CN106651027A - Internet regular bus route optimization method based on social network - Google Patents
Internet regular bus route optimization method based on social network Download PDFInfo
- Publication number
- CN106651027A CN106651027A CN201611187726.6A CN201611187726A CN106651027A CN 106651027 A CN106651027 A CN 106651027A CN 201611187726 A CN201611187726 A CN 201611187726A CN 106651027 A CN106651027 A CN 106651027A
- Authority
- CN
- China
- Prior art keywords
- cluster
- data
- point
- user
- working
- 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
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000005457 optimization Methods 0.000 title claims abstract description 8
- 238000013480 data collection Methods 0.000 claims abstract description 4
- 238000003064 k means clustering Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000002123 temporal effect Effects 0.000 claims description 5
- 239000012141 concentrate Substances 0.000 claims description 4
- 238000009412 basement excavation Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000011524 similarity measure Methods 0.000 claims description 2
- 239000004744 fabric Substances 0.000 claims 1
- 230000003993 interaction Effects 0.000 claims 1
- 238000005065 mining Methods 0.000 abstract 2
- 238000013461 design Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 241001520299 Phascolarctos cinereus Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000010009 beating Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G06Q50/40—
Abstract
The invention discloses an internet regular bus route optimization method. The method comprises the steps: step I, a social software data collection step, wherein social software data collection includes acquiring information data of social software and collecting information related to a regular bus route from the information data; step II, mining the data collected in the data collection step to obtain OD (Origin to Destination) points of the user at different outgoing time periods; step III, mining a target user relationship, after a target user is found, looking for a user having the similar outgoing law with the target user according to the friend relationship of a microblog, and further broadening the range of the target user; and step IV, clustering a target user group to obtain a final clustering center (uk, rk) of the user group, wherein the value can be used as the OD points of the internet regular bus route. The potential target users are mined by use of massive data in a social network, the method can help the user to make an individualized outgoing scheme in accordance with the practical requirement of the user, and the scheme is easily accepted by the user.
Description
Technical field
The present invention relates to field of traffic.Set in particular to a kind of internet regular service route optimization based on social networks
Meter method.
Background technology
Along with the popularization at full speed and development of information technology especially internet, Public Transport Service also rise to one it is new
Level.The appearance of the O2O taxi-hailing softwares with excellent step (UBER), drop drop as representative is very easy to the work and life of people
It is living.In the line city of Beijing, Shanghai etc., because go off daily is far away, public transport is crowded to capacity, people are remained further need for than beating
Car is less expensive, shuttle bus service more comfortable than public transport, subway, and then drop drop bus, hello colleague, koala regular bus etc. are large quantities of mutual
Networking regular bus arises at the historic moment.But, internet regular bus of the prior art is high, steady in a long-term due to circuit repetition, empty-seat rate
Traveller is very few, causes operation cost and remains high, and many operators are hard to carry on.Additionally, each road class for occurring suddenly
Car also creates the heavy congestion in many focus sections, has run counter to the original intention of its Green Travel advocated.
Cause the reason for difficulty is matched between internet regular bus and user's request can be attributed to it is following some：
(1) abundant digging user demand is failed, the reason for the trip rule for grasping user is most critical.Internet regular bus pair
The work of city dweller, living condition lack necessary statistics, general only according to general knowledge and experience in quality design regular bus operation line
Road, considers not enough to the individual demand of user's trip, and causing a lot " minority's demands " cannot cover.In fact, in urban human
Mouthful continuous today of expansion, even " minority's demand " also not minority in absolute quantity.And meet the energy of " minority's demand "
Power, the exactly mark of shuttle bus service level.In addition, internet regular bus is often strict in accordance with " late towards nine at runtime
Five ", do not consider the situation of certain user Jing often overtime works, therefore can be lost in a part of potential customers yet.
(2) circuit site design granularity is too big, does not provide fine point-to-point services.Many internets regular bus is in design lines
Granularity is excessively thick during road, and the scope that beginning and end is covered is too big, causes user's inconvenience by bus, therefore lost part use
Family.For example hot topic regular service route Back Long View in Beijing is to Shangdi, it is commonly known that Back Long View and Shangdi are the large-scale communities in two, Beijing,
Area coverage is very wide, and although many users belong to this region, but far apart from regular bus point, and this circuit will not be selected naturally.
Only accomplish that regular bus point is all close from the place on and off duty of user, such circuit just can receive an acclaim.
