is provided operator's industry product mouth described above Upright stone tablet
Marketing effectiveness identification systems, this method comprise the following steps：
A:Screened to being not suitable for use in Praise effect identification user；
B:Communication bill between user is described using the wide table module of data；
C:Quantum chemical method is carried out to intimate degree between user using contact cohesion identification module；
D:Using checking validity module to meeting necessarily intimately user's accounting difference of degree threshold values in user's relationship cycle Carry out checking validity module.
Further, this method comprises the following steps：
E:Screened to being not suitable for use in Praise effect identification user；
F:Motion track feature between user is described using signaling positional information table；
G:Quantum chemical method is carried out to the space overlap degree between user using space overlap degree identification module；
H:It is poor to the user's accounting for meeting certain space degree of overlapping threshold values in user's relationship cycle using checking validity module Different carry out checking validity.
Beneficial effect：The present invention compared with the prior art, this have the advantage that：
1st, the present invention combines checking validity module, the intimate model of structure relationship cycle and space overlap degree model, compared to biography System method, more accurate science, science objective assessment is made to product Praise effect；
2nd, the inventive method can graduate from old-type opera school according to user's relationship cycle and space-time trace information to product Praise effect Objective assessment is learned, marketing effectiveness qualification result is more accurate；
3rd, the present invention utilizes the existing data of telecommunications, carries out many demonstrations by computer technology, greatly reduces Outgoing call investigation cost and manpower, it is more efficient faster；
If the 4, identifying the Praise significant effect of certain product, it can be the interconnection networking of telecommunication product, find Correct approach, the input in terms of other advertisements is reduced, reduces cost.
Brief description of the drawings
Fig. 1 is the inventive method flow chart of steps.
Fig. 2 is that user associates schematic diagram.
Fig. 3 is the overlapping schematic diagram of user's space.
Embodiment
The present invention is further described below in conjunction with the accompanying drawings.
Embodiment one：
Operator's industry product Praise effect identification system of the present embodiment, including data acquisition module, data are deposited Store up module, data processing module and data identification and analysis module；
Wherein, the data acquisition module is used to obtain related data, the packet for research object and research product The of that month communication bill of research object and signaling position data are included, wherein communication bill data include the duration of call, short message/coloured silk Creed number, communication time period, communication festivals or holidays type and communication type；
The data memory module is used for data storage；
The data processing module is used to handle data storage；
The data authentication analysis module is used to carry out identification and analysis to data, obtains effect identification result.
The data authentication analysis module includes contact cohesion identification module, space overlap degree identification module and conspicuousness Correction verification module；
Wherein, the contact cohesion identification module, based on communication bill data, build user network relation, and to Family relationship cohesion is identified, by the accounting for judging use research product user in strong tie cohesion user group Size identifies the effect of oral marketing；
The space overlap degree identification module, based on signaling position data, each user is portrayed in different time different location Appearance track, the space overlap degree between user is identified, by judging in strong association user colony that use research produces The accounting size of product user identifies the effect of oral marketing；
The checking validity module, using Pearson came Chi-square Test method, in confirmatory experiment group user's intercommunion, with User's accounting of experimental group use research product and the user's accounting difference for compareing use research product in user's intercommunion circle It is whether notable.
The data acquisition module is acquired using powercenterETL instruments to data.
The data memory module uses the data warehouse that IBM DB2 are built.
Data are handled using QuestCentralForDB2 softwares.
Data authentication analysis is carried out using using PC.
Customer relationship contact cohesion is entered using contact cohesion identification model in the contact cohesion identification module Row identification, shown in the contact cohesion identification model such as formula (one)：
Wherein, diFor festivals or holidays weight, XiFor type weight, WiFor period weight, piFor the duration of call or short message/multimedia message Bar number, n are number of communications.
The space overlap degree identification module is entered using space overlap degree identification model to the space overlap degree between user Row identification, shown in the space overlap degree identification model such as formula (two)：
Wherein, djFor festivals or holidays weight, XjFor type weight, WjFor period weight, TjIt is same to be appeared in the identical period The duration of base station, n are the base station number that user occurs simultaneously.
A kind of operator's industry product Praise effect identification method, there is provided operator's industry product mouth described above Upright stone tablet
Marketing effectiveness identification systems, this method comprise the following steps：
A:Screened to being not suitable for use in Praise effect identification user；
B:Communication bill between user is described using the wide table module of data；
C:Quantum chemical method is carried out to intimate degree between user using contact cohesion identification module；
D:Using checking validity module to meeting necessarily intimately user's accounting difference of degree threshold values in user's relationship cycle Carry out checking validity module.
This method comprises the following steps：
E:Screened to being not suitable for use in Praise effect identification user；
F:Motion track feature between user is described using signaling positional information table；
G:Quantum chemical method is carried out to the space overlap degree between user using space overlap degree identification module；
H:It is poor to the user's accounting for meeting certain space degree of overlapping threshold values in user's relationship cycle using checking validity module Different carry out checking validity.
