CN102118706A - Mobile phone advertising method based on subdivision of mobile phone advertisement users - Google Patents
Mobile phone advertising method based on subdivision of mobile phone advertisement users Download PDFInfo
- Publication number
- CN102118706A CN102118706A CN2010106035237A CN201010603523A CN102118706A CN 102118706 A CN102118706 A CN 102118706A CN 2010106035237 A CN2010106035237 A CN 2010106035237A CN 201010603523 A CN201010603523 A CN 201010603523A CN 102118706 A CN102118706 A CN 102118706A
- Authority
- CN
- China
- Prior art keywords
- mobile phone
- data
- user
- users
- consumption
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Telephonic Communication Services (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a mobile phone advertising method based on the subdivision of mobile phone advertisement users, belonging to the field of mobile phone advertisements. The method comprises the following steps: 1) extracting sample user data from a user database by using a mobile phone advertising server; 2) acquiring consumption characteristic data information of sample users; 3) according to the consumption characteristics, carrying out cluster analysis on the sample user data to obtain consumption characteristic-subdivided groups having different consumption characteristics; 4) based on the consumption characteristic-subdivided groups, establishing a telecommunication characteristic-derived subdivision class discriminative model; 5) using the discriminative model to mark each user in the user database with a corresponding consumption characteristic-subdivided group class label; and 6) selecting the users by a mobile phone advertiser in accordance with the subdivided class labels to perform advertising operation. The invention implements the multi-dimensional and comprehensive subdivision of the users, so that the subdivision of the mobile phone advertisement users is more accurate and scientific, and the mobile phone advertiser can select the users and perform the advertising operation more conveniently.
Description
Technical field
The present invention relates to a kind of mobile phone advertisement put-on method based on the mobile phone advertisement subscriber segmentation.Belong to mobile phone advertisement, telecommunication service, data mining and analysis field.
Background technology
When mobile phone becomes very universal, when becoming in everybody life an indispensable part, based on the information service of mobile phone also very rich and varied.The variation very of its form of expression has note, multimedia message, CRBT, video or the like.The instrument that mobile phone is not only a conversation and is linked up more by numerous service innovations, becomes everybody and obtains one of main source of bulk information.Especially mobile phone newspaper, multimedia message newspaper have facilitated everybody to obtain the custom of information by mobile phone in recent years fast development greatly.To such an extent as to scream in the industry: mobile phone has become five medium arranged side by side with newspaper, broadcasting, TV, this four media giant of the Internet.
Along with determining of mobile media attribute, the mobile phone advertisement industry also naturally is born and is flourish.2006, U.S.'s admob founding of the company of being absorbed in the mobile phone advertisement operation was in California, USA.In November, 2009, the Internet giant Google announces with 7.5 hundred million dollars of purchase admob.In January, 2010, famous Apple announces to purchase the Quattro Wireless of another mobile phone advertisement specialized company of family with 2.75 hundred million U.S. dollars.Very fast, domestic the imitator appearred, include a meter net, and posture is wireless etc.More than, be that the industry strength that has nothing to do with mobile operator is at the independent operation mobile phone advertisement.In countries such as Japan and Korea S, mobile operator has played the part of a positive role therein.Their above relatively these mobile phone advertisements network operator has sizable advantage of customer-side.And simultaneously, throw in technology, cellphone subscriber's analytical technology and the difference of beginning along with each emphasis of these different participants in the industry about mobile phone, and progress.
