CN106022810A - Advertisement audience targeting and putting system based on driving behavior of car owner - Google Patents
Advertisement audience targeting and putting system based on driving behavior of car owner Download PDFInfo
- Publication number
- CN106022810A CN106022810A CN201510077256.7A CN201510077256A CN106022810A CN 106022810 A CN106022810 A CN 106022810A CN 201510077256 A CN201510077256 A CN 201510077256A CN 106022810 A CN106022810 A CN 106022810A
- Authority
- CN
- China
- Prior art keywords
- car owner
- behavior
- driving
- advertisement
- label
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to an advertisement audience targeting and putting system based on the driving behavior of a car owner, and the system comprises the following modules: a car owner behavior label module which forms driving data according to the driving behavior of the car owner, uploads the driving data to a cloud platform through a GPS of a car, carries out the analysis of the driving data through the cloud platform, and pastes a corresponding behavior label for the user according to an analysis result; a car owner attribute defining module which carries out the traversal of all behavior labels of the car owner, quantitates the importance of each behavior to the car owner through weighted normalization, and forms the attribute definition of the car owner; a car owner clustering module which carries out the clustering of the car owner according to the attribute definition of the car owner, and enables the car owners with the similar attributes to be classified into one class; and an advertisement directional putting module which carries out the mapping and matching of the car owner cluster and an advertisement target group, and carries out the directional putting of advertisement for a matched car owner through a navigator. The system can precisely put an advertisement through the navigator.
Description
Technical field
The invention belongs to advertisement putting field, relate to a kind of advertisement crowd orientation and jettison system,
It is specifically related to a kind of advertisement crowd based on car owner's driving behavior orientation and jettison system.
Background technology
Crowd's orientation is exactly target group's orientation, in marketing, product orientation and people in fact
Group's segmentation is very important theory, and this theory have also been obtained the accreditation in market, therefore
Each product just has determined that the target group of oneself at the beginning of design, production.From us
Advertisement putting, from the point of view of the market propaganda, be necessarily intended to carry out to target group, spend
Popularization outside target group is all waste.
But in conventional medium, it is desirable to identify user completely, to determine whether target group
It is not easy thing, does not accomplishes the most completely, can only be by difference
Medium means go close target group as far as possible.
Internet era, by the strength of technology, can be unlimited close, be close to accurately
Judge everyone attribute, thus orient service for advertiser target group.
If in existing automobile, advertiser is thrown in by radio, but is thrown by radio
Put advertisement to have the following disadvantages:
1, the most deep in China's media research, to emerge in large numbers from media, media data market is gradually
Formed today, broadcast audience rate data lack relatively, and exist the quality of data the highest, analyze
Not science, be formed without system, there is no the problems such as unified specification so that numerous advertising companies
When doing media input plan, it is absorbed in the awkward situation not having data to be referred to.Can owing to lacking
Broadcast audience rate data, cause broadcasting a lot of advantage and be not found, advertiser hope and
Stepping back, broadcasting station disorderly competition, radio advertising run off in a large number, and the investigation of broadcast audience rate puts into
Limited, radio advertising operation has been absorbed in vicious cycle.For car owner's free time in car,
Suitable advertisement is not had to push.The FM ad quality and the effect that push are the most very poor.
2, radio ads user group can not segment, if certain car owner driving BMW 740
Turning on radio, the automotive advertising heard is probably Chevrolet Sail, the precision extreme difference of advertisement,
Experience bad luck.
Summary of the invention
Goal of the invention: the problem that the present invention is directed to the existence of above-mentioned prior art makes improvement, i.e. originally
A kind of advertisement crowd based on car owner's driving behavior of disclosure of the invention orientation and jettison system.