Embodiment two：
On the basis of embodiment one, operator's industry product Praise effect identification system of the present embodiment,
The system includes data acquisition module, data processing module, contact cohesion identification module, the identification of space overlap degree Module and checking validity module；
Data acquisition module gathers related data according to research method mainly for research object.Data acquisition module leads to Cross following steps and realize function：
The product of selected Praise effect identification, and determine the period of research object and observation；
Screen the user of improper progress Praise effect identification；Such as in relationship cycle rete mirabile user product type without Method obtains, and can reject the rete mirabile user in relationship cycle；
Using powercenter ETL instruments to voice ticket, short message ticket and the letter in the research object this month being related to Make position data be extracted, clean, changing and be loaded into IBMDB29.5 databases；
Data processing module carries out geometric precision correction, noise reduction enhancing, Data Fusion to the data of centralized procurement, passes through quantitative solution Map data mining platform of the output with research object and effective evaluation evaluation index information is translated, the information is the accessible word of computer Segment information.Data processing module realizes function by following steps：
For the relationship cycle data of centralized procurement, using calling subscribe and called subscriber as major key, the contacts built between user refer to Number, including calling subscribe, called subscriber, call account phase, festivals or holidays type, voice call duration, short message bar number, multimedia message bar number Deng；
For the user trajectory data of collection, appeared in same time in the customer group of same base based on two two users Key, build the wide table between user, including user A, user B, date, period, base station, sector, residence time etc..
For the user in two tables of data of above-mentioned foundation, associated with the user information table in crm system, identify use The product type that family uses；
Contact cohesion identification module memory storage has cohesion computational methods between user and meets the parent of certain threshold values Density user is contacted cohesion identification module and is realized work(by following steps using user's proportion computing technology of identification product Energy：
For the contacts index between user, useMeter Calculate the cohesion coefficient between two two.
Wherein diFor festivals or holidays weight, XiFor type weight, WiFor period weight, piFor the duration of call or short message/multimedia message bar Number, n is number of communications；
Specific weight parameter using EXCEL softwares sketch out data distribution situation and combine expert interviewing result after set It is as follows：
d：Legal festivals and holidays：D=1.0；Two-day weekend：D=0.9；Working day:D=0.8；
w：The point of morning 7. -9：W=0.8；The point of the morning 9. -12：W=0.7；The point of 12 noon -14：W=0.8；Afternoon 14 Point -18 point：W=0.7；18. -23 point at night：W=0.9；The point of night 23. -7：W=1.0；
x：Voice：X=0.6；Short message：X=0.8；Multimedia message：X=1.0；
The user that cohesion coefficient is more than certain threshold values is filtered out, is calculated in the user group, uses the use of identification product Family accounting.
Space overlap degree identification module memory storage has the computational methods of the space overlap degree between user and meets certain weight The computational methods of the accounting of the identification product are used in the user of folded degree, space overlap degree identification module is realized by following steps Function：
For the contacts index between user, use Calculate the cohesion coefficient between two two.
Wherein, djFor festivals or holidays weight, XjFor type weight, WjFor period weight, TjIt is same to be appeared in the identical period The duration of base station, n are the base station number that user occurs simultaneously.Specific weight parameter sketches out the distribution of data using EXCEL softwares Setting is as follows after situation and combination expert interviewing result：
d：Legal festivals and holidays：D=0.6；Two-day weekend：D=0.8；Working day：D=1.0；
w：The point of morning 7. -9：W=0.8；The point of the morning 9. -12：W=0.7；The point of 12 noon -14：W=0.8；Afternoon 14 Point -18 point：W=0.7；18. -23 point at night：W=0.9；The point of night 23. -7：W=1.0；
x：CBD：X=0.8；Station：X=0.3；Scenic spot：X=0.5；Cell x=1.0；Campus x=0.7；Other x= 0.8；
The user that cohesion coefficient is more than certain threshold values is filtered out, is calculated in the user group, uses the use of identification product Family accounting.
Checking validity module is to use user's accounting of like products with compareing with experimental group in confirmatory experiment group user It is whether notable using user's accounting difference of experimental group like products in user.Checking validity module is realized by following steps Function：
Establish null hypothesis and selected hypothesis：
H0：Using the accounting of experimental group product user with being used in control group user's relationship cycle in experimental group user's relationship cycle The accounting of experimental group product user is equal.
H1:Using the accounting of experimental group product user with being used in control group user's relationship cycle in experimental group user's relationship cycle The accounting of experimental group product user is unequal.
In spss statistics softwares in variable view, definition meets the form that independent sample T is examined, and in number The ratio for using experimental products user according to being inputted in view in experimental group and control group；
In the analysis menu of spss statistics softwares, independent sample T is selected to examine in relatively average；
Input variable in test variable " ratio for using experimental products user ", packet variable input " whether experimental group " become Amount, level of confidence selection 95%；
Embodiment three：
Based on embodiment two, a kind of operator's industry product Praise effect identification method of the present embodiment, reference picture 1, carried out by following steps：
The first step selectes experimental group and contrast group objects
Selected first experimental group and contrast group objects and scale.
, one transitional period of user's number of changing presence, then can not be just in view of research object (user) if the newly-increased time is too short The really behavior of reflection user, the accuracy assessed so as to impact effect；
