CN104376327B - A kind of clustering method of public bicycles lease point - Google Patents
A kind of clustering method of public bicycles lease point Download PDFInfo
- Publication number
- CN104376327B CN104376327B CN201410616030.5A CN201410616030A CN104376327B CN 104376327 B CN104376327 B CN 104376327B CN 201410616030 A CN201410616030 A CN 201410616030A CN 104376327 B CN104376327 B CN 104376327B
- Authority
- CN
- China
- Prior art keywords
- point
- lease
- curve
- coding
- vehicle
- 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.)
- Expired - Fee Related
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
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of vehicle clustering method of public bicycles lease point, belong to public transport cluster field.This method comprises the following steps：Pass through the described lease point indicatrix that vehicle number changes in one day and curve is split according to specific segmentation demand, the encoded radio of this section is determined using the coding and quantization method of curve in divided each section, corresponding similarity is calculated by similarity function again, lease point the most similar is classified as by a class using clustering method according to the value of similarity.The invention enables the cluster of public bicycles lease vertex type is more convenient, the lease point distribution effectively alleviated and the unreasonable problem sorted out, also the present situation of " borrowing car difficult; difficulty of returning the car " is preferably alleviated, the service level of public bicycles system, people are improved for the total satisfactory grade of public bicycles and the utilization rate of public bicycles.
Description
Technical field
This method belong to public transport cluster field, can be applied to public bicycles lease vertex type cluster, with faster,
The lease vertex type of public bicycles is clustered with easily mode, it is proposed that the lease based on coding and similarity function
Point clustering method.
Background technology
Although urban public transport is quickly grown, people's trip increasingly facilitates, either subway station or bus station,
A segment distance is always had from our destinations to be gone.This segment distance is often neither long nor short, calls a taxi uneconomical, transfer public transport less
It is convenient, walk again a little remote, i.e., so-called " last one kilometer " problem.Therefore, public bicycles are a kind of preferable solution party
Formula.Meanwhile, the call of the Green Travel proposed in response to government, the public bicycles lease service based on general public is just
Each big city at home and abroad is quietly risen, and bicycle storing, land occupation is few when using, it is possible to increase path resource
Utilization rate, alleviates congestion in road, can promote energy-saving and emission-reduction again, reduce tail gas pollution, moreover it is possible to physical fitness, raising city quality.
Therefore, public bicycles leasing market is cultivated, public bike renting system is realized, is to alleviate traffic pressure, reduces environment dirty
The effective way of dye, is also to solve one of good method of citizens' activities.
However, as public bicycles system scale gradually increases, frequency of use gradually increases, and gives public bicycles system
Management and service also bring a series of problems.Most primary is exactly to show in the satisfaction of client, for example, during peak period
Section, all lock studs of some lease points are in vacant state overlong time, cause user to borrow less than car；All locks of some lease points
Stake is long in full position state for time, causes user's also not car.These problems cause user satisfaction drastically to decline, customer complaint
Amount constantly rises, and directly have impact on query of the people for the practical function of public bicycles.
The content of the invention
In order to increase the satisfaction of user, the popular accreditation serviced for public bicycles is improved, public bicycles are improved
The service level of system, it is necessary to clustered to lease point using reasonable manner.It is an object of the invention to propose one kind
The clustering method of the similarity function changed over time based on vehicle number, by analyzing existing lease point feature, with
The vehicle number of each lease point carries out coded quantization with the changing rule of period, is leased same class with the mode of more convenient and quicker
Point is classified as a class.