Technical scheme: a kind of advertisement crowd based on car owner's driving behavior orientation and jettison system,
Comprise with lower module:
Car owner's behavior label model, forms, according to the behavior of driving of car owner, data of driving, logical
Cross the GPS of vehicle data of driving and upload to cloud platform, by cloud platform to driving
Data are analyzed, and stick the behavior label of correspondence to user according to analysis result；
Car owner's attribute definition module, whole behavior labels of traversal car owner, pass through weighting normalization
Quantify each behavior label significance level for car owner, form the attribute definition of car owner；
Car owner's cluster module, clusters car owner according to the attribute definition of car owner, will have phase
It is attributed to same class like the car owner of attribute；
Advertisement orientation putting module, to car owner cluster with advertising objective crowd carry out map and
Join, be oriented advertisement putting by navigator to mating qualified car owner.
As a kind of advertisement crowd based on car owner's driving behavior orientation in the present invention and input system
A kind of preferred version of system: data of driving include running region, running time section, traveling
Rule, level of driving, custom of driving, haunt, movable place and temporal regularity.
As a kind of advertisement crowd based on car owner's driving behavior orientation in the present invention and input system
A kind of preferred version of system: behavior label includes the explanation of tag identifier, label and label number of times.
As a kind of advertisement crowd based on car owner's driving behavior orientation in the present invention and input system
A kind of preferred version of system: cluster includes that car owner clusters mark, the total attribute character of cluster and car
Main set.
Beneficial effect: the invention discloses a kind of advertisement crowd based on car owner's driving behavior orientation
And jettison system, it has the advantages that
1, there is no advertisement accurately jettison system in current car, the when that car owner driving, only have radio reception
Machine, wireless effect of advertising are poor, by this invention, logical in the time that can car owner be driven
Cross navigator and precisely throw in advertisement；
2, the demand of car owner can be understood accurately, throw in advertisement based on demand.
Accompanying drawing explanation
Fig. 1 is the flow chart labeing behavior according to car owner's behavior of driving.
Detailed description of the invention:
Below the detailed description of the invention of the present invention is described in detail.
As it is shown in figure 1, a kind of advertisement crowd based on car owner's driving behavior orientation and input system
System, comprises with lower module:
Car owner's behavior label model, forms, according to the behavior of driving of car owner, data of driving, logical
Cross the GPS of vehicle data of driving and upload to cloud platform, by cloud platform to driving
Data perform an analysis, and stick the behavior label of correspondence to user according to analysis result；
Car owner's attribute definition module, whole behavior labels of traversal car owner, pass through weighting normalization
Quantify each behavior label significance level for car owner, form the attribute definition of car owner；
Car owner's cluster module, clusters car owner according to the attribute definition of car owner, will have phase
It is attributed to same class like the car owner of attribute；
Advertisement orientation putting module, to car owner cluster with advertising objective crowd carry out map and
Join, be oriented advertisement putting by navigator to mating qualified car owner.
In the present embodiment, data of driving include running region, running time section, travel rule
Rule, level of driving, custom of driving, haunt, movable place and temporal regularity.
In the present embodiment, behavior label includes the explanation of tag identifier, label and label number of times.
In the present embodiment, cluster includes that car owner clusters mark, the total attribute character of cluster and car owner
Set.
During use, when car owner is during driving, the GPS of its vehicle can be by row of driving
Sailing data and upload in the big data of cloud platform, the analysis module of cloud platform can analyze user data,
According to behavioral datas such as the running region of user, running times, user is done behavior label and distinguishes,
These labels include but not limited to car owner like going window-shopping, like to steep in wine, like having Western food, in the recent period
There is house-purchase demand etc., after specifying behavior label to user, by the behavior label got,
Can progressively set up complete car owner's Analysis model of network behaviors, excavate and there is similar behavior characteristics
Car owner, refines the attribute character of car owner, finally does advertisement accurately according to these attribute character and throw
Put.
The driving data according to car owner will be told about below the most in detail be car owner and orient district
The flow process divided and module:
Car owner's behavior label model
Car owner's behavior label model mainly (includes travelling district according to the data of driving of car owner
Territory, running time section, travel rule, level of driving, custom of driving, haunt, live
Dynamic place and temporal regularity), data are performed an analysis, then according to analysis result, to user
Stick the behavior label of correspondence；
Behavior tag definition is tuple more than, and content is as follows:
1.tagid tag identifier, each label has a unique id the most corresponding；
2. label explanation: go window-shopping, bar, remove building, run weekend long-distance grade one of them；
3. label number of times: the frequency of single month behavior label.
During behavior tag processes, investigate multiple typical case car owner's behavior, such as: geographical position,
Zone of action, activity attributes, the activity time.After having investigated car owner's behavior, it is possible to determine
Whether this car owner meets the condition of this label, if it is satisfied, then this car owner has this label, with
The number of times of this label of Shi Tongji.
Meanwhile, by living habit and the log-on message of website of car owner, car owner can be got
Personal information, such as: age bracket (young, middle age), sex, occupation, educational background etc..
It these information is each a label.The most each car owner can corresponding multiple labels.
Car owner's attribute definition module
Car owner's behavior label model has processed car owner's behavioural information, by adding behavioural information
Power normalized, whole behavior labels of traversal car owner, quantify each behavior label for car
Main significance level, is the formation of the attribute definition to car owner.
Car owner's attribute definition be one group with the behavior label of tagid sort ascending and label pair
The set of the weight answered, car owner's attribute is also tuple more than
1.pointid: car owner attribute id；
2. audient is designated the i-th behavior label model of pointid；
3. behavior label weight mapping function, portrays the behavior significance level for car owner's attribute；
4. the scope of weight coefficient.
What car owner's attribute definition was portrayed is the behavior characteristics of car owner, and its dimension is exactly different row
For label, generally up to tens dimensions.Mapping function makes the behavior of label be weighted normalizing
Change, can be designed according to no business demand.It is said that in general, purchasing behavior is the strongest
Behavior label, weight is the highest；It is other that driving habits then belongs to time intensity level, shows the subjectivity of user
Characteristic.
Assuming that the behavior label of car owner kandy has two, one is always to go to sell for nearest one month
Building portion, another one label is that he always brings to a halt, and for first label, belongs to purchase
Behavior label, weight is the highest, and second label belongs to driving habits, and weight is more weak.So far may be used
To show that kandy may have purchase of property desire, driving habits is bad correspondence may become rich secondary.
So far,
The interest attribute of car owner and feature, by the behavior of car owner, be abstracted through above-mentioned steps
With reconstruct out.
Car owner's cluster module
After accumulation car owner's attribute data, just there are enough data volumes that big data process, connect down
Just can cluster car owner according to the attribute of car owner, the car owner that will have like attribute does
Classifying accurately, each car owner may belong in multiple different classification.Car owner's cluster is adopted
With the clustering algorithm of data mining.If cluster is a process that data set is divided into Ganlei.
The objective of cluster and judgment criteria are so that homogeneous object similarity is big as far as possible, and each class it
Between similarity the least.
Each cluster of car owner is made up of 3 parts
1.clusterid car owner clusters mark, and each car owner cluster has unique one
Id is the most corresponding；
2. the total attribute character of cluster, such as, all likes eating Guangdong dishes, or likes upper bar；
3. car owner's set, all belongs to the car owner's object set being designated clusterid.
Car owner's clustering algorithm of this invention has used for reference CURE hierarchical clustering algorithm, and on its basis
On with reference to other clustering algorithms, merge and improve, make whole clustering algorithm more conform to reality
Advertising business demand.Layered approach is combined together by CURE algorithm with division methods.It
Overcome deflection and find similar size and the problem of round-shaped cluster；Processing high dimension simultaneously
According to also show to obtain more efficient stable during abnormal data.Compare other algorithms, CURE algorithm
Mainly there is a following advantage:
1. by using " representing a little " to represent a cluster, so that clustering algorithm can
Adapt to various erose class, and be not only spherical；
2. by the factor, representative point is shunk to cluster centre, so that clustering algorithm supports
The ability of anti-outlier is higher；
3. by stochastic sampling and partitioning technology so that CURE algorithm can process ultra-large
Data, to accelerate the execution efficiency of clustering algorithm.
After clustering processing, understand that certain car owner belongs to that several clusters, often the most clearly
Individual cluster is all any characteristic.Such as kandy may belong to following cluster: purchases house in the recent period, happiness
Joyous upper bar, likes Guangdong dishes etc..