Because mobile phone is people's medium one by one,, cellphone subscriber's analysis and interruption-free mechanism is become extremely important so no matter be the sort of mode.Because the individual media attribute of mobile phone has determined that mobile phone advertisement can be more accurate, meet advertiser's input requirement more.But at present, normal for cellphone subscriber's analysis just from telecommunication service data and consensus data, provide the input reference to the advertiser.Such as the mobile phone tail number is 2345 user, has opened the Fetion service, has opened the monthly payment that surfs the web, and the mobile phone terminal model is Nokia or the like, and this is the telecommunications characteristic; This user is the male sex in other words, and the age is 35 years old, and this is demographic feature.In this time, the advertiser can only face these very basic, very specialized user data, carries out advertisement putting.Concrete this personal consumption ability how, the tendency of consumption like giving a discount commodity or like name brands, whether like social, be important life style attribute such as unmarried or married, in fact for the advertiser, have more and be worth and the actual meaning of throwing in.But present in the industry, excavate and subdivide technology for user's data, still only rest on the user is hived off based on telecommunication service data and consensus data.
Summary of the invention
From top analysis, carry out advertisement putting based on user's life style feature, more meaningful concerning the advertiser in fact.Especially consider actual industrial environment, demand side is to various advertiser from different industries the time, and is all the more so.But how according to user's telecommunications feature, demographic feature, drawing user's consumption feature, is to have much challenging problem in the industry.
The invention provides a mobile phone advertisement put-on method that comprises the mobile phone advertisement subscriber segmentation of life style feature, having solved this problem to a certain extent. the innovative point of this technology is: the telecommunications feature and the demographic feature that not only combine the mobile phone advertisement user, also added the personal consumption feature, application data statistics and the method excavated are carried out various dimensions to the user then, comprehensively segmentation, only remedied in traditional subdivide technology according to telecom consumption feature or the limitation only segmented according to demographic feature, make that the mobile phone advertisement subscriber segmentation is more accurate, science more, convenient mobile phone advertisement master is to user's selection and advertisement putting.The specific implementation process of this technology: at first, obtain the telecommunications feature and the demographic characteristic of sample of users by the methods of sampling, comprehensive these two parts data, add user's personal consumption characteristic, sample of users is subdivided into the consumption feature segmentation colony of consumption feature difference with clustering method, then, the analyzing samples data are set up with other discrimination model of telecommunications feature derivation disaggregated classification.At last this discrimination model is applied to full database data, thereby segments colony's class label, and the disaggregated classification distinguishing label is applied to mobile phone advertisement master's advertisement putting for each user stamps corresponding consumption feature.
The inventive method may further comprise the steps (as shown in Figure 1, 2):
(1) at first from whole customer data bases (abbreviating full database data as), obtains the sample of users database, and therefrom draw telecommunications characteristic and demographic characteristic by the methods of sampling;
(2) obtain this part user's personal consumption feature then from other approach.Clustering method during then maintenance data excavates and analyzes, the consumption feature that sample of users is subdivided into consumption feature difference is segmented colony.
(3) analyze set up this moment comprise telecommunications feature, demographic feature, and life style characteristic, and be divided into the sample of users database of consumption feature segmentation colony is set up with other discrimination model of telecommunications feature derivation disaggregated classification.
(4) at last this discrimination model is applied to full database data, thereby segments colony's class label, and the disaggregated classification distinguishing label is applied to mobile phone advertisement master's advertisement putting for each user stamps corresponding consumption feature.
For achieving the above object, the present invention adopts following technical scheme.
A kind of mobile phone advertisement put-on method based on the mobile phone advertisement subscriber segmentation the steps include:
1) at first obtain the sample of users database by the methods of sampling from whole customer data bases, these data comprise user's telecommunication service data and consensus data;
2),, obtain this part user's personal consumption feature such as at methods such as these users' street corner investigation, phone investigations then from other approach.If can not obtain certain user's personal consumption feature, then in following process of cluster analysis, reject these users, in order to avoid pollute user's sample;
3),, sample of users is subdivided into the consumption feature segmentation colony of consumption feature difference according to the clustering method in excavation of consumption feature maintenance data and the analysis in conjunction with above 3 class data;
4) analyze this consumption feature segmentation colony, set up with other discrimination model of telecommunications feature derivation disaggregated classification; Setting up discrimination model is the known technology of this area.