The technical solution adopted by the present invention is as follows：
A kind of vehicle clustering method of public bicycles lease point, comprises the following steps：
(1) indicatrix of each lease point is set up according to time series, and Image Segmentation Methods Based on Features is carried out to curve, multiple points are obtained
Cutpoint；
(2) coding and quantization method of point feature curve are leased：
Step 21：Encoded radio is determined, the change of obtained indicatrix in the vehicle number of cut-point is split according to step (1)
Trend, that is, rise or decline, determine encoded radio；
Step 22：Encoding scheme is determined, determines which kind of coded system used according to the feature that curve is described；
Step 23：The quantization of curve encoding：The variable quantity of each cut-point is calculated according to the encoding scheme in step 22：Profit
Use formula：(vehicle number of the vehicle number of start time period-end time the period)/lease point slot total quantity, is obtained
Go out certain lease one day specific coding value of point；
(3) similarity of two lease points is calculated using similarity function：
Step 31：It is determined that the coding of the two lease points compared；
Step 32：The distance of two websites is calculated with edit distance approach；
Step 33：The similarity of two lease points, similarity function formula are calculated according to similarity function：Similarity=
1- editing distances/coding number；
(4) lease point is clustered：
Step 41：5 type of codings are selected in all encoded radios obtained using step (2), the first of each type is used as
Beginning central point；
Step 42：Calculate remaining each be encoded to the editing distances of this 5 initial center points, respectively by each lease dot-dash
Assign in the type where the minimum initial center point of editing distance；
Step 43：The editing distance leased two-by-two between point coding in each type is calculated, one is drawn into this type
Other minimum points of lease point editing distance, as new lease point, recalculate all lease points into 5 newly produced
The distance of heart point, repartitions the midpoint belonging to each lease point；
Step 44：Repeat step 43 is untill the process restrains；
Step 45：Finally according to the feature of lease vertex type, determine that above-mentioned 5 central points are belonging respectively to which kind.
The specific method of curve segmentation is in above-mentioned steps (1)：Step 11：It is between at the beginning of one day normal operation of definition
Ts, the end time is Te；Step 12：The time of lease point operation in one day is carried out being divided into N sections, then every of time is T
=(Te-Ts)/N；Step 13：Divided according to the period above, by curve segmentation into N number of point, this N number of point is encoded with specific
To represent the variation characteristic of curve here, the Time Calculation each put is as follows：Ti=Ts+T × i, wherein i=1,2 ..., N.
The specific steps of above-mentioned steps (2) include：Step 21 determines encoded radio：Song is obtained according to step (1) dividing method
Line also enters in cut-point, lends or in the variation tendency of station vehicle number, determine encoded radio；Step 22 determines encoding scheme：Compile
The concrete numerical value of code is determined by the amplitude of variation of this section of curve, if coding is to portray lending or the vehicle also entered change
Curve, then the coding formula of every section of vehicle change is as follows：
Here, CiCoding for i-th section simultaneously takes the upper bound, is represented with one, BiChange sum for i-th section of vehicle, W is should
Website parking stall sum；
If coding is to portray website in station number of vehicles change, the coding formula of every section of vehicle change is as follows：
Here,For i-th section lending vehicle number,For i-th section of vehicle number also entered；At station, vehicle number encoder needs two
Position represents that because there are positive and negative values, wherein first use 1 of negative is represented, first use 0 of positive number is represented.
The present invention is to be based on time series, on the basis of description, analysis and the segmentation to each lease point feature curve,
Each lease point is encoded, then lease point clustered by encoding, is respectively rented in public bicycles system so as to realize
The quick clustering rented a little.For similar lease point, it is convenient for such as predicting, optimizes various analyses, to cause matching somebody with somebody for website
Put and reach optimal, the trip for the public bicycles that are convenient for people to use.The encoding amount that the present invention passes through the indicatrix to lease point
Change, can be more straightforward show it is each lease point the characteristics of, with lease faster, with easily mode to public bicycles
Vertex type is clustered, the lease point distribution effectively alleviated and the unreasonable problem sorted out, and is also preferably alleviated and " is borrowed car
The present situation of difficulty, difficulty of returning the car ", improve the service level of public bicycles system, people for public bicycles total satisfactory grade
And the utilization rate of public bicycles.
Brief description of the drawings
Fig. 1 is coding, the quantization flow figure of present invention lease point feature curve；
Fig. 2 is the influence factor classification chart of the number of vehicles change of lease point；
Fig. 3 be in the fall, in the case of normal weather, date type one, study column area the change of lease point bicycle vehicle it is bent
Line chart；
Fig. 4 be in the fall, in the case of normal weather, date type one, the bicycle vehicle change of different lease vertex types
Curve map；
Fig. 5 be in the fall, under normal circumstances, same lease point, the bicycle vehicle change curve of different date types；
Fig. 6 is in the case of normal weather, same lease point, date type one, the bicycle vehicle change song of Various Seasonal
Line chart；
Fig. 7 be in the fall, same lease point, date type one, the bicycle vehicle change curve in the case of different weather
Figure；
Fig. 8 is that (Tuesday, weather is normal, and the total vehicle number of website is 20) one day on November 20th, 2012 for certain primary school lease point
Available vehicle number curve map；
Fig. 9 is the form that the occurrence after partition encoding is carried out to Fig. 8 curves；
Figure 10 is the cluster flow chart of present invention lease point.