(3) the effect of publicity is undesirable, and user's participation is very low.Although internet regular bus carries out daily publicity by internet
And operation, but do not give full play to the propagation advantage of internet.Regular bus operator is made after circuit, by mobile phone A PP
Issuing route allows user to register, and decides whether to open circuit according to the registration situation of user.But due to APP population coverage not
Height, therefore this is also lost in most potential user.And these APP do not have with the society of main flow mostly without social functions yet
Hand over software effectively integrated, even if therefore have good circuit, its social propagation efficiency is greatly reduced.
Summary, internet regular bus fail at present effectively solving user ride pain spot the reason for, do not lie in demand deficiency, and
Demand is excavated in effective channel is lacked, demand is matched, demand is met.Internet regular bus wants the pain by bus of effectively solving user
Point, must just find a kind of method, and user's request is captured exactly, according to user's request come personalized formulation rational routes, enter
One step improves the level of service, here it is the background of this patent is located.
The content of the invention
This patent is based on the above-mentioned situation of prior art and proposes, this patent technical problem to be solved is to provide
A kind of internet regular bus route optimization method based on social networks, it is potential in order to accurately finding and positioning internet regular bus
User's request.
In order to solve the above problems, this patent is provided：
A kind of internet regular bus route optimization method, methods described includes：
Step one, social software data collection steps, the social software data acquisition includes obtaining the letter of social software
Breath data, and therefrom collect the information relevant with regular service route.In this step, arranging crawl condition by way of from obtaining
The open interface that the social software is provided is taken, the relevant information data in the microblogging issued is captured；The condition includes geography
Positional information, temporal information, friend's incidence relation.Such as, in geographical location information, Beijing can be set；The data of crawl
Temporal information in, the data volume of crawl is at least one month, it is possible to constantly captures in real time and updates.
Step 2, the data that the data collection steps are collected are excavated, obtain user's different trip periods
OD points (terminus).In this step, the relevant information data for collecting is carried out using K-means clustering algorithms
Excavate；It is in this step vector by i-th data definition collected in the data：xi=(xi1, xi2, xi3), wherein
xiRepresent vector, xi1Represent i-th point of longitude, xi2Represent i-th point of latitude, xi3Represent i-th point of time；In meter
Calculate and classified first with the time before the cluster centre of place, according to issuing time x of the social softwarei3By Segmentation of Data Set
For working collection and next collection；Work as xi3∈ (5, when 9), by xiPoint puts working collection into and carries out place cluster, works as xi3∈ (16, when 20),
By xiPoint puts next concentration into carries out place cluster, and remainder data point is considered as Null Spot and is filtered；This step is realized tentatively
Data filtering with classification, be conducive to the cluster of subsequent step；Then first time place cluster is carried out, be on duty respectively Ji Hexia
Class concentrates and respectively selects 3 points alternately point, that is, determine cluster centre number k=3；K-th place cluster of first definition working collection
Center is：
Wherein, distribution coefficientThe meaning of the formula is will to take part in k-th
The calculation of longitude ＆ latitude mean value of the point of working collection place cluster, cluster centre of the longitude and latitude for calculating as k-th working collection
Coordinate；The n is the number at gathered significant figure strong point；
Define i-th point xiWith k-th cluster centre μkThe distance between be：
Dik=(xi-μk)T(xi-μk)
Then the calculation procedure of the first time place cluster includes working clustering and next clustering, the working collection
Cluster includes：(1) first random initializtion cluster mean μ1、μ2、μ3；(2) to each point xiAll finding makes DikMinimum k, by i points
Ju Daogai centers, and distribution coefficient z is setik=1：(3) if all of zikIt is not changed in last iteration, then stops
Cluster, exports μ1、μ2、μ3；(4) otherwise μ is updated according to 1. formula1、μ2、μ3；With can be with the same method of the working clustering
Obtain the three cluster centre ρ for collecting that come off duty1、ρ2、ρ3；
Step 3, targeted customer's relation excavation, after targeted customer is found, according to the friends of its microblogging, find
The user similar to its rule of going on a journey, so as to further expand the scope of targeted customer.The circle of friends relation of traversal targeted customer,
The all emphasis friends for meeting predetermined condition are found out, the working clustering center μ of each friend s is then calculateds=(μ1、
μ2、μ3) and next clustering center ρs=(ρ1、ρ2、ρ3)；Then cosine similarity can define working collection circuit Similarity Measure
Each emphasis friend s and targeted customer d (μd、ρd) circuit similarity：
The cos θ are threshold value, if cos θ are more than predetermined numerical value, then it is assumed that the circuit phase of friend s and targeted customer d
Seemingly, all similar friends are put in a new set, carries out the secondary cluster of the 4th step；
Step 4, targeted user population is clustered, obtain the final cluster centre (μ of customer groupk, ρk), this value
As the OD points of internet regular service route.In this step, using K-means clustering algorithms to the correlation for collecting
Information data is excavated；(μ, the ρ) for the customer group that step 3 is obtained is divided into two set according to working collection with next collection, point
Ji Suan not two cluster points；Because the similarity through step 3 judges, so cluster centre number is set in this step as 1,
If μ=(μ1, μ2..., μ3n), as k=1, can release final working clustering center according to the formula of step 2 is：
Next clustering center can be calculated in the same manner is：
The OD points for finally giving k-th customer group are (μk, ρk), to set up internet regular service route according to this point.