Therefore the research object increased newly using the month before last is used as experimental group, the domestic consumer that the month before last increases newly as comparison group, its In the data source of of that month relationship cycle and positional information as research.
Second step obtains evaluation index information figure layer
By the understanding to business and the situation of system data, powercenterETL instruments are utilized at charging and network optimization Voice ticket, short message ticket and the signaling position data of that month to research object cleaned, is changed and is loaded into IBM DB2 and takes In the data warehouse built, and geometric precision correction, noise reduction enhancing, Data Fusion are carried out using QuestCentralForDB2 softwares Afterwards, exported by DB2 SQL multilist corresponding technologies quantitative interpretations with research object and effective evaluation evaluation index information Figure layer.
It is as shown in table 1 including ticket relationship cycle tablet menu and signaling position movement table, ticket relationship cycle tablet menu：
Table 1
Because user is being communicated with base station all the time, data volume is extremely huge, therefore by dividing the time Section, reaches reduction data scale, so as to lift the purpose for calculating data efficiency.
Time slice is segmented in DB2 database using the CASE WHEN grammers of SQL statement to the time, output 1, The point of morning 7. -9.2, the point of the morning 9. -12.3, the point of 12 noon -14.4, afternoon 14. -18 point.5,18. -23 points of evening.6, The point of night idle 23. -7
Signaling position unusual fluctuation table tool cuticle topography is as shown in table 2 below:
Table 2
Relationship cycle recognition effect is analyzed and evaluated using DB2 database and sql like language for 3rd step
When integrating voice call number between user and user, the duration of call, call by the table corresponding technology of sql like language Section and short message bar number, short message period, build the relational network between user and user；
Then the ticket behaviors such as the voice of SNA algorithms and user, short message, multimedia message are combined, build social networks circle, And according to talk times, the duration of call, talk period, call number of days, calculate contact cohesion between two two communication targets, Mathematical modeling is as follows：
Wherein diFor festivals or holidays weight, XiFor type weight, WiFor period weight, piFor the duration of call or short message/multimedia message bar Number, n is number of communications.Specific weight parameter sketches out the distribution situation of data using EXCEL softwares and combines expert interviewing knot Set after fruit as follows：
d：Legal festivals and holidays：D=1.0；Two-day weekend：D=0.9；Working day:D=0.8；
w：The point of morning 7. -9：W=0.8；The point of the morning 9. -12：W=0.7；The point of 12 noon -14：W=0.8；Afternoon 14 Point -18 point：W=0.7；18. -23 point at night：W=0.9；The point of night 23. -7：W=1.0；
x：Voice：X=0.6；Short message：X=0.8；Multimedia message：X=1.0；
And the strength of association between network of personal connections is identified, reflected by judging the accounting size of strong 2 experimental group user Determine the effect of oral marketing, its cohesion is higher more than the instruction manual of average level for contact cohesion (CC).
Except by mobile phone communication propagate in addition to, may also be propagated between each other by aspectant mouth ear, for example, neighbours it Between go out to run into the possibility of propagation will be produced when chatting, and the generation not communicated with each other between this user, according to Above-mentioned calling exponential model, which is difficult to portray Praise, have propagated, but can be sentenced according to the stacked degree of user's motion track The intimate degree broken between them, occurs if appearing at identical base station location and such case in both same time Frequency reach certain threshold values, then may determine that between them that there is intimate contact, so as to can by what is taught orally face-to-face Energy property is bigger.
So this research passes through the table corresponding technology of sql like language, integration user cutting in different base station in different time points Change path, the information such as extraction user mark, time, position, residence time, and by calculating between user and user when identical Between section appear in the time length of same base to describe the space cohesion between disseminator, by judging strong cohesion user The accounting size for testing group objects in colony identifies the effect of oral marketing.Specific formula for calculation is as follows：
Wherein djFor festivals or holidays weight, XjFor type weight, WjFor period weight, TjTo be appeared in together in the identical period
The duration of one base station, n are the base station number that user occurs simultaneously.Specific weight parameter is delineated using EXCEL softwares
Setting is as follows after going out the distribution situation of data and combining expert interviewing result：
d：Legal festivals and holidays：D=0.6；Two-day weekend：D=0.8；Working day：D=1.0；
w：The point of morning 7. -9：W=0.8；The point of the morning 9. -12：W=0.7；The point of 12 noon -14：W=0.8；Afternoon 14 Point -18 point：W=0.7；18. -23 point at night：W=0.9；The point of night 23. -7：W=1.0；
x：CBD：X=0.8；Station：X=0.3；Scenic spot：X=0.5；Cell x=1.0；Campus x=0.7；Other x= 0.8；
And the space overlap degree between user is identified, by judging experimental group user in strong association user colony Accounting size identifies the effect of oral marketing, and it is higher that space overlap degree exceedes the instruction manual of average level its cohesion.
4th step carries out significant effect inspection using the Pearson came Chi-square Test method of spssstatistics softwares
This step utilizes the Pearson came Chi-square Test method of spssstatistics softwares, and confirmatory experiment group user intimately hands over Into, the production identical with using experimental group in control user user's intercommunion circle of user's accounting of like products is used with experimental group Whether user's accounting difference of product is notable.
First, set null hypothesis and selected hypothesis
H0：Using the accounting of experimental group product user with being used in control group user's relationship cycle in experimental group user's relationship cycle The accounting of experimental group product user is equal.
H1:Using the accounting of experimental group product user with being used in control group user's relationship cycle in experimental group user's relationship cycle The accounting of experimental group product user is unequal.
Secondly, in spss statistics softwares in variable view, definition meets the form that independent sample T is examined, and Inputted in Data View in experimental group and control group and use the ratio of experimental products user；