Embodiment
The present invention is illustrated below in conjunction with accompanying drawing.It may be noted that described embodiment is only deemed as the mesh of explanation
, rather than the limitation to invention.
The coding method of the lease point feature curve proposed in the present invention, with reference to the analysis to curve division time point
On assign corresponding encoded radio according to the variation tendency and amplitude of curve, reach the purpose to the coded quantization of indicatrix.Fig. 1
Give the present invention coding, the quantization flow figure of lease point feature curve：
Step 101：The indicatrix of lease point is depicted according to the statistics of lease point；
Step 102：The indicatrix for leasing point is cut according to the cutting method of curve；
Step 103：The tendency that curve is determined at each cut point is rising or decline；
Step 104：It is determined that rising；
Step 105：Sign bit represents with 0, bits of coded CiRepresent (CiCoding for i-th section)；
Step 106：It is determined that declining；
Step 107：Sign bit represents with 1, bits of coded CiRepresent (CiCoding for i-th section)；
Step 108：Continue every section of coding until terminating；
Step 109：Curve is obtained completely to encode.
1st, the influence factor sorting technique of the number of vehicles change of lease point
According to the lending of the bicycle of lease point, give back, in line number, and room number etc. statistical analysis, find lease
The bicycle number of point is relevant with the factors such as vertex type are leased by season, weather, working day, specific influence factor such as Fig. 2 institutes
Show.The feature curve analysis for combining specific lease point for each influence factor is as follows：
Fig. 3 be in the fall, in the case of normal weather, date type one, study column area the change of lease point bicycle vehicle it is bent
Line, analyzes and understands, the trip peak period (go to work, go to school) of people at 9 points or so, and therefore, the lease point is available at this moment
Bicycle number reaches the low peak period in one day；(come off duty, classes are at top in being reached at 18 points or so with bicycle number one day
Etc. reason).
Fig. 4 be in the fall, in the case of normal weather, date type one, the bicycle vehicle change of different lease vertex types
Curve, analysis understands that the bicycle vehicle change curve of different types of lease point is different.Wherein residential block in the morning on
Class's time reaches low ebb, and the quitting time peaks；Study column area is then that ebb is reached at 10 points or so, and 12 points and 18 points or so reach
To peak the reason for (go to school, classes are over)；Shopping centre is peaked at 10 points or so, and 15 points or so reach ebb；Scenic spot and hospital
Changing rule it is a bit similar, it is little in 9 points and 15 points or so its amplitude of variation that peak
Fig. 5 is that in the fall, in the case of normal weather, same lease point, the bicycle vehicle change of different date types is bent
Line, analysis understands that the date three is then different than relatively similar for the changing rule on date one and date two.Crack cause includes the date one
And during two belong to regular working day, and the date two is slightly gentle close to weekend variation tendency relative-date one, the date one and
Two peak value difference；Date three belongs to day off, and the travel time of people is than working day relatively a little later.
Fig. 6 is same lease point, date type one, the bicycle vehicle change of Various Seasonal in the case of normal weather
Curve, analysis understands that spring, summer, autumn, the rule of four season vehicle change curves of winter are peaks that is similar, can simply reaching
Value is different.The main cause of this phenomenon is formed because in season in spring and autumn proper temperature selection public bicycles trip
People is relatively more, is less suitable in winter and summer temperature, causes the use of public bicycles to reduce
Fig. 7 be in the fall, same lease point, date type one, the bicycle vehicle change curve in the case of different weather
Analysis understands that the vehicle change curve of bicycle is similar under different weather situation.Only due to anomalous weather in the case of,
Weather is comparatively relatively more severe, and people's trip number is reduced, and therefore, the vehicle change curve of bicycle is relatively gentle.
2nd, the dividing method of curve
Curve segmentation mainly for curve encoding service, so how it is appropriate by curve carry out segmentation be coding
Premise.The curve of bicycle amount change, typically all histogram, histogram there are multiple Wave crest and wave troughs or Wave crest and wave trough is special
Unconspicuous situation is levied, so we can split according to its kurtosis feature to curve.Here we are not it is desirable that
Change number in intraday vehicles passing in and out with website, then this curve is carried out quantization segmentation by the curve of simulating vehicle trip.