5) this discrimination model is applied to full database data, successively at each user's data, each consumer label is carried out discriminating processing, draw a numerical value at this label, this numerical value is the probable value that this user satisfies this consumer label.Thereby segment colony's class label for each user stamps corresponding consumption feature, and the disaggregated classification distinguishing label is applied to mobile phone advertisement master's advertisement putting.
In the described method " methods of sampling ", mainly be meant " systemic sampling method ", because quite huge at cellphone subscriber's database, number of individuals is more;
In the described method " cluster analysis ", be exactly that index is according to indication a kind of statistical analysis technique that things of a kind come together, people of a mind fall into the same group in excavation and the analytical technology.Cluster analysis comes down to seek a kind of statistic that can objectively respond close and distant relation between the element, then according to this statistic the element divide into several classes.Used cluster statistic comprises distance coefficient and similarity factor 2 classes in this method.
In the described method " discrimination model ", be exactly the condition model, or the conditional probability model.The target that he will reach is, at a non-sample user, differentiate through this model, with user's data and the customer consumption category feature of wanting to differentiate be input, the computing through this model can draw a numerical value.This numerical value is the probable value that this user has this type of consumption feature model.
Compared with prior art, advantage of the present invention and good effect are:
In current industrial environment, the present invention has well solved at the advertiser and has carried out mobile phone advertisement when throwing in, the cellphone subscriber of required selection lacks the data problem of consumption feature, thereby improved advertiser's input enthusiasm, and because being used of the existence of these data, cause ad content to meet user's potential demand probably more, improved cellphone subscriber's experience.For the effect of the inventive method is described, we are example with an actual scene.Suppose that certain apparel brand producer shop wants by certain cellular telephone companies, thrown in mobile phone advertisement at 1,000,000 potential buyers of nationwide.This apparel brand is located high-end business people.
In this time,, suppose to have a discrimination model of checking by analysis according to our method.This model has provided following differentiation conclusion: every month telephone expenses are more than 300 yuan, often roaming (these two are the telecommunication service data), male sex's (this is the consensus data) and in life, pursue quality, the cellphone subscriber of be busy with one's work (these two are the life style data), he is that high-end business people's probability is 0.8.And this model has been used in the full database data, has stamped the label of " high-end business people " to the stakeholder.
At this moment, for this brand producer, he directly clicks " high-end business people " this label in advertisement delivery, and specifies and oneself will throw in 1,000,000 quantity and get final product! Very efficient, be convenient to understand.
Certainly, if there is not " high-end business people " this label in the database, also it doesn't matter.He can select close " a commercial man " or " successful personage " or the like.In specific implementation, determining of consumer feature tag is through investigating and directly and after the advertiser of the every profession and trade communication establishing.Perhaps, can come to offer more accurately the advertiser and select by the combination of label.
Description of drawings
Fig. 1 has illustrated subscriber segmentation method flow diagram of the present invention;
Fig. 2 has illustrated mobile phone advertisement put-on method flow chart of the present invention.
Embodiment:
Illustrate that below by a concrete example how implementing the described method of this patent obtains any one cellphone subscriber's consumption feature, and practice is in the thin field of mobile phone advertisement user.Suppose that we have built the environment of a whole set of user data excavation and data analysis according to this method, and an advertisement putting environment.And a brand clothing manufacturer, New Year's Day, be to nationwide potential user's advertisement delivery.
At first we can through systematic sampling, draw 100,000 data such as in 1,000 ten thousand cellphone subscriber's data from full database data.In the middle of these data, have only user's telecommunication service data and simple consensus data.The relevant questionnaire of our well-designed life styles then, by to the visit of the phone at this 10 general-purpose family or concentrate variety of way such as interview, obtain their life style data, such as consumption propensity, whether often take exercises, whether make the shopping place that applies some make up, often go and shopping frequency or the like.