Then, in the analysis menu of spss statistics softwares, independent sample T is selected to examine in relatively average；
Finally, whether input variable " ratio for using experimental products user " in test variable, packet variable input " test Group " variable, level of confidence selection 95%；
Examined by independent sample T, if double tail significance P=0.000<0.05, refuse null hypothesis H0, receive standby Select and assume H1, it was demonstrated that the Praise significant effect that experimental group is propagated by relationship cycle.
Ibid, the Pearson came Chi-square Test method of spssstatistics softwares, confirmatory experiment group user's space weight are utilized In the high user of folded degree, with experimental group using making in user's accounting of like products user high with compareing user's space degree of overlapping It is whether notable with user's accounting difference of experimental group like products.
First, set null hypothesis and selected hypothesis
H0：Using the accounting of experimental group product user with being used in control group user's relationship cycle in experimental group user's relationship cycle The accounting of experimental group product user is equal.
H1:Using the accounting of experimental group product user with being used in control group user's relationship cycle in experimental group user's relationship cycle The accounting of experimental group product user is unequal.
Secondly, in spss statistics softwares in variable view, definition meets to intersect tableau format, and in data Inputted in view in experimental group and control group and use the ratio of experimental products user
Again, in data menu, selection weighting case, frequency variable selection " overlapping user number " variable
Then, in the analysis menu of spss statistics softwares, the selection intersection form in descriptive statistics
Finally, " whether experimental subjects " variable is chosen in being expert at, identifies experimental group and validation group, selection is " empty in row Between overlapping user whether identify product user ", identify in space overlap user using identification product user quantity, Chi is chosen in statistics, remaining parameter presses system default
By Pearson came Chi-square Test, if double tail significance P=0.000<0.05, refuse null hypothesis H0, connect
By alternative hypothesis H1, it was demonstrated that the Praise significant effect that experimental group passes through spatial.
The description of method flow diagram in Fig. 1：
S01：Required according to research and the understanding to telecommunication service, the user for selecting the month before last new development are used as research pair As wherein the user for handling research product handles other products as a control group as experimental group.
S02：Using powerCenter ETL instruments, extracted at charging and network optimization, clean research object this month voice words List, short message ticket, multimedia message ticket and signaling position data, and be loaded into the data warehouse built by DB2.And utilize QuestCentralForDB2 softwares and sql like language carry out the processing such as geometric precision correction, noise reduction enhancing, data fusion to data.
S03：Using sql like language, the user bill data come in loading are handled, build a user and user it Between the wide table of contacts index, comprising calling number, called number, the duration of call is talk period, the call date, short message bar number, short Believe the fields such as period, short message date, multimedia message bar number, multimedia message period, multimedia message date.
S04:Using sql like language, the subscriber signaling data come in loading are handled, build a user and user it Between location track wide table, include the fields such as user A, user B, base station, date, period, residence time.
S05：The ticket behaviors such as the voice of SNA algorithms and user, short message, multimedia message are combined, build social networks circle, And according to talk times, the duration of call, talk period, by number of days, calculate contact cohesion between two two communication targets.
S06:By the table corresponding technology of sql like language, integrate user in different time points on the switching road of different base station The information such as footpath, extraction user mark, time, position, residence time, and by calculating between user and user in same time period The time length of same base is appeared in describe the space cohesion between disseminator, by judging strong cohesion user group The accounting size of middle experiment group objects identifies the effect of oral marketing；
S07:Utilize the Pearson came Chi-square Test method of spssstatistics softwares, confirmatory experiment group user's intercommunion In, experimental group like products is used in user user's intercommunion circle with compareing using user's accounting of like products with experimental group User's accounting difference it is whether notable
S08：With S07 modules, using the Pearson came Chi-square Test method of spssstatistics softwares, confirmatory experiment group is used Family space overlap degree>In 8 user, user's accounting of like products is used with compareing user's space degree of overlapping with experimental group>8 It is whether notable using user's accounting difference of experimental group like products in user.
The point of black represents the object using experimental products in Fig. 2, and the point of grey represents the object that experimental products are not used, Connection is established by modes such as voice, short message, multimedia messages between them, so as to build social networks collection of illustrative plates.Pass through experiment with computing group The user that experimental products are used in user's ratio of experimental products and the relationship cycle of control group objects is used in the relationship cycle of object Ratio, to identify the effect of Praise；Such as wherein experimental group user one, its relationship cycle include 1 experimental group user and 4 Individual control group user, then the use of user's ratio of experimental products is 2/6 in the relationship cycle of experiment group objects；Wherein control group user One, its relationship cycle includes 2 experimental group users and 1 control group user, then compares in the relationship cycle of group objects using experiment User's ratio of product is 2/4, the ratio used higher than experimental products in experimental group user's relationship cycle；
Represent appearing in the user of same base station at the same time in Fig. 3, including the use using experimental products Family and the non-user using experimental products, by user's ratio that experimental products are used in the space tracking of experiment with computing group objects With control group objects space tracking in use experimental products user's ratio, to identify the effect of Praise.