The step of curve segmentation, is as follows：
Step one：It is Ts (such as, 6 points of morning, are defined as Ts=6) between at the beginning of one day normal operation of definition, terminates
Time is Te (such as, at 22 points in evening, are defined as Te=22)；
Step 2：The time of lease point operation in one day is carried out being divided into N sections, then every of time is T=(Te-
Ts)/N；
Step 3：Divided according to period above, by curve segmentation into N number of point, this N number of point is with specifically encoding come table
Show the variation characteristic of curve here.The Time Calculation each put is as follows：
Ti=Ts+T × i i=1,2 ..., N
3rd, curve encoding method
The coding of curve mainly has the tendency of curve relevant with amplitude of variation, how accurately that the feature of a certain website is bent
It is the key for carrying out type of site division that line, which carries out rational coded representation,.The step of coding, is as follows：
Step one：Determine encoded radio
Curve is obtained according to dividing method also to enter, lend or in the variation tendency of station vehicle number (rise in cut-point
Or decline), determine encoded radio.
Step 2：Determine encoding scheme
The concrete numerical value of coding determines by the amplitude of variation of this section of curve, if coding be portray curve lending or
It is the vehicle change curve also entered, then the coding formula of every section of vehicle change is as follows：
Here, CiCoding for i-th section simultaneously takes the upper bound, is represented with one, BiChange sum for i-th section of vehicle, W is should
Website parking stall sum；
If coding is to portray website in station number of vehicles change, the coding formula of every section of vehicle change is as follows：
Here,For i-th section lending vehicle number,For i-th section of vehicle number also entered.Online vehicle number encoder needs two
Position represents that because there are positive and negative values, wherein first use 1 of negative is represented, first use 0 of positive number is represented.
Step 3：Specific coding
Example：It is that (Tuesday, weather is normal, and the total vehicle number of website is on November 20th, 2012 for certain primary school lease point such as Fig. 8
20) curve map of the available vehicle number of one day, segmentation sum is 32 (i.e. N=30), calculates T values, i.e., with per half an hour
Variable quantity is a coding, draws specific coding such as Fig. 9 of the lease point, table using the calculation formula of step 2 by analyzing
The result positive number of its in lattice represents that vehicle number is reduced in vehicle number increase, this short time of negative number representation in this period.
Then the lease point one day is encoded to：
0011011400011302110203150102000502120105040300041314000012030100
4th, similarity function calculation procedure
1) similarity function computational methods
Editing distance (Edit Distance)：Also known as Levenshtein distances, refer between two word strings, are turned by one
Minimum edit operation number of times needed for into another.The edit operation of license includes a character being substituted for another character,
A character is inserted, a character is deleted.Similarity function computational methods are as follows：
Similarity=1- editing distances/coding number.
The words of kitten mono- are for example changed into sitting：
●sitten(k→s)
●sittin(e→i)
●sitting(→g)
I.e. editing distance is 3.
2) the step of calculating similarity：
Step one：It is determined that the coding of two websites compared；
Step 2：The distance of two websites is calculated with editing distance；
Step 3：Calculate the similarity of two websites.
5th, lease point clustering method
1) define：Cluster is exactly the process that set of data objects is classified according to its similitude, and homogeneous object is similar
Property it is high, inhomogeneous objects similarity is small.
2) cluster process generally comprises 5 parts：
(1) data prepare；(2) feature selecting；(3) feature extraction；(4) cluster (or packet)；(5) cluster result is assessed.
3) clustering method
Step one：Set up the indicatrix coding of lease point；
Step 2：5 type of codings (cell, school, hospital, commercial center, tourism scape are selected in all codings
Point), it is used as the initial center point of each type；
Step 3：Calculate remaining each be encoded to the editing distances of this 5 initial center points, respectively by each lease point
It is divided into the type where the minimum initial center point of editing distance；
Step 4：The editing distance leased two-by-two between point coding in each type is calculated, one is drawn into this type
Other minimum points of lease point editing distance, as new lease point, recalculate all lease points into 5 newly produced
The distance of heart point, repartitions the midpoint belonging to each lease point；
Step 5：Repeat step four is untill the process restrains；
Step 6：Finally according to the feature of lease vertex type, determine that above-mentioned 5 central points are belonging respectively to which kind.
Figure 10 gives the cluster flow chart of present invention lease point：
Step 201：Set up the indicatrix coding of all lease points；
Step 202：Choose the cluster centre (randomly selecting 5 lease points for the first time) of type in 5；
Step 203：A lease point is calculated to the editing distance of cluster centre；
Step 204：If the editing distance of other points in each type in current cluster centre to this type is not most
Short then repeat step 202 and step 203, untill convergence；
Step 205：Cluster process terminates if judging to meet condition.
The coding method of the lease point feature curve proposed in the present invention, with reference to the analysis to curve division time point
On assign corresponding encoded radio according to the variation tendency and amplitude of curve, reach the purpose to the coded quantization of indicatrix.
The lease point clustering method proposed in the present invention, with reference to factors such as date, the weather of influence lease point feature curve,
Three different characteristic curves of each lease point are drawn, the replacement for finding three indicatrixes with the method for editing distance is bent
Line, using the curve as lease, the indicatrix of point uses editing distance to be clustered.
Claims (3)
1. a kind of vehicle clustering method of public bicycles lease point, it is characterised in that the method comprises the following steps：
(1) indicatrix of each lease point is set up according to time series, and Image Segmentation Methods Based on Features is carried out to curve, multiple segmentations are obtained
Point；
(2) coding and quantization method of point feature curve are leased：
Step 21：Encoded radio is determined, obtained indicatrix is split according to step (1) and is become in the change of the vehicle number of cut-point
Gesture, that is, rise or decline, determine encoded radio；
Step 22：Encoding scheme is determined, determines which kind of coded system used according to the feature that curve is described；
Step 23：The quantization of curve encoding：The variable quantity of each cut-point is calculated according to the encoding scheme in step 22：Utilize public affairs
Formula：(vehicle number of the vehicle number of start time period-end time the period)/lease point slot total quantity, draws certain
The lease point specific coding value of one day；
(3) similarity of two lease points is calculated using similarity function：
Step 31：It is determined that the coding of the two lease points compared；
Step 32：The distance of two websites is calculated with edit distance approach；
Step 33：The similarity of two lease points, similarity function formula are calculated according to similarity function：Similarity=1- is compiled
Collect distance/coding number；
(4) lease point is clustered：
Step 41：In all encoded radios obtained using step (2) select 5 type of codings, as each type it is initial in
Heart point；
Step 42：Calculate remaining each be encoded to the editing distances of this 5 initial center points, each lease point is divided into respectively
In type where the minimum initial center point of editing distance；
Step 43：Calculate in each type the editing distance between lease point coding two-by-two, draw one into this type other
The minimum point of lease point editing distance, as new lease point, recalculates all lease points to 5 central points newly produced
Distance, repartition each lease point belonging to midpoint；
Step 44：Repeat step 43 is untill the process restrains；
Step 45：Finally according to the feature of lease vertex type, determine that above-mentioned 5 central points are belonging respectively to which kind.
2. a kind of vehicle clustering method of public bicycles lease point according to claim 1, it is characterised in that step
(1) specific method of curve segmentation is in：
Step 11：It is Ts between at the beginning of one day normal operation of definition, the end time is Te；
Step 12：The time of lease point operation in one day is carried out being divided into N sections, then every of time is T=(Te-Ts)/N；
Step 13：Divided according to the period above, by curve segmentation into N number of point, this N number of point represents bent with specific coding
The variation characteristic of line here, the Time Calculation each put is as follows：Ti=Ts+T × i, wherein i=1,2 ..., N.
3. a kind of vehicle clustering method of public bicycles lease point according to claim 1 or 2, it is characterised in that step
Suddenly the specific steps of (2) include：
Step 21 determines encoded radio：Curve is obtained according to step (1) dividing method also to enter in cut-point, lend or in station vehicle
Several variation tendencies, determines encoded radio；
Step 22 determines encoding scheme：The concrete numerical value of coding is determined by the amplitude of variation of this section of curve, if coding is to carve
The vehicle change curve that lending either also enters is drawn, then the coding formula of every section of vehicle change is as follows：
Here, CiCoding for i-th section simultaneously takes the upper bound, is represented with one, BiChange sum for i-th section of vehicle, W is the website
Parking stall sum；
If coding is to portray website in station number of vehicles change, the coