CN110020666A - A kind of public transport advertisement placement method and system based on passenger behavior mode - Google Patents
A kind of public transport advertisement placement method and system based on passenger behavior mode Download PDFInfo
- Publication number
- CN110020666A CN110020666A CN201910128113.2A CN201910128113A CN110020666A CN 110020666 A CN110020666 A CN 110020666A CN 201910128113 A CN201910128113 A CN 201910128113A CN 110020666 A CN110020666 A CN 110020666A
- Authority
- CN
- China
- Prior art keywords
- passenger
- data
- trip
- public transport
- vector
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a kind of public transport advertisement placement methods based on passenger behavior mode, and packet pre-processes collected passenger data, and builds passenger's trip data library;Data in passenger's trip data library are gathered, the vector of related passenger's trip is formed;The vector is clustered;Data after cluster are divided into N number of data class, according to the center vector of each data class, pattern-recognition is carried out to each data class by investigating and simulating;N≥2;Advertisement is accurately launched in public transport according to the result of pattern-recognition.Such method can launch advertisement according to the different demands of passenger with a definite target in view, improve the accuracy of advertisement dispensing, improve the utilization rate of advertising resource.
Description
Technical field
The present invention relates to the processing technology fields of bus passenger data, and in particular to a kind of public affairs based on passenger behavior mode
Transit advertising put-on method and system altogether.
Background technique
For the operator of advertiser and public transport, and the passenger of public transport is taken, is desirable in public transport
It is upper that it can be seen that meeting the advertisement of passenger's actual demand precisely launched, passenger will have preferably to be experienced by bus, advertiser and
Public transport operator can also therefrom improve the income of oneself, reach the win-win of businessman and consumer.
Current bus advertisement is launched substantially in a manner of " alignment determines portion's timing ", and businessman and advertising company consult throwing
Put advertisement route, launch advertisement position and launch advertisement period, launch route, the selection of release time substantially with
Route approach website situation and volume of the flow of passengers size are selection gist.Such putting mode effect is little, and and resource is not implemented
It is complete to utilize.It is variable LED electronic display in public transit system, vehicle-mounted for example, with the continuous development of internet new media
The advertisement that the injected volume presentation of TV etc. increases substantially trend, but is lack of pertinence at present launch mode make LED display,
In-car TV not playing its due dynamic propaganda function really, instead with the plane formula exploitation difference of public transport vehicle body
Less.Furthermore not high for the trip prediction accuracy and timeliness of public transport passenger at this stage.Therefore, it is badly in need of grinding in industry
It sends out a kind of based on passenger behavior model prediction public transport passenger trip and then targetedly launches the method and system of advertisement.
Summary of the invention
The purpose of the invention is to overcome above the shortcomings of the prior art, provide a kind of based on passenger behavior mould
The public transport advertisement placement method of formula.
It is another object of the present invention to provide a kind of based on passenger's row to overcome above the shortcomings of the prior art
For the public transport advertisement delivery system of mode.
The purpose of the present invention is realized by the following technical solution:
A kind of public transport advertisement placement method based on passenger behavior mode, comprising:
S1 pre-processes collected passenger data, and builds passenger's trip data library;
S2 gathers the data in passenger's trip data library, forms the vector of related passenger's trip;
S3 clusters the vector;
Data after cluster are divided into N number of data class by S4, according to the center vector of each data class, by investigation and mould
It is quasi- that pattern-recognition is carried out to each data class;N≥2;
S5 accurately launches advertisement according to the result of pattern-recognition in public transport.
Preferably, step S2 includes: and in the database extracts all data of same card number;Wherein extract
Data be the corresponding passenger of same card number all trips;Whole trips of the passenger are divided into N number of period, are obtained
The trip number of each period;The trip number of each period is quantified, related passenger is obtained in the corresponding period and goes out line number
The vector of amount.
Preferably, step S3 includes: to be clustered the vector based on K mean cluster algorithm.
Preferably, the passenger data includes cardholder information, consumption information, multiplied line information and a letter of riding
Breath.
It preferably, will be before the pretreated step of collected passenger data progress further include: by collected ridership
According to carrying out pretreatment screening;Wherein, the data filtered out include card number, Card Type, by bus date, riding time, multiplied line
Road, car number.
It preferably, include: that passenger data is acquired by subway gate and/or is taken pubic transport by bus card before step S1
Vehicle acquires passenger data.
It preferably, include: to be carried out in advance according to Shepard interpolative prediction algorithm to collected passenger data after step S1
It surveys.