Example IV：
A kind of operator's industry product Praise effect identification method of the present embodiment, based on embodiment three；
Using the Ifree products that China Telecom newly releases as research object, illustrate that the authentication method of Praise effect exists The application of telecommunications industry.
The first step selectes experimental group and the contrast group objects present invention using 09 month 2016 newly-increased Ifree card user as in fact Test group, newly-increased domestic consumer group as a comparison, the relationship cycle in its after 1 month (i.e. in November, 2016) and positional information conduct The data source of research.Study sample and data are totally as shown in table 3 below：
Table 3
Because the user that part of in September, 2016 increases newly can not possibly be broadcast to the user to network before, and his net can not be judged User whether Ifree cards, so only research had call behavior to the present invention with the Home Network storage user to be networked before in September, 2016 0.7 ten thousand experimental groups and 23.8 ten thousand contrast groups.
Second step obtains evaluation index information figure layer
Using powercenterETL instruments to the of that month voice ticket of research object, short message ticket at charging and network optimization And signaling position data is cleaned, changes and be loaded into the data warehouse that IBM DB2 are built, and utilize After QuestCentralForDB2 softwares carry out geometric precision correction, noise reduction enhancing, Data Fusion, band is exported by quantitative interpretations The figure layer of research object and effective evaluation evaluation index information.
It is as shown in table 4 below respectively including ticket relationship cycle tablet menu and signaling position movement table：
Table 4
Counted using SQL, the accumulative development day wing user 88.5 ten thousand of 09 month 2016 the whole province, wherein Ifree card users 3.0 Ten thousand, domestic consumer 85.5 ten thousand, in the movement tickets of in November, 2016 common property life 4081.6 ten thousand, calling peer user number up to 379.1 ten thousand, Wherein Ifree card users produce 23.6 tickets per family, and relationship cycle number is 2.5 per family, and domestic consumer produces 46.9 per family Individual ticket, relationship cycle number is 4.3 per family.User after being networked due to September part is networked before can not possibly being broadcast to September part User, and can not judge his network users whether Ifree cards, so this problem only studies the Home Network with being networked before September part Storage user had 0.7 ten thousand Ifree card users and 23.8 ten thousand domestic consumers of call behavior, this certain customers and 85.5 ten thousand Home Networks This province movement storage generates call behavior, Ifree cards 2.9 relationship cycles per family, domestic consumer's 3.5 relationship cycles per family, such as Shown in table 5.
Table 5
Because user is being communicated with base station all the time, data volume is extremely huge, therefore by dividing the time Section, reach reduction data scale, lifting calculates the purpose of data, and tool cuticle topography is as shown in table 6 below:
Table 6
Due to calculate 09 month 2016 88.4 newly-increased general-purpose families with the mobile subscriber of the whole province more than 2,200 ten thousand in 6.1 ten thousand base stations The space density of sector each period, operand has exceeded the computing capability of platform, therefore this model is only using some regional A as research Object.The A of in September, 2016 the areas accumulative development day wing 23.7 ten thousand, wherein IFREE card users 0.5 ten thousand, domestic consumer 23.1 ten thousand, In regional 4.9 ten thousand sectors of visiting A in November, up to 1.28 hundred million times with this certain customers space overlap occurs for visiting record number User have 304.0 ten thousand, base station visiting number occurs 32.9 hundred million times, as shown in table 7；
Table 7
Relationship cycle recognition effect is analyzed and evaluated using DB2 database and sql like language for 3rd step
When integrating voice call number between user and user, the duration of call, call by the table corresponding technology of sql like language Section and short message bar number, short message period, build the relational network between user and user.Then by SNA algorithms and the language of user The ticket behavior such as sound, short message, multimedia message combines, build social networks circle, and according to talk times, the duration of call, call when Section, by number of days, calculate contact cohesion between two two communication targets.
And the strength of association between network of personal connections is identified, by judging IFREE user in strong association user colony Accounting size identifies the effect of oral marketing, and its cohesion of the contact instruction manual of cohesion (CC) more than 3 is higher, such as table Shown in 8, following data are obtained by statistics.
Table 8
Except by mobile phone communication propagate in addition to, may also be propagated between each other by aspectant mouth ear, for example, neighbours it Between go out to run into the possibility of propagation will be produced when chatting, and the generation not communicated with each other between this user, according to Above-mentioned calling exponential model, which is difficult to portray Praise, have propagated, but can be sentenced according to the stacked degree of user's motion track The intimate degree broken between them, occurs if appearing at identical base station location and such case in both same time Frequency reach certain threshold values, then may determine that between them that there is intimate contact, so as to can by what is taught orally face-to-face Energy property is bigger.
So this research by the table corresponding technology of sql like language, integrate user in different time points in different base station The information such as toggle path, extraction user mark, time, position, residence time, and by calculating between user and user identical Period appears in the time length of same base to describe the space cohesion between disseminator, by judging that strong cohesion is used The accounting size of ifree user identifies the effect of oral marketing in the colony of family.
And the space overlap degree between user is identified, by judging IFREE user in strong association user colony Accounting size identifies the effect of oral marketing, the higher of space overlap degree its degree of overlapping of the instruction manual more than 8
By calculating, following data are obtained, as shown in table 9：
Table 9
4th step carries out significant effect inspection using the Pearson came Chi-square Test method of spssstatistics softwares