Car owner orients putting module
Final step before formally throwing in as car owner's targeted ads, car owner's cluster and advertising objective
Crowd carries out mapping and mating.This mapping is that the registration of both utilizations attribute realizes.
When crowd's orthoselection " luxury goods " or " automobile " of an advertisement serving policy,
Advertisement allows to be designated and is delivered to belong to luxury goods and the car owner of automobile cluster.
Above embodiments of the present invention are elaborated.On but the present invention is not limited to
State embodiment, in the ken that art those of ordinary skill is possessed, also
Can make a variety of changes on the premise of without departing from present inventive concept.
Claims (4)
1. advertisement crowd based on car owner's driving behavior orientation and a jettison system, its feature
It is, comprises with lower module:
Car owner's behavior label model, forms, according to the behavior of driving of car owner, data of driving, logical
Cross the GPS of vehicle data of driving and upload to cloud platform, by cloud platform to driving
Data are analyzed, and stick the behavior label of correspondence to user according to analysis result；
Car owner's attribute definition module, whole behavior labels of traversal car owner, pass through weighting normalization
Quantify each behavior label significance level for car owner, form the attribute definition of car owner；
Car owner's cluster module, clusters car owner according to the attribute definition of car owner, will have phase
It is attributed to same class like the car owner of attribute；
Advertisement orientation putting module, to car owner cluster with advertising objective crowd carry out map and
Join, be oriented advertisement putting by navigator to mating qualified car owner.
A kind of advertisement crowd based on car owner's driving behavior orients
And jettison system, it is characterised in that data of driving include running region, running time section,
Travel rule, level of driving, custom of driving, haunt, movable place and time rule
Rule.
A kind of advertisement crowd based on car owner's driving behavior orients
And jettison system, it is characterised in that behavior label includes the explanation of tag identifier, label and label
Number of times.
A kind of advertisement crowd based on car owner's driving behavior orients
And jettison system, it is characterised in that cluster includes that car owner clusters mark, cluster total attribute spy
The car owner that seeks peace gathers.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510077256.7A CN106022810A (en) | 2015-02-13 | 2015-02-13 | Advertisement audience targeting and putting system based on driving behavior of car owner |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510077256.7A CN106022810A (en) | 2015-02-13 | 2015-02-13 | Advertisement audience targeting and putting system based on driving behavior of car owner |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106022810A true CN106022810A (en) | 2016-10-12 |
Family
ID=57082385
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510077256.7A Pending CN106022810A (en) | 2015-02-13 | 2015-02-13 | Advertisement audience targeting and putting system based on driving behavior of car owner |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106022810A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993966A (en) * | 2018-01-02 | 2019-07-09 | 中国移动通信有限公司研究院 | A kind of method and device of building user portrait |
| CN110275167A (en) * | 2019-06-03 | 2019-09-24 | 浙江吉利控股集团有限公司 | A kind of control method of radar detection, controller and terminal |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101427279A (en) * | 2006-04-25 | 2009-05-06 | 丰田自动车株式会社 | Vehicle environmental service system |
| CN103218442A (en) * | 2013-04-22 | 2013-07-24 | 中山大学 | Method and system for life mode analysis based on mobile device sensor data |
| CN103778555A (en) * | 2014-01-21 | 2014-05-07 | 北京集奥聚合科技有限公司 | User attribute mining method and system based on user tags |
| CN104158846A (en) * | 2014-07-16 | 2014-11-19 | 深圳市凯伦圣科技有限公司 | Driving behavior analysis method and system |
-
2015
- 2015-02-13 CN CN201510077256.7A patent/CN106022810A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101427279A (en) * | 2006-04-25 | 2009-05-06 | 丰田自动车株式会社 | Vehicle environmental service system |