Our comprehensive above 3 class data adopt clustering method then.Can draw this 10 general-purpose family and be clustered into 8 class life style colonies, simultaneously this telecommunication service characteristic based on certain user determines the discrimination model that this user belongs to which class life style user of colony in this 8 class and also sets up.
We utilize this discrimination model then, at this 1000 general-purpose user data of all databases, differentiate.Each user has been included into a kind of in this 8 class life style label.Wherein " business people " this class crowd just has 1,200,000.
Like this, in the advertisement putting environment, when this producer wants to throw in, found " business people " this user group, he thinks that this colony and he's targeted customer is very identical, then just very happy selection this colony, and 1,000,000 people, everyone advertisement have once directly been thrown in.
After the input, one of them user is such as the senior executive who is exactly certain company.Seen the advertisement of this brand clothing, very happy.Because with regard to the fan of this brand, perhaps many friends at one's side are exactly the fan of this brand at ordinary times for he.
Whole process is very convenient, efficient.And user experience is fine.
Claims (7)
1. the mobile phone advertisement put-on method based on the mobile phone advertisement subscriber segmentation the steps include:
1) mobile phone advertisement publisher server sample drawn user data from customer data base, described sample data comprises user's telecommunication service data message and consensus data's information;
2) obtain the consumption feature data message of sample of users and be entered into the mobile phone advertisement publisher server;
3) the mobile phone advertisement publisher server carries out cluster analysis according to consumption feature to the sample of users data, obtains the consumption feature segmentation colony of consumption feature difference;
4) the mobile phone advertisement publisher server utilizes described consumption feature to segment colony, sets up with other discrimination model of telecommunications feature derivation disaggregated classification;
5) the mobile phone advertisement publisher server utilizes described discrimination model that each user in the customer data base is stamped corresponding consumption feature and segments colony's class label;
6) the mobile phone advertisement main root is chosen the user according to the disaggregated classification distinguishing label and is carried out advertisement putting.
2. the method for claim 1 is characterized in that described consumption feature comprises: consuming capacity, consumption habit, shopping place.
3. the method for claim 1 is characterized in that obtaining by the method for street corner investigation or phone investigation the consumption feature data of sample of users.
4. method as claimed in claim 3 is characterized in that carrying out cluster analysis then if the consumption feature data of some sample of users of obtaining for empty, are then deleted the respective sample user data in the sample of users data.
5. as claim 1 or 4 described methods, it is characterized in that utilizing distance coefficient and similarity factor to carry out cluster analysis.
6. the method for claim 1, it is characterized in that user's data and this customer consumption category feature are imported described discrimination model carries out computing, draw the probable value that this user has this type of consumption feature model, if probable value is greater than setting threshold then mark the user as corresponding consumption feature segmentation colony class label.