It is an advantage of the current invention that：
1st, the present invention is based on social networks, and using the mass data in social networks potential targeted customer is excavated, and finds
Meet the regular bus route of user's request and be pushed to user, the behavior that this active is close to the users, than merely by passenger's registration
Passive approach, can more effectively excavate potential user and solve user and ride pain spot, while Consumer's Experience is more preferable.
2nd, social networks can provide abundant user context information：Such as user job, the place of life, when on and off duty
Between, or even the hobby including user etc..According to these information, we can help, and user is customized more to meet user's actual need
Personalized trip scheme, it is easy to be easily accepted by a user.
Description of the drawings
Fig. 1 is a kind of original of the internet regular service route optimization method based on social networks in the specific embodiment of the invention
Reason figure；
Fig. 2 is the flow chart of the internet regular service route method for optimizing based on social networks of the present invention.
Specific embodiment
Specific implementation of the patent mode is described in detail below in conjunction with the accompanying drawings.It is pointed out that the concrete reality
It is only the citing to this patent optimal technical scheme to apply mode.The restriction to this patent protection domain can not be interpreted as.Its
Purpose is that the present invention is described in further detail, to make those skilled in the art to implement according to this with reference to specification.
As shown in Figure 1 and Figure 2.It is excellent that this specific embodiment provides a kind of internet regular service route based on social networks
Change method, methods described comprises the steps：
Step one, social software data collection steps, the social software data acquisition includes obtaining the letter of social software
Breath data, and therefrom collect the information relevant with regular service route.
In this step, by taking microblog data as an example, it is possible to use in the microblogging that the open interface crawl that microblogging is provided is issued
Relevant information data.Above-mentioned data can be obtained by way of setting crawl condition, the condition includes geographical position
Information, temporal information, friend's incidence relation etc..
Such as, in geographical location information, Beijing can be set；In the temporal information of the data of crawl, the data of crawl
Amount is at least one month, it is possible to constantly captures in real time and updates.
Step 2, the data that the data collection steps are collected are excavated, obtain user's different trip periods
OD points (terminus).
In this step, the data for collecting are excavated using K-means clustering algorithms.The K-means
Clustering algorithm includes：K-means algorithms be very typically based on distance clustering algorithm, using distance as similitude evaluation
Index, that is, think that the distance of two objects is nearer, and its similarity is bigger.The algorithm thinks that cluster is by apart from close object group
Into, therefore using obtaining compact and independent cluster as final goal.
The selection of k initial classes cluster centre point has large effect to cluster result, because in the algorithm first step
In be any k object of random selection as the center of initial clustering, initially represent a cluster.The algorithm is in each iteration
In remaining each object is concentrated to data, each object is assigned to again according to itself and the distance at each cluster center nearest
Cluster.After all data objects have been investigated, an iteration computing is completed, and new cluster centre is computed.If once
Before and after iteration, the value of distance does not change, and illustrates that algorithm has been restrained.
In this example, i-th data for collecting can be defined as vector：xi=(xi1, xi2, xi3) (bold-type letter is represented
Vector, similarly hereinafter), wherein xi1Represent i-th point of longitude, xi2Represent i-th point of latitude, xi3Represent i-th point of time.