This step utilizes the Pearson came Chi-square Test method of spssstatistics softwares, and checking IFREE user intimately hands over Into, whether ifree user's accounting and IFREE user's accounting difference in domestic consumer user intercommunion circle are notable.
The contact relationship cycle user of cohesion (CC) more than 3 is calculated for above-mentioned, experiment with computing set product user's accounts for Than as shown in table 10.
Table 10
Verified by calling exponential model, the month before last ifree user effectively studied that newly networks is 6243 families, of that month Index is associated in relationship cycle>3 Home Network mobile call object has 9156 families, and wherein IFREE user is 1210, accounting 13.9%. It is 209476 families that in September, 2016, which newly networks and effectively studies number of users domestic consumer, and index is associated in the relationship cycle in this month>3 Home Network mobile call object has 246977 families, and wherein IFREE user is 793, and accounting is only 0.3%, is handed over less than experimental group user The ratio of ifree cards into circle, whether difference is notable, and we are verified by independent sample T.Comprise the following steps that：
First, set null hypothesis and selected hypothesis
H0：Using the accounting of experimental group product user with making in control group user's relationship cycle in experimental group user's relationship cycle
Accounting with experimental group product user is equal.
H1:Using the accounting of experimental group product user with making in control group user's relationship cycle in experimental group user's relationship cycle
Accounting with experimental group product user is unequal.
Secondly, in spss statistics softwares in variable view, definition meets the form that independent sample T is examined, and Inputted in Data View in experimental group and control group and use the ratio of experimental products user.Variable view is as shown in table 11 below：
Table 11
Data View sample is as shown in table 12 below：
pd_inst_ID IFREE_F PEER_IFREE_CNT PEER_CNT IFREE_PRNC
1.01227E+11 N 0 2 0
1.01247E+11 N 0 1 0
1.0126E+11 N 0 1 0
1.01266E+11 N 0 1 0
Table 12
Then, in the analysis menu of spss statistics softwares, independent sample T is selected to examine in relatively average
Finally, whether input variable " ratio for using experimental products user " in test variable, packet variable input " test Group " variable, level of confidence selection 95%, as shown in table 13.
Table 13
It is as follows to click on the output result obtained after determining, shown in table 14, table 15：
Group statistics
Table 14
Independent sample is examined
Table 15
Examined by independent sample T, double tail significance P=0.000<0.05, refuse null hypothesis H0, receive alternative vacation If H1, it was demonstrated that the Praise significant effect that experimental group passes through man-machine propagation.
Ibid, the Pearson came Chi-square Test method of spssstatistics softwares, confirmatory experiment group user's space weight are utilized Folded degree>IFREE card users accounting is with compareing user's space degree of overlapping in 8 user>IFREE card users accounting is poor in 8 user It is different whether notable.
For user of the Spatial Overlap degree more than 8, the accounting using experimental group product user is calculated, as shown in table 16；
Table 16
Verified by space density model, in September, the 2016 experimental group user effectively studied that newly networks is 2653 families, 11 Spatial Overlap degree in base station location signaling month in and month out>8 mobile subscriber has 674937 families, and the number of wherein IFREE cards is 27473, accounting 4.1%.The control group user effectively studied that newly networks in November, 2016 is 83116 families, in 11 base station month in and month out Spatial Overlap degree in position signaling>8 mobile subscriber has 1965030 families, and wherein IFREE numbers are 32341, accounting 1.8%, Less than the ratio of IFREE in experimental group user's relationship cycle.Whether difference is notable, and we are tested by Pearson came Chi-square Test Card.Comprise the following steps that：
First, set null hypothesis and selected hypothesis
H0：Using the accounting of experimental group product user with being used in control group user's relationship cycle in experimental group user's relationship cycle The accounting of experimental group product user is equal.
H1:Using the accounting of experimental group product user with being used in control group user's relationship cycle in experimental group user's relationship cycle The accounting of experimental group product user is unequal.
Secondly, in spss statistics softwares in variable view, definition meets to intersect tableau format, and in data Inputted in view in experimental group and control group and use the ratio of experimental products user.
Variable view is as shown in table 17 below：
Title Type Value
Research object whether IFREE user Character string { N, common research object } ...
Space overlap user whether IFREE user Character string { N, domestic consumer } ...
Overlapping user number Numerical value Nothing
Table 17
Data View is as shown in table 18 below：
Research object whether IFREE user Space overlap user whether IFREE user Overlapping user number
Y Y 27473
Y N 647464
N Y 35341
N N 1929689
Table 18
Again, in data menu, selection weighting case, frequency variable selection " overlapping user number " variable, such as the institute of table 19 Show；
Table 19
Then, in the analysis menu of spss statistics softwares, the selection intersection form in descriptive statistics
Finally, " whether experimental subjects " variable is chosen in being expert at, identifies experimental group and validation group, selection is " empty in row Between overlapping user whether identify product user ", identify in space overlap user using identification product user quantity, Chi is chosen in statistics, remaining parameter presses system default, as shown in table 20；
Table 20
Click on after determining, export following result, as shown in table 21：
Dealing by single is made a summary
Research object whether IFREE user's * space overlaps user whether IFREE user's crosstab
Chi-square Test
A.0 expected count that individual cell (0.0%) has is less than 5.Minimum expected is counted as 16059.10.
B. it is only that 2x2 forms calculate
Table 21
By Pearson came Chi-square Test, double tail significance P=0.000<0.05, refuse null hypothesis H0, receive alternative Assuming that H1, it was demonstrated that the remarkable anthrochorous Praise significant effect of Ifree cartoons.The hardware environment of implementation steps is as follows：
Function Software Hardware
ETL instruments PowerCenter 9.6 Ibmx86 servers 2
Database IBM DB2 9.5 Ibmx86 servers 16
Analytic statistics software spss statistics 22.0 PC
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 (10)