7. as claim 1 or 6 described methods, it is characterized in that described discrimination model is condition model or conditional probability model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106035237A CN102118706A (en) | 2010-12-14 | 2010-12-14 | Mobile phone advertising method based on subdivision of mobile phone advertisement users |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106035237A CN102118706A (en) | 2010-12-14 | 2010-12-14 | Mobile phone advertising method based on subdivision of mobile phone advertisement users |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102118706A true CN102118706A (en) | 2011-07-06 |
Family
ID=44217264
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010106035237A Pending CN102118706A (en) | 2010-12-14 | 2010-12-14 | Mobile phone advertising method based on subdivision of mobile phone advertisement users |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102118706A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102592236A (en) * | 2011-12-28 | 2012-07-18 | 北京品友互动信息技术有限公司 | Internet advertising crowd analysis system and analysis method |
| CN104063801A (en) * | 2014-06-23 | 2014-09-24 | 广州优蜜信息科技有限公司 | Mobile advertisement recommendation method based on cluster |
| CN105095346A (en) * | 2015-06-05 | 2015-11-25 | 中国联合网络通信集团有限公司 | Service push method and apparatus |
| CN105227575A (en) * | 2015-10-26 | 2016-01-06 | 刘永锋 | A kind of telecommunication user personal identification method |
| CN105744005A (en) * | 2016-04-30 | 2016-07-06 | 平安证券有限责任公司 | Client positioning and analyzing method and server |
| CN105868243A (en) * | 2015-12-14 | 2016-08-17 | 乐视网信息技术（北京）股份有限公司 | Information processing method and apparatus |
| CN106294812A (en) * | 2016-08-16 | 2017-01-04 | 中国联合网络通信有限公司吉林省分公司 | Number washes in a pan self-service screening service system |
| CN107203772A (en) * | 2016-03-16 | 2017-09-26 | 阿里巴巴集团控股有限公司 | A kind of user type recognition methods and device |
| CN107403019A (en) * | 2017-08-15 | 2017-11-28 | 重庆邮电大学 | A kind of vehicle owner identification method based on mobile data |
| CN107563807A (en) * | 2017-08-29 | 2018-01-09 | 重庆邮电大学 | A kind of regional advertisement supplying system based on data mining |
| CN108765015A (en) * | 2018-05-30 | 2018-11-06 | 苏州介观软件技术有限公司 | Advertisement accurately jettison system in city |
| US10248527B1 (en) | 2018-09-19 | 2019-04-02 | Amplero, Inc | Automated device-specific dynamic operation modifications |
| CN109583964A (en) * | 2018-12-07 | 2019-04-05 | 中国银行股份有限公司 | Advertisement placement method and device |
| CN110009401A (en) * | 2019-03-18 | 2019-07-12 | 康美药业股份有限公司 | Advertisement placement method, device and storage medium based on user's portrait |
| CN110263862A (en) * | 2019-06-21 | 2019-09-20 | 北京字节跳动网络技术有限公司 | Method for pushing, device, electronic equipment and the readable storage medium storing program for executing of information |
| CN110995839A (en) * | 2019-12-03 | 2020-04-10 | 北京搜狐新媒体信息技术有限公司 | Method and device for analyzing performance of advertisement system and computer storage medium |
| CN111127091A (en) * | 2019-12-20 | 2020-05-08 | 西安万像电子科技有限公司 | Advertisement putting method, device and system |
-
2010
- 2010-12-14 CN CN2010106035237A patent/CN102118706A/en active Pending
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102592236A (en) * | 2011-12-28 | 2012-07-18 | 北京品友互动信息技术有限公司 | Internet advertising crowd analysis system and analysis method |
| CN104063801A (en) * | 2014-06-23 | 2014-09-24 | 广州优蜜信息科技有限公司 | Mobile advertisement recommendation method based on cluster |
| CN104063801B (en) * | 2014-06-23 | 2016-05-25 | 有米科技股份有限公司 | A kind of moving advertising recommend method based on cluster |
| CN105095346A (en) * | 2015-06-05 | 2015-11-25 | 中国联合网络通信集团有限公司 | Service push method and apparatus |
| CN105227575A (en) * | 2015-10-26 | 2016-01-06 | 刘永锋 | A kind of telecommunication user personal identification method |
| CN105868243A (en) * | 2015-12-14 | 2016-08-17 | 乐视网信息技术（北京）股份有限公司 | Information processing method and apparatus |
| CN107203772A (en) * | 2016-03-16 | 2017-09-26 | 阿里巴巴集团控股有限公司 | A kind of user type recognition methods and device |
| CN105744005A (en) * | 2016-04-30 | 2016-07-06 | 平安证券有限责任公司 | Client positioning and analyzing method and server |
| CN106294812A (en) * | 2016-08-16 | 2017-01-04 | 中国联合网络通信有限公司吉林省分公司 | Number washes in a pan self-service screening service system |
| CN107403019B (en) * | 2017-08-15 | 2020-08-18 | 重庆邮电大学 | Vehicle owner identity recognition method based on mobile data |
| CN107403019A (en) * | 2017-08-15 | 2017-11-28 | 重庆邮电大学 | A kind of vehicle owner identification method based on mobile data |
| CN107563807A (en) * | 2017-08-29 | 2018-01-09 | 重庆邮电大学 | A kind of regional advertisement supplying system based on data mining |
| CN108765015A (en) * | 2018-05-30 | 2018-11-06 | 苏州介观软件技术有限公司 | Advertisement accurately jettison system in city |
| US10248527B1 (en) | 2018-09-19 | 2019-04-02 | Amplero, Inc | Automated device-specific dynamic operation modifications |
| CN109583964A (en) * | 2018-12-07 | 2019-04-05 | 中国银行股份有限公司 | Advertisement placement method and device |
| CN110009401A (en) * | 2019-03-18 | 2019-07-12 | 康美药业股份有限公司 | Advertisement placement method, device and storage medium based on user's portrait |
| CN110263862A (en) * | 2019-06-21 | 2019-09-20 | 北京字节跳动网络技术有限公司 | Method for pushing, device, electronic equipment and the readable storage medium storing program for executing of information |
| CN110263862B (en) * | 2019-06-21 | 2021-05-07 | 北京字节跳动网络技术有限公司 | Information pushing method and device, electronic equipment and readable storage medium |
| CN110995839A (en) * | 2019-12-03 | 2020-04-10 | 北京搜狐新媒体信息技术有限公司 | Method and device for analyzing performance of advertisement system and computer storage medium |
| CN111127091A (en) * | 2019-12-20 | 2020-05-08 | 西安万像电子科技有限公司 | Advertisement putting method, device and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102118706A (en) | Mobile phone advertising method based on subdivision of mobile phone advertisement users | |
| Izquierdo-Yusta et al. | Attitudes toward mobile advertising among users versus non-users of the mobile Internet | |
| Okazaki | What do we know about mobile Internet adopters? A cluster analysis | |
| Bawm et al. | A Conceptual Model for effective email marketing | |
| CN108537578A (en) | Advertisement sending method based on big data and device | |
| CN106960354A (en) | Method and device is recommended in a kind of precision based on customer life cycle | |
| US20070239538A1 (en) | Incentivized relationship-data communication to select targeted content method and system | |
| CN107103485B (en) | Automatic advertisement recommendation method and system according to cinema visitor information | |
| CN112418932B (en) | Marketing information pushing method and device based on user tag | |
| CN102592236A (en) | Internet advertising crowd analysis system and analysis method | |
| CN106296314A (en) | Push the method and system of targeting advertisement | |
| CN106202371B (en) | Media file processing method and device and advertisement analysis method | |
| US20150248501A1 (en) | Content analytics | |
| Zhang et al. | Mining target users for mobile advertising based on telecom big data | |
| CN115249170A (en) | Marketing customer group reach method and device based on 5G message | |
| Chaparro-Peláez et al. | May I have your attention, please? An investigation on opening effectiveness in e-mail marketing | |
| KR20130113251A (en) | Multiple structures data standardization processing system for integrated analysis of sns | |
| CN116934372A (en) | Store operation customer data management method and system | |
| CN110020123A (en) | A kind of promotion message put-on method, device, medium and equipment | |
| Prakash et al. | A model on understanding mobile buying behaviour of migrant workers with consideration to Kerala region | |
| CN111382343A (en) | Label system generation method and device | |
| Tanase | Responding to Customer Behaviour in Real Time: From Campaign Management to Marketing Automation | |
| CN110866651B (en) | Advertisement click rate prediction method | |
| KR101743217B1 (en) | Multi-media network service system based on big data, service method and computer readable medium having computer program recorded therefor | |
| CN107818471A (en) | One kind is based on O2O electric business user data method for digging under big data environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110706 |