Can be classified first with the time before place cluster centre is calculated, according to xi3I.e. the issuing time of microblogging divides data set
It is segmented into working collection and next collection.5 are concentrated in view of the work hours of most people:00～9:00, the quitting time concentrates on 16:
00～20:00, so working as xi3∈ (5, when 9), by xiPoint puts working collection into and carries out place cluster, works as xi3∈ (16, when 20), by xi
Point puts next concentration into carries out place cluster, and remainder data point is considered as Null Spot and is filtered.This step realizes preliminary number
According to filtering and classification, be conducive to the cluster of subsequent step.
Followed by first time place cluster, it is on duty to collect and come off duty to concentrate respectively and respectively selects 3 points alternately point,
Determine cluster centre number k=3.First defining the k-th place cluster centre for collecting of going to work is：
Wherein, distribution coefficientThe meaning of the formula is to take part on k-th
The calculation of longitude ＆ latitude mean value of the point of Ban Ji places cluster, the longitude and latitude for calculating is used as the cluster centre of k-th working collection
Coordinate.
Then i-th point x is definediWith k-th cluster centre μkThe distance between be：
Dik=(xi-μk)T(xi-μk)
Calculation procedure：
(1) first random initializtion cluster mean μ1、μ2、μ3；
(2) to each point xiAll finding makes DikMinimum k, by i Dian Judaogai centers, and arranges distribution coefficient zik=1；
(3) if all of zikIt is not changed in last iteration, then stops cluster, exports μ1、μ2、μ3；
(4) otherwise μ is updated according to 1. formula1、μ2、μ3；
Can obtain coming off duty the three cluster centre ρ for collecting with same method1、ρ2、ρ3。
The function of data screening can have both been realized with above-mentioned k-means clustering algorithms, it is also possible to realize the work(of data clusters
Can, cluster centre number k is set to into 3 can avoid impact of the discrete point to cluster centre, for example, due to mobile phone positioning precision limit
System if the user while has also sent out several microbloggings, at this moment coordinate is possible to greatly deviate residence, so having when going window-shopping
It is necessary that cluster centre number is set greater than into 1, and excessive cluster centre can be produced if k settings are excessive, it is unfavorable for next
The secondary cluster of step.It is relatively more effective in this patent when test of many times finds that k takes 3, based on the side that above-mentioned first time clusters
Method is conducive to excavating the real place of abode of user and job site, is that second cluster of next step has carried out data standard
It is standby.
Step 3, targeted customer's relation excavation, after targeted customer is found, according to the friends of its microblogging, find
The user similar to its rule of going on a journey, so as to further expand the scope of targeted customer.
Because social software is a big social media, such as microblogging, the present embodiment targeted customer is on its microblogging
Friend, it is likely that be exactly him in real-life relatives, friend, colleague, it is likely that and targeted customer has similar trip
Rule, have ready conditions the targeted customer for becoming new.Can relatively rapid determine multiple targets by way of excavating targeted customer
User, efficiency so can be saved for the calculating of magnanimity is carried out and the accuracy of identification is improved.
Further, because the targeted customer in the present embodiment may have many friends in microblogging, if being directed to its friend
Friend analyzes one by one its trip rule, can equally bring a large amount of calculating.Therefore it is excellent in the present embodiment so when its circle of friends is traveled through
Selection of land, focal selection meets the social software data of " the emphasis friend User " of following condition and is analyzed, and to improve it is found
The operation efficiency of his targeted customer:Condition 1, and targeted customer is mutual concern relation；Condition 2, and targeted customer is in microblogging
It is frequently interactive；Condition 3, and targeted customer has common friend.It can be the pass of sum between conditions above 1, condition 2, condition 3
System, or or relation, it is also possible to selected section or be all combined in three conditions.
The circle of friends relation of traversal targeted customer, finds out all emphasis friends for meeting above-mentioned condition, then calculates every
The working clustering center μ of one friend ss=(μ1、μ2、μ3) and next clustering center ρs=(ρ1、ρ2、ρ3).Then calculate every
One emphasis friend s and targeted customer d (μd、ρd) circuit similarity.Because cosine similarity has computing side in sorting algorithm
Just the characteristics of, effect is obvious, so working collection circuit similarity can be defined according to cosine similarity：
If cos θ are more than certain threshold value (this value can be obtained by experiment), then it is assumed that the circuit phase of friend s and targeted customer d
Seemingly, all similar friends are put in a new set, carries out the secondary cluster of the 4th step.
Step 4, targeted user population is clustered, obtain the final cluster centre (μ of customer groupk, ρk), this value
As the OD points of internet regular service route.