  1. A kind of 1. operator's industry product Praise effect identification system, it is characterised in that：Including data acquisition module, data Memory module, data processing module and data identification and analysis module；
    Wherein, the data acquisition module is used to obtain related data for research object and research product, and the data include grinding Study carefully the of that month communication bill of object and signaling position data, wherein communication bill data include the duration of call, short message/multimedia message bar Number, communication time period, communication festivals or holidays type and communication type；
    The data memory module is used for data storage；
    The data processing module is used to handle data storage；
    The data authentication analysis module is used to carry out identification and analysis to data, obtains effect identification result.
  2. 2. operator's industry product Praise effect identification system according to claim 1, it is characterised in that：The number Include contact cohesion identification module, space overlap degree identification module and checking validity module according to identification and analysis module；
    Wherein, the contact cohesion identification module, based on communication bill data, user network relation is built, and user is closed System's contact cohesion is identified, by the accounting size for judging use research product user in strong tie cohesion user group To identify the effect of oral marketing；
    The space overlap degree identification module, based on signaling position data, portray each user going out in different time different location Existing track, the space overlap degree between user is identified, by judging that use research product is used in strong association user colony The accounting size at family identifies the effect of oral marketing；
    The checking validity module, using Pearson came Chi-square Test method, in confirmatory experiment group user's intercommunion, with experiment Group use research product user's accounting with compare use research product in user's intercommunion circle user's accounting difference whether Significantly.
  3. 3. operator's industry product Praise effect identification system according to claim 1, it is characterised in that：The number Data are acquired using powercenterETL instruments according to acquisition module.
  4. 4. operator's industry product Praise effect identification system according to claim 1, it is characterised in that：The number The data warehouse built according to memory module using IBM DB2.
  5. 5. operator's industry product Praise effect identification system according to claim 1, it is characterised in that：Using QuestCentralForDB2 softwares are handled data.
  6. 6. according to operator's industry product Praise effect identification system described in claim 1, it is characterised in that：Using use PC carries out data authentication analysis.
  7. 7. according to operator's industry product Praise effect identification system described in claim 2, it is characterised in that：The contact Customer relationship contact cohesion is identified using contact cohesion identification model in cohesion identification module, the contact parent Shown in density identification model such as formula (one)：
    Wherein, diFor festivals or holidays weight, XiFor type weight, WiFor period weight, piFor the duration of call or short message/multimedia message bar number, n For number of communications.
  8. 8. according to operator's industry product Praise effect identification system described in claim 2, it is characterised in that：The space The space overlap degree between user is identified using space overlap degree identification model for degree of overlapping identification module, the space weight Shown in folded degree identification model such as formula (two)：
    Wherein, djFor festivals or holidays weight, XjFor type weight, WjFor period weight, TjTo appear in same base station in the identical period Duration, n be user simultaneously occur base station number.
  9. A kind of 9. operator's industry product Praise effect identification method, it is characterised in that：Methods described provides a kind of as weighed Profit requires operator's industry product Praise effect identification system described in any one claim, this method bag in 1 to 8 Include following steps：
    A:Screened to being not suitable for use in Praise effect identification user；
    B:Communication bill between user is described using the wide table module of data；
    C:Quantum chemical method is carried out to intimate degree between user using contact cohesion identification module；
    D:Using checking validity module to meeting that user's accounting difference of certain intimate degree threshold values is carried out in user's relationship cycle Checking validity module.
  10. 10. operator's industry product Praise effect identification method according to claim 9, it is characterised in that：The party Method comprises the following steps：
    E:Screened to being not suitable for use in Praise effect identification user；
    F:Motion track feature between user is described using signaling positional information table；
    G:Quantum chemical method is carried out to the space overlap degree between user using space overlap degree identification module；
    H:Using checking validity module to meeting that user's accounting difference of certain space degree of overlapping threshold values is entered in user's relationship cycle Row checking validity.
CN201710739922.8A 2016-08-18 2017-08-18 System and method for identifying public praise marketing effect of operator industry product Active CN107527240B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610679961.9A CN106296300A (en) 2016-08-18 2016-08-18 A kind of authentication method of telecommunications industry mobile product Praise effect
CN2016106799619 2016-08-18