It preferably, include: the then freshly harvested passenger when freshly harvested passenger data amount is greater than K between step S1 and S2
Passenger data in data cover step S1.
Another object of the present invention is realized by the following technical solution:
A kind of public transport advertisement delivery system based on passenger behavior mode, comprising: the database building being sequentially connected
Module, data processing module, k mean cluster analysis module, passenger behavior identification module and advertisement putting module;The database
Module is built, for pre-processing collected passenger data, and builds passenger's trip data library;The data processing
Module forms the vector of related passenger's trip for gathering the data in passenger's trip data library;The k mean value is poly-
Alanysis module, for clustering the vector;The passenger behavior identification module, for being divided into the data after cluster
N number of data class carries out pattern-recognition to each data class by investigating and simulating according to the center vector of each data class;N≥
2;The advertisement putting module, for accurately launching advertisement in public transport according to the result of pattern-recognition.
Preferably, data processing module includes: data extracting unit, Time segments division unit and quantifying unit;The data
Extraction unit, in the database extracting all data of same card number;The data wherein extracted are as same
All trips of the corresponding passenger of one card number;The Time segments division unit, it is N number of for being divided into whole trips of the passenger
Period obtains the trip number of each period;The quantifying unit is obtained for quantifying the trip number of each period
To in relation to passenger corresponding period trip quantity vector.
The present invention has the advantage that compared with the existing technology
This programme forms the vector of related passenger's trip by gathering the data in passenger's trip data library;It will
The vector is clustered;Data after cluster are divided into N number of data class, according to the center vector of each data class, through toning
It looks into and simulates and pattern-recognition is carried out to each data class;N≥2;Had according to the result of pattern-recognition according to the different demands of passenger
Put arrow launch advertisement, improve advertisement dispensing accuracy, improve the utilization rate of advertising resource;It can also be according to passenger's
The change of demand or travel amount and type and the time for changing advertisement dispensing, this is not only that advertising company, enterprise bring income
Raising, also meet the demand of passenger, killing three birds with one stone.
In addition, this programme carries out accurate timesharing to collected passenger data also according to Shepard interpolative prediction algorithm
Section divides the public bus network passenger flow estimation of passenger type, sufficiently combines the working day property on date, class hour property, festivals or holidays, day
The factors such as gas, compared to traditional neural network method and support vector machine method, the method precision of prediction with higher and
Reliability, while there is lower parameter dependence, so that reality can show the accuracy that advertisement is launched, it is also the passenger of public transport
The classification ways and means that precisely prediction provides.
Detailed description of the invention
Fig. 1 is the flow diagram of the public transport advertisement placement method of the invention based on passenger behavior mode.
Fig. 2 is the structural schematic diagram of the public transport advertisement delivery system of the invention based on passenger behavior mode.
Specific embodiment
Present invention will be further explained below with reference to the attached drawings and examples.
It is to improve the invention discloses a kind of public transport advertisement placement method and system based on passenger behavior mode
For the purpose of utilization rate is launched in public transport advertisement, using existing public transport advertisement input situation as background, the public transport of passenger is utilized
IC card brushing card data is accurate using being carried out based on the passenger behavior pattern-recognition of clustering and classification passenger flow forecast method
The composed structure of passenger flow at times prediction and trip characteristics analysis.By using accurate passenger flow estimation technology, public transport IC is utilized
Card brushing card data classifies to the behavior pattern of passenger, then predicts the volume of the flow of passengers composition of every kind of passenger type at times,
There is targetedly advertisement serving policy so as to be formed, in addition to this can be used for improving service standard, improve driving specification
Etc., it is final to realize the target for being conducive to the more excellent configuration of traffic resource.For example, being largely in passenger's composition of morning sessions
Go to work commuting crowd, can choose for this crowd and launches the advertisements such as product, real estate, finance product that disappear fastly, in addition to this, this
The time value and opportunity cost of one crowd is higher, and subway and transit operator can be considered using the diminution departure interval, fit
When improve speed scheme to meet passenger the needs of.
Referring to Fig. 1, a kind of public transport advertisement placement method based on passenger behavior mode, comprising:
S1 pre-processes collected passenger data, and builds passenger's trip data library;The passenger data packet
Include cardholder information, consumption information, multiplied line information and multiplied information of vehicles.Collected passenger data is pre-processed
The step of before further include: collected passenger data is subjected to pretreatment screening;Wherein, the data filtered out include that card is compiled
Number, Card Type, by bus date, riding time, multiplied route, car number.