(μ, the ρ) for the customer group that step 3 is obtained is divided into two set according to working collection with next collection, and two are calculated respectively
Individual cluster point.Because the similarity through step 3 judges that, so the data in the step can be concentrated more, abnormity point is less,
So it is 1 that can reduce cluster centre number, so as to reduce the cost of decision making of enterprise.If μ=(μ1, μ2..., μ3n), work as k=1
When, can release final working clustering center according to the formula of step 2 is：
Next clustering center can be calculated in the same manner is：
The OD points for finally giving k-th customer group are (μk, ρk), enterprise can set up internet regular bus according to this point
Circuit.
Claims (3)
1. a kind of internet regular service route optimization method based on social networks, it is characterised in that methods described includes following step
Suddenly：
Step one, social software data collection steps, the social software data acquisition includes obtaining the Information Number of social software
According to, and therefrom collect the information relevant with regular service route.
In this step, from the open interface that the social software is provided is obtained by way of setting crawl condition, crawl is sent out
Relevant information data in the microblogging of cloth；The condition includes geographical location information, temporal information, friend's incidence relation.
Step 2, the data that the data collection steps are collected are excavated, obtain the OD points of user's different trip periods
(terminus).
In this step, the relevant information data for collecting is excavated using K-means clustering algorithms；At this
In step by i-th data definition collected in the data for vector：
xi=(xi1, xi2, xi3), wherein xiRepresent vector, xi1Represent i-th point of longitude, xi2Represent i-th point of latitude, xi3
Represent i-th point of time；
Classified first with the time before place cluster centre is calculated, according to issuing time x of the social softwarei3By number
Working collection and next collection are divided into according to collection；Work as xi3∈ (5, when 9), by xiPoint puts working collection into and carries out place cluster, works as xi3∈
(16, when 20), by xiPoint puts next concentration into carries out place cluster, and remainder data point is considered as Null Spot and is filtered；This step
Preliminary data filtering and classification are realized, is conducive to the cluster of subsequent step；
Then first time place cluster is carried out, is on duty to collect and come off duty to concentrate respectively and is respectively selected 3 points alternately point, that is, determined
Cluster centre number k=3；First defining the k-th place cluster centre for collecting of going to work is：
Wherein,The meaning of the formula is will to take part in k-th working collection
The calculation of longitude ＆ latitude mean value of the point of place cluster, coordinate of the longitude and latitude for calculating as the cluster centre of k-th working collection；
The n is the number at gathered significant figure strong point；
Define i-th point xiWith k-th cluster centre μkThe distance between be：
Then the calculation procedure of the first time place cluster includes working clustering and next clustering, the working clustering
Including：(1) first random initializtion cluster mean μ1、μ2、μ3；(2) to each point xiAll finding makes DikMinimum k, i points are gathered
The center, and distribution coefficient z is setik=1；(3) if all of zikIt is not changed in last iteration, then stops cluster,
Output μ1、μ2、μ3；(4) otherwise μ is updated according to 1. formula1、μ2、μ3；
Can obtain coming off duty the three cluster centre ρ for collecting with the method same with the working clustering1、ρ2、ρ3；
Step 3, targeted customer's relation excavation, after targeted customer is found, according to the friends of its microblogging, find and it
The similar user of trip rule, so as to further expand the scope of targeted customer.
The circle of friends relation of traversal targeted customer, finds out all emphasis friends for meeting predetermined condition, then calculates each
The working clustering center μ of friend ss=(μ1、μ2、μ3) and next clustering center ρs=(ρ1、ρ2、ρ3)；Then cosine similarity
Working collection circuit Similarity Measure each emphasis friend s and targeted customer d (μ can be definedd、ρd) circuit similarity：
The cos θ are threshold value, if cos θ are more than predetermined numerical value, then it is assumed that friend s is similar with the circuit of targeted customer d, will
All similar friends are put in a new set, carry out the secondary cluster of the 4th step；
Step 4, targeted user population is clustered, obtain the final cluster centre (μ of customer groupk, ρk), this value can be used as
The OD points of internet regular service route.
In this step, the relevant information data for collecting is excavated using K-means clustering algorithms；Will step
(μ, ρ) of rapid three customer groups for obtaining is divided into two set according to working collection with next collection, and two cluster points are calculated respectively；Due to
Judge through the similarity of step 3, so cluster centre number is set in this step as 1, if μ=(μ1, μ2..., μ3n), when
During k=1, can release final working clustering center according to the formula of step 2 is：
Next clustering center can be calculated in the same manner is：
The OD points for finally giving k-th customer group are (μk, ρk), to set up internet regular service route according to this point.
2. method according to claim 1, it is characterised in that the social networks includes microblogging.
3. the method told according to claim 1, it is characterised in that in the step 3, the predetermined condition includes：Condition
1, and targeted customer is mutual concern relation；Condition 2, and targeted customer's frequently interaction in microblogging；Condition 3, and targeted customer
There is common friend.Can be between conditions above 1, condition 2, condition 3 sum relation, or or relation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611187726.6A CN106651027B (en) | 2016-12-21 | 2016-12-21 | Internet regular bus route optimization method based on social network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611187726.6A CN106651027B (en) | 2016-12-21 | 2016-12-21 | Internet regular bus route optimization method based on social network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106651027A true CN106651027A (en) | 2017-05-10 |
| CN106651027B CN106651027B (en) | 2021-08-20 |
Family
ID=58835167
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611187726.6A Active CN106651027B (en) | 2016-12-21 | 2016-12-21 | Internet regular bus route optimization method based on social network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106651027B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109238271A (en) * | 2018-08-23 | 2019-01-18 | 中国科学院电子学研究所苏州研究院 | A kind of time-based line fitting method |
| CN109858567A (en) * | 2019-03-06 | 2019-06-07 | 上海赢科信息技术有限公司 | The judgment method and system of the trip form of car owner |
| CN109919354A (en) * | 2019-01-15 | 2019-06-21 | 深圳大学 | A kind of share-car optimization method, index structure and system based on social big data |
| CN109978209A (en) * | 2017-12-28 | 2019-07-05 | 北京京东尚科信息技术有限公司 | Line design method and device |
| CN110009175A (en) * | 2018-12-25 | 2019-07-12 | 阿里巴巴集团控股有限公司 | The performance estimating method and device of OD demand analysis algorithm |
| CN110598755A (en) * | 2019-08-19 | 2019-12-20 | 北京建筑大学 | OD flow clustering method based on vector constraint |
| CN112132236A (en) * | 2020-11-20 | 2020-12-25 | 深圳市城市交通规划设计研究中心股份有限公司 | Demand subarea dividing and line planning method and device based on clustering algorithm |
| CN112447054A (en) * | 2019-08-30 | 2021-03-05 | 比亚迪股份有限公司 | Method and apparatus for controlling vehicle travel |
| CN112948663A (en) * | 2019-12-10 | 2021-06-11 | 京东数字科技控股有限公司 | Information pushing method and device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2772894A1 (en) * | 2013-02-28 | 2014-09-03 | HERE Global B.V. | Method and Apparatus for Automated Service Schedule Derivation and Updating |
| CN104811367A (en) * | 2015-04-02 | 2015-07-29 | 常青竹（北京）科技有限公司 | Data communication connecting method by gathering geographic positions |
| CN105117595A (en) * | 2015-08-19 | 2015-12-02 | 大连理工大学 | Floating car data based private car travel data integration method |
| CN105243128A (en) * | 2015-09-29 | 2016-01-13 | 西华大学 | Sign-in data based user behavior trajectory clustering method |
-
2016
- 2016-12-21 CN CN201611187726.6A patent/CN106651027B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2772894A1 (en) * | 2013-02-28 | 2014-09-03 | HERE Global B.V. | Method and Apparatus for Automated Service Schedule Derivation and Updating |
| CN104811367A (en) * | 2015-04-02 | 2015-07-29 | 常青竹（北京）科技有限公司 | Data communication connecting method by gathering geographic positions |
| CN105117595A (en) * | 2015-08-19 | 2015-12-02 | 大连理工大学 | Floating car data based private car travel data integration method |
| CN105243128A (en) * | 2015-09-29 | 2016-01-13 | 西华大学 | Sign-in data based user behavior trajectory clustering method |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109978209B (en) * | 2017-12-28 | 2022-04-12 | 北京京东尚科信息技术有限公司 | Line design method and device |
| CN109978209A (en) * | 2017-12-28 | 2019-07-05 | 北京京东尚科信息技术有限公司 | Line design method and device |
| CN109238271A (en) * | 2018-08-23 | 2019-01-18 | 中国科学院电子学研究所苏州研究院 | A kind of time-based line fitting method |
| CN109238271B (en) * | 2018-08-23 | 2020-08-14 | 中国科学院电子学研究所苏州研究院 | Line fitting method based on time |
| CN110009175A (en) * | 2018-12-25 | 2019-07-12 | 阿里巴巴集团控股有限公司 | The performance estimating method and device of OD demand analysis algorithm |