Publications (2)

Publication Number Publication Date
CN107527240A true CN107527240A (en) 2017-12-29
CN107527240B CN107527240B (en) 2021-02-19

Family

ID=57678709

Family Applications (2)

Application Number Title Priority Date Filing Date
CN201610679961.9A Pending CN106296300A (en) 2016-08-18 2016-08-18 A kind of authentication method of telecommunications industry mobile product Praise effect
CN201710739922.8A Active CN107527240B (en) 2016-08-18 2017-08-18 System and method for identifying public praise marketing effect of operator industry product

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN201610679961.9A Pending CN106296300A (en) 2016-08-18 2016-08-18 A kind of authentication method of telecommunications industry mobile product Praise effect

Country Status (1)

Country Link
CN (2) CN106296300A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108615198A (en) * 2018-04-26 2018-10-02 北京小米移动软件有限公司 Methods of exhibiting, device and the storage medium that social networking application releases news
CN110544114A (en) * 2019-08-16 2019-12-06 北京市天元网络技术股份有限公司 Method and device for identifying and rating user group aiming at marketing preference
CN111917574A (en) * 2020-07-21 2020-11-10 上海阿尔卡特网络支援系统有限公司 Social network topology model and construction method thereof, user confidence degree and intimacy degree calculation method and telecommunication fraud intelligent interception system
CN112446541A (en) * 2020-11-26 2021-03-05 上海浦东发展银行股份有限公司 Fusion classification model establishing method, marketing conversion rate gain prediction method and system

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7106851B2 (en) * 2017-12-12 2022-07-27 富士フイルムビジネスイノベーション株式会社 Information processing device and program
CN110312202B (en) * 2018-03-23 2021-11-23 中兴通讯股份有限公司 User searching method, device, system and storage medium
CN110162683B (en) * 2019-04-16 2023-08-29 平安科技（深圳）有限公司 Dynamic early warning method, device, computer equipment and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102083010A (en) * 2009-11-26 2011-06-01 中国移动通信集团公司 Method and equipment for screening user information
CN103914781A (en) * 2013-01-09 2014-07-09 索尼公司 Information processing apparatus, information processing method, program and terminal apparatus
CN105260931A (en) * 2015-10-10 2016-01-20 苏州工业园区凌志软件股份有限公司 Financial service platform system based on MOT module

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102083010A (en) * 2009-11-26 2011-06-01 中国移动通信集团公司 Method and equipment for screening user information
CN103914781A (en) * 2013-01-09 2014-07-09 索尼公司 Information processing apparatus, information processing method, program and terminal apparatus
CN105260931A (en) * 2015-10-10 2016-01-20 苏州工业园区凌志软件股份有限公司 Financial service platform system based on MOT module

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108615198A (en) * 2018-04-26 2018-10-02 北京小米移动软件有限公司 Methods of exhibiting, device and the storage medium that social networking application releases news
CN110544114A (en) * 2019-08-16 2019-12-06 北京市天元网络技术股份有限公司 Method and device for identifying and rating user group aiming at marketing preference
CN111917574A (en) * 2020-07-21 2020-11-10 上海阿尔卡特网络支援系统有限公司 Social network topology model and construction method thereof, user confidence degree and intimacy degree calculation method and telecommunication fraud intelligent interception system
CN112446541A (en) * 2020-11-26 2021-03-05 上海浦东发展银行股份有限公司 Fusion classification model establishing method, marketing conversion rate gain prediction method and system

Also Published As

Publication number Publication date
CN106296300A (en) 2017-01-04
CN107527240B (en) 2021-02-19

Similar Documents

Publication Publication Date Title
CN107527240A (en) A kind of operator&#39;s industry product Praise effect identification system and method
CN102591854B (en) For advertisement filtering system and the filter method thereof of text feature
Umayaparvathi et al. A survey on customer churn prediction in telecom industry: Datasets, methods and metrics
Bose et al. Exploring business opportunities from mobile services data of customers: An inter-cluster analysis approach
CN105117422A (en) Intelligent social network recommender system
Al-Zuabi et al. Predicting customer’s gender and age depending on mobile phone data
CN105869012A (en) Marketing system based on or including shopper credit system and cellphone contact list
Blumenstock et al. Who’s calling? Demographics of mobile phone use in Rwanda
CN104915397A (en) Method and device for predicting microblog propagation tendencies
CN111274338B (en) Pre-outbound user identification method based on mobile big data
CN102081774A (en) Card-raising identification method and system
Globa et al. Ontology model of telecom operator big data
Guo et al. GroupMe: Supporting group formation with mobile sensing and social graph mining
WO2022057108A1 (en) Federated-learning-based personal qualification evaluation method, apparatus and system, and storage medium
Hu et al. BTG: A Bridge to Graph machine learning in telecommunications fraud detection
Manley et al. New forms of data for understanding urban activity in developing countries
CN110197426A (en) A kind of method for building up of credit scoring model, device and readable storage medium storing program for executing
CN108776857A (en) NPS short messages method of investigation and study, system, computer equipment and storage medium
WO2024001102A1 (en) Method and apparatus for intelligently identifying family circle in communication industry, and device
CN111062422A (en) Method and device for systematic identification of road loan
Liu et al. Female employment data analysis based on decision tree algorithm and association rule analysis method
Khedra et al. Social network analysis through big data platform review
CN114254161A (en) Graph database-based social network platform user behavior analysis method and system
Yihua Vip customer segmentation based on data mining in mobile-communications industry
Khan et al. Correlation between Triadic Closure and Homophily Formed over Location‐Based Social Networks

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