It include: that passenger data is acquired by subway gate and/or is taken by bus card before the present embodiment, step S1
Bus acquires passenger data.
S2 gathers the data in passenger's trip data library, forms the vector of related passenger's trip;Specifically, it walks
Rapid S2 includes: in the database to extract all data of same card number;The data wherein extracted are same card
All trips of number corresponding passenger;Whole trips of the passenger are divided into N number of period, obtain the trip time of each period
Number;The trip number of each period is quantified, the vector in relation to passenger in corresponding period trip quantity is obtained.For example, example
It such as first can be divided into 8 periods one day, then repartition working day and festivals or holidays.In the database by all of same card number
Data extract, and the data extracted are all trips of some passenger in May.Then again this passenger's
All trip is divided into 16 periods of working day and festivals or holidays, has also just obtained the trip number of each period in this way.In order to
Different passengers is clustered, the trip record of each passenger is carried out vectorization by us.Each vector represents one
Passenger, vector have 16 elements, and each element has respectively represented that (period on working day 1 swipes the card number, and the period on working day 2 swipes the card time
Number, period on working day 3 swipe the card number, and the period on working day 4 swipes the card number, and the period on working day 5 swipes the card number, and the period on working day 6 brushes
Card number, period on working day 7 swipe the card number, and the period on working day 8 swipes the card number, and period festivals or holidays 1 swipes the card number, period festivals or holidays
2 swipe the card number, and period festivals or holidays 3 swipes the card number, and period festivals or holidays 4 swipes the card number, and period festivals or holidays 5 swipes the card number, festivals or holidays
Period 6 swipes the card number, and period festivals or holidays 7 swipes the card number, and period festivals or holidays 8 swipes the card number).
Between the present embodiment, step S1 and S2 include: when freshly harvested passenger data amount is greater than K, then it is freshly harvested
Passenger data in passenger data covering step S1.
S3 clusters the vector;Specifically, step S3 include: based on K mean cluster algorithm by the vector into
Row cluster.The process that the set of physics or abstract object is grouped into the multiple clusters being made of similar object is referred to as poly-
Class.By clustering the set that cluster generated is one group of data object, these objects and the object in the same cluster are similar to each other, with
Object in other clusters is different.What K mean cluster algorithm solved is that the set two containing n data point (entity) is divided into k
The problem of a class cluster.Algorithm randomly selects initial cluster center of the k number strong point as k class cluster, each data in set first
Point is divided into it among the class cluster where nearest cluster center, forms the initial distribution of k cluster.To distributing
Each class cluster calculate new cluster center, then proceed by data allocation process, such iteration several times after, if cluster center
Be no longer changed, then illustrate in the class cluster where data object is fully allocated to oneself, clustering criteria function convergence, otherwise after
It is continuous to be iterated process, until convergence.Here clustering criteria function is generally using cluster error sum of squares criterion function.K is equal
One feature of value clustering algorithm is exactly that will be adjusted to the distribution of all data points in iterative process each time, so
After recalculate cluster center, into iterative process next time, if the position of all data points does not have in certain iterative process
It changes, corresponding cluster center does not also change, and indicates that clustering criteria function has been restrained at this time, algorithm terminates.
Data after cluster are divided into N number of data class by S4, according to the center vector of each data class, by investigation and mould
It is quasi- that pattern-recognition is carried out to each data class;N≥2;After the cluster that have passed through kmeans, multiple small gather has been splitted data into
Class (data class), then according to export each small cluster the characteristics of, each group has the center vector of oneself, in conjunction with each
The case where center vector of group, by expert and appraises the investigation of group through discussion and simulates the mode knowledge carried out for each classification
Not, after classification, the new client (in the case where database degree of variation is relatively low) that the later period is added is also according to European
Range formula is identified, is included into the type divided.For example, can be by observing the flat of user in each cluster
Equal center vector obtains some basic features of the cluster, analyzes each parameter of these vectors and says the meaning represented to carry out
The identification of class of subscriber.It is qualitatively analyzed in conjunction with existing data first, by contacting the expert of relevant industries and department,
And an analysis generally is obtained for the sample investigation of passenger condition as a result, being then sampled investigation on the spot, pass through
It fills in questionnaires, and the problem of to questionnaire design and fills out the essential information of writer and carry out the processing of system and analysis obtains accordingly
Quantitative analysis results, comprehensive and quantitative analysis obtain last pattern classification with qualitative analysis, complete the identification work of mode.Finally
According to identification as a result, passing through the feature of occupation, the purpose of trip, the composition of family and social status income level etc.
Many aspects carry out the explanation and illustration for each classification, and expansion work is expanded in analysis of this data partly obtained for the later period
Make.
S5 accurately launches advertisement according to the result of pattern-recognition in public transport.
Include: in the present embodiment, after step S1 according to Shepard interpolative prediction algorithm to collected passenger data into
Row prediction.The algorithm model constructs main thinking: the first step extracts date attribute relevant to passenger flow, is quantified and is returned
One change processing, establishes the multidimensional property matrix for being directed to the date;Second step evaluate correlation between each attribute and passenger flow with it is quick
Perception extracts the dimension being effectively predicted, and assigns and weigh to each valid dimension;In the attribute matrix of third step after the pre-treatment,
On the basis of historical data, the period passenger flow of target date is predicted using Shepard interpolative prediction algorithm;Finally carry out
Prediction result quality evaluation.The prediction technique is more accurate and ineffectiveness more pre- than traditional statistics with the reasonability degree of prediction
Survey method is more preferable.Traditional statistics prediction technique is counted merely from the angle analysis passenger flow rule of data statistics
Prediction, forecast quality relies heavily on Quality of Statistical Data, thus such method precision is not high, and reliability is low.Traditional
Machine learning prediction technique improves precision of prediction and reliability, but all has model complexity, and parameter dependence is big, to training number
The defects of high according to mass dependence.And the interpolation prediction method is also to have precision using relatively more extensive a kind of prediction technique
Height, the small advantage of parameter dependence.Meanwhile interpolation prediction method has preliminary research in terms of forecasting traffic flow and achieves one
Fixed achievement, and different conditions, such as festivals or holidays, the period, in class hour, temperature, extreme weather etc. is for the different volumes of the flow of passengers, no
Crowd with trip characteristics has different influence degree, this is not available for the prediction model of the maturation such as time series analysis
Function, only by multi-dimensional interpolation prediction algorithm, the volume of the flow of passengers of the classifying type at times obtained in this way just can be more accurate, more accords with
Close reality.
Referring to fig. 2, the above-mentioned public transport advertisement placement method based on passenger behavior mode be applicable in based on passenger behavior
The public transport advertisement delivery system of mode, comprising: database building module, data processing module, the k mean value being sequentially connected are poly-
Alanysis module, passenger behavior identification module and advertisement putting module;The database building module, for multiplying collected
Objective data are pre-processed, and build passenger's trip data library;The data processing module is used for passenger's trip data library
In data gathered, form the vector of related passenger trip;The k mean cluster analysis module is used for the vector
It is clustered;The passenger behavior identification module, for the data after cluster to be divided into N number of data class, according to each data class
Center vector, through investigation and simulate to each data class carry out pattern-recognition;N≥2;The advertisement putting module, is used for
Advertisement is accurately launched in public transport according to the result of pattern-recognition.
In the present embodiment, data processing module includes: data extracting unit, Time segments division unit and quantifying unit;It is described
Data extracting unit, in the database extracting all data of same card number;The data wherein extracted are i.e.
For all trips of the corresponding passenger of same card number;The Time segments division unit, for dividing whole trips of the passenger
To N number of period, the trip number of each period is obtained;The quantifying unit, for by the trip number amount of progress of each period
Change, obtains the vector in relation to passenger in corresponding period trip quantity.
Above-mentioned specific embodiment is the preferred embodiment of the present invention, can not be limited the invention, and others are appointed
The change or other equivalent substitute modes what is made without departing from technical solution of the present invention, are included in protection of the invention
Within the scope of.
Claims (10)
1. a kind of public transport advertisement placement method based on passenger behavior mode characterized by comprising
S1 pre-processes collected passenger data, and builds passenger's trip data library;
S2 gathers the data in passenger's trip data library, forms the vector of related passenger's trip;
S3 clusters the vector;
Data after cluster are divided into N number of data class by S4, according to the center vector of each data class, by investigating and simulating pair
Each data class carries out pattern-recognition;N≥2;
S5 accurately launches advertisement according to the result of pattern-recognition in public transport.
2. the public transport advertisement placement method according to claim 1 based on passenger behavior mode, which is characterized in that step
Suddenly S2 includes:
All data of same card number are extracted in the database;The data wherein extracted are that same card number is corresponding
Passenger all trips;
Whole trips of the passenger are divided into N number of period, obtain the trip number of each period;
The trip number of each period is quantified, the vector in relation to passenger in corresponding period trip quantity is obtained.
3. the public transport advertisement placement method according to claim 1 based on passenger behavior mode, which is characterized in that step
Suddenly S3 includes:
The vector is clustered based on K mean cluster algorithm.
4. the public transport advertisement placement method according to claim 1 based on passenger behavior mode, which is characterized in that institute
Stating passenger data includes cardholder information, consumption information, multiplied line information and multiplied information of vehicles.
5. the public transport advertisement placement method according to claim 4 based on passenger behavior mode, which is characterized in that will
Collected passenger data carries out before pretreated step further include:
Collected passenger data is subjected to pretreatment screening;Wherein, the data filtered out include card number, Card Type, ride
Date, riding time, multiplied route, car number.
6. the public transport advertisement placement method according to claim 1 based on passenger behavior mode, which is characterized in that step
Include: before rapid S1
It acquires passenger data by subway gate and/or is taken bus by bus card and acquire passenger data.
7. the public transport advertisement placement method according to claim 1 based on passenger behavior mode, which is characterized in that step
Include: after rapid S1
Collected passenger data is predicted according to Shepard interpolative prediction algorithm.
8. the public transport advertisement placement method according to claim 1 based on passenger behavior mode, which is characterized in that step
Include: between rapid S1 and S2
When freshly harvested passenger data amount is greater than K, then the passenger data in freshly harvested passenger data covering step S1.
9. a kind of public transport advertisement delivery system based on passenger behavior mode characterized by comprising the number being sequentially connected
Module, data processing module, k mean cluster analysis module, passenger behavior identification module and advertisement putting module are built according to library;
The database building module for pre-processing collected passenger data, and builds passenger's trip data
Library;
The data processing module forms related passenger's trip for gathering the data in passenger's trip data library
Vector;
The k mean cluster analysis module, for clustering the vector;
The passenger behavior identification module, for the data after cluster to be divided into N number of data class, according to the center of each data class
Vector carries out pattern-recognition to each data class by investigating and simulating;N≥2;
The advertisement putting module, for accurately launching advertisement in public transport according to the result of pattern-recognition.
10. a kind of public transport advertisement delivery system based on passenger behavior mode, which is characterized in that data processing module packet
It includes: data extracting unit, Time segments division unit and quantifying unit;
The data extracting unit, in the database extracting all data of same card number;Wherein extract
Data be the corresponding passenger of same card number all trips;
The Time segments division unit obtains the trip of each period for whole trips of the passenger to be divided into N number of period
Number;
The quantifying unit obtains related passenger in corresponding period trip for quantifying the trip number of each period
The vector of quantity.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910128113.2A CN110020666B (en) | 2019-02-21 | 2019-02-21 | Public transport advertisement putting method and system based on passenger behavior mode |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910128113.2A CN110020666B (en) | 2019-02-21 | 2019-02-21 | Public transport advertisement putting method and system based on passenger behavior mode |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110020666A true CN110020666A (en) | 2019-07-16 |
CN110020666B CN110020666B (en) | 2023-06-20 |
Family
ID=67189036
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910128113.2A Active CN110020666B (en) | 2019-02-21 | 2019-02-21 | Public transport advertisement putting method and system based on passenger behavior mode |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110020666B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110533483A (en) * | 2019-09-05 | 2019-12-03 | 中国联合网络通信集团有限公司 | A kind of occupant classification method and system based on trip characteristics |
CN110956505A (en) * | 2019-12-04 | 2020-04-03 | 腾讯科技(深圳)有限公司 | Advertisement inventory estimation method and related device |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012088871A (en) * | 2010-10-19 | 2012-05-10 | Hitachi Solutions Ltd | Advertisement distribution system in means of public transportation |
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 |
WO2016045195A1 (en) * | 2014-09-22 | 2016-03-31 | 北京交通大学 | Passenger flow estimation method for urban rail network |
CN105718946A (en) * | 2016-01-20 | 2016-06-29 | 北京工业大学 | Passenger going-out behavior analysis method based on subway card-swiping data |
CN106951976A (en) * | 2016-10-12 | 2017-07-14 | 华南理工大学 | A kind of bus passenger flow Forecasting Methodology based on pattern classification |
KR20180061992A (en) * | 2016-11-30 | 2018-06-08 | 동의대학교 산학협력단 | System and Method for Transmitting target advertisement based on Customer in Public transportation |
CN108171970A (en) * | 2017-12-11 | 2018-06-15 | 北京奇虎科技有限公司 | A kind of method and device for obtaining passenger on public transport track |
CN108171533A (en) * | 2017-12-11 | 2018-06-15 | 北京奇虎科技有限公司 | Ads on Vehicles public bus network based on Intelligent bus card data recommends method and device |
-
2019
- 2019-02-21 CN CN201910128113.2A patent/CN110020666B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012088871A (en) * | 2010-10-19 | 2012-05-10 | Hitachi Solutions Ltd | Advertisement distribution system in means of public transportation |
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 |
WO2016045195A1 (en) * | 2014-09-22 | 2016-03-31 | 北京交通大学 | Passenger flow estimation method for urban rail network |
CN105718946A (en) * | 2016-01-20 | 2016-06-29 | 北京工业大学 | Passenger going-out behavior analysis method based on subway card-swiping data |
CN106951976A (en) * | 2016-10-12 | 2017-07-14 | 华南理工大学 | A kind of bus passenger flow Forecasting Methodology based on pattern classification |
KR20180061992A (en) * | 2016-11-30 | 2018-06-08 | 동의대학교 산학협력단 | System and Method for Transmitting target advertisement based on Customer in Public transportation |
CN108171970A (en) * | 2017-12-11 | 2018-06-15 | 北京奇虎科技有限公司 | A kind of method and device for obtaining passenger on public transport track |
CN108171533A (en) * | 2017-12-11 | 2018-06-15 | 北京奇虎科技有限公司 | Ads on Vehicles public bus network based on Intelligent bus card data recommends method and device |
Non-Patent Citations (1)
Title |
---|
彭舰等: "基于马尔科夫链的轻轨乘客轨迹预测新算法", 《电子科技大学学报》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110533483A (en) * | 2019-09-05 | 2019-12-03 | 中国联合网络通信集团有限公司 | A kind of occupant classification method and system based on trip characteristics |
CN110956505A (en) * | 2019-12-04 | 2020-04-03 | 腾讯科技(深圳)有限公司 | Advertisement inventory estimation method and related device |
CN110956505B (en) * | 2019-12-04 | 2021-03-16 | 腾讯科技(深圳)有限公司 | Advertisement inventory estimation method and related device |
Also Published As
Publication number | Publication date |
---|---|
CN110020666B (en) | 2023-06-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106846538B (en) | Cross vehicle record treating method and apparatus | |
CN107045788B (en) | Traffic road condition prediction method and device | |
Echaniz et al. | Modelling perceived quality for urban public transport systems using weighted variables and random parameters | |
CN106897420B (en) | Mobile phone signaling data-based user travel resident behavior identification method | |
Förster et al. | Data-driven identification of characteristic real-driving cycles based on k-means clustering and mixed-integer optimization | |
Marcucci et al. | Regional airport choice: Consumer behaviour and policy implications | |
CN107633035B (en) | Shared traffic service reorder estimation method based on K-Means and LightGBM model | |
CN105488046A (en) | Big data analysis system based on vehicle insurance services | |
CN108062857B (en) | Prediction technique for cab-getter's trip purpose | |
CN109002492B (en) | Performance point prediction method based on LightGBM | |
KR20160071990A (en) | Customer data analysis and verification system | |
CN106779808A (en) | Consumer space's behavior analysis system and method in a kind of commercial circle | |
CN107392389A (en) | Taxi dispatching processing method based on ARIMA models | |
CN111784022A (en) | Short-time adjacent fog prediction method based on combination of Wrapper method and SVM method | |
CN111899059A (en) | Navigation driver revenue management dynamic pricing method based on block chain | |
CN110020666A (en) | A kind of public transport advertisement placement method and system based on passenger behavior mode | |
CN109243173A (en) | Track of vehicle analysis method and system based on road high definition bayonet data | |
CN111680382A (en) | Grade prediction model training method, grade prediction device and electronic equipment | |
CN107274066B (en) | LRFMD model-based shared traffic customer value analysis method | |
CN111353094A (en) | Information pushing method and device | |
CN110852792B (en) | Route value evaluation method based on big data analysis and related products | |
CN110459050B (en) | Short-term bus passenger flow prediction method based on mixed decision tree | |
Chen et al. | RFM model and K-means clustering analysis of transit traveller profiles: A case study | |
CN112949926B (en) | Income maximization ticket amount distribution method based on passenger demand re-identification | |
CN112836996B (en) | Method for identifying potential ticket buying demand of passenger |
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 |