| WO2020147597A1 (en) * | 2019-01-15 | 2020-07-23 | 深圳大学 | Social big data-based carpool optimization method, index structure and system |
| CN109919354A (en) * | 2019-01-15 | 2019-06-21 | 深圳大学 | A kind of share-car optimization method, index structure and system based on social big data |
| CN109919354B (en) * | 2019-01-15 | 2022-04-22 | 深圳大学 | Car pooling optimization method, index structure and system based on social big data |
| CN109858567A (en) * | 2019-03-06 | 2019-06-07 | 上海赢科信息技术有限公司 | The judgment method and system of the trip form of car owner |
| CN110598755A (en) * | 2019-08-19 | 2019-12-20 | 北京建筑大学 | OD flow clustering method based on vector constraint |
| CN110598755B (en) * | 2019-08-19 | 2022-07-15 | 北京建筑大学 | OD flow clustering method based on vector constraint |
| CN112447054A (en) * | 2019-08-30 | 2021-03-05 | 比亚迪股份有限公司 | Method and apparatus for controlling vehicle travel |
| CN112948663A (en) * | 2019-12-10 | 2021-06-11 | 京东数字科技控股有限公司 | Information pushing method and device, equipment and storage medium |
| CN112132236A (en) * | 2020-11-20 | 2020-12-25 | 深圳市城市交通规划设计研究中心股份有限公司 | Demand subarea dividing and line planning method and device based on clustering algorithm |
| CN112132236B (en) * | 2020-11-20 | 2021-03-26 | 深圳市城市交通规划设计研究中心股份有限公司 | Demand subarea dividing and line planning method and device based on clustering algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106651027B (en) | 2021-08-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106651027A (en) | Internet regular bus route optimization method based on social network | |
| CN107134142B (en) | A kind of urban road method for predicting based on multisource data fusion | |
| CN107241512B (en) | Intercity Transportation trip mode judgment method and equipment based on data in mobile phone | |
| CN104484993B (en) | Processing method of cell phone signaling information for dividing traffic zones | |
| CN106912015B (en) | Personnel trip chain identification method based on mobile network data | |
| CN106447573A (en) | Spatial accessibility analysis method and device based on public facility differences | |
| CN105206048B (en) | A kind of city dweller colony transfer mode based on traffic OD data finds system and method | |
| Liu et al. | Understanding intra-urban trip patterns from taxi trajectory data | |
| JP5214810B2 (en) | POSITION INFORMATION ANALYSIS DEVICE AND POSITION INFORMATION ANALYSIS METHOD | |
| CN110111574B (en) | Urban traffic imbalance evaluation method based on flow tree analysis | |
| CN104809112B (en) | A kind of city bus development level integrated evaluating method based on multi-source data | |
| CN109583640A (en) | A kind of Urban Traffic passenger flow attribute recognition approach based on multi-source location data | |
| CN108564226A (en) | A kind of public bus network optimization method based on taxi GPS and mobile phone signaling data | |
| CN110176141B (en) | Traffic cell division method and system based on POI and traffic characteristics | |
| CN105142106A (en) | Traveler home-work location identification and trip chain depicting method based on mobile phone signaling data | |
| CN108053240B (en) | Method and device for generating vehicle-mounted advertisement delivery bus route scheme | |
| CN110796337B (en) | System for evaluating service accessibility of urban bus stop | |
| CN110716935A (en) | Track data analysis and visualization method and system based on online taxi appointment travel | |
| CN109978224A (en) | A method of analysis obtains the Trip Generation Rate of heterogeneity building | |
| CN102609616A (en) | Dynamic population distribution density detecting method based on mobile phone positioning data | |
| CN107656987A (en) | A kind of subway station function method for digging based on LDA models | |
| CN108307316B (en) | User's high-precision method of discrimination is fast moved based on S1-MME data | |
| CN105509758A (en) | Route planning method and system based on mobile terminal and mobile terminal | |
| CN109389243A (en) | In conjunction with the freeway traffic condition survey website distribution method of networked fee collection data | |
| CN110263109A (en) | A kind of family's amount evaluation method merging internet information and GIS technology |
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 |