| CN103218442A (en) * | 2013-04-22 | 2013-07-24 | 中山大学 | Method and system for life mode analysis based on mobile device sensor data |
| CN103778555A (en) * | 2014-01-21 | 2014-05-07 | 北京集奥聚合科技有限公司 | User attribute mining method and system based on user tags |
| CN104158846A (en) * | 2014-07-16 | 2014-11-19 | 深圳市凯伦圣科技有限公司 | Driving behavior analysis method and system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993966A (en) * | 2018-01-02 | 2019-07-09 | 中国移动通信有限公司研究院 | A kind of method and device of building user portrait |
| CN110275167A (en) * | 2019-06-03 | 2019-09-24 | 浙江吉利控股集团有限公司 | A kind of control method of radar detection, controller and terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110335126B (en) | Automobile industry big data passenger obtaining method | |
| Punel et al. | Using Twitter network to detect market segments in the airline industry | |
| US10423983B2 (en) | Determining targeting information based on a predictive targeting model | |
| Rixey | Station-level forecasting of bikesharing ridership: Station network effects in three US systems | |
| CN104796481B (en) | Intelligent audio and video selection method | |
| Gruber et al. | Reject or embrace? Messengers and electric cargo bikes | |
| CN1987916A (en) | Method and device for releasing network advertisements | |
| CN102609523A (en) | Collaborative filtering recommendation algorithm based on article sorting and user sorting | |
| Heryati et al. | The application of data mining by using K-means clustering method in determining new students’ admission promotion strategy | |
| CN104217353A (en) | Webpage advertisement directional pushing system | |
| CN106530094A (en) | Vehicle insurance assessment system and assessment method | |
| CN107077498A (en) | The presentation-entity relation in online advertisement | |
| CN106227900A (en) | Recommending scenery spot equipment, method and system | |
| KR102456425B1 (en) | Brand image and influencer matching method, apparatus and system based on the category of brand image | |
| Lipyanina et al. | Targeting Model of HEI Video Marketing based on Classification Tree. | |
| CN106022810A (en) | Advertisement audience targeting and putting system based on driving behavior of car owner | |
| Leao et al. | Validating crowdsourced bicycling mobility data for supporting city planning | |
| Sun et al. | Discovering vehicle usage patterns on the basis of daily mobility profiles derived from floating car data | |
| Ergun et al. | Creating a geodemographic classification model within geo-marketing: the case of Eskişehir province | |
| Mohiuddin et al. | Examining market segmentation to increase bike-share use: The case of the Greater Sacramento Region | |
| Lin et al. | Using multivariate statistical methods to analyze high-quality bicycle path service systems: A case study of popular bicycle paths in Taiwan | |
| Wafa et al. | Latent-segmentation-based approach to investigating spatial transferability of activity-travel models | |
| Zhang et al. | An automatic approach for transit advertising in public transportation systems | |
| Saputra et al. | Selective federated learning for on-road services in internet-of-vehicles | |
| Akbar et al. | Analysis Of Satisfaction and Priority Levels of Jaya Makmur Road Development in Supporting Kurik District as A Rice Surplus Area |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20180723 Address after: 511458 9, Nansha District Beach Road, Guangzhou, Guangdong, 9 Applicant after: Rui Chi intelligent automobile (Guangzhou) Co., Ltd. Address before: 610000 Sichuan Chengdu high tech Zone No. 1 Tianfu 19 street 1 1 unit 14 level 1403 Applicant before: Chengdu easy Technology Co., Ltd. |
|
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 511458 9, Nansha District Beach Road, Guangzhou, Guangdong, 9 Applicant after: Hengda Faraday future intelligent vehicle (Guangdong) Co., Ltd. Address before: 511458 9, Nansha District Beach Road, Guangzhou, Guangdong, 9 Applicant before: Rui Chi intelligent automobile (Guangzhou) Co., Ltd. |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20190319 Address after: 100015 Building No. 7, 74, Jiuxianqiao North Road, Chaoyang District, Beijing, 001 Applicant after: FAFA Automobile (China) Co., Ltd. Address before: 511458 9, Nansha District Beach Road, Guangzhou, Guangdong, 9 Applicant before: Hengda Faraday future intelligent vehicle (Guangdong) Co., Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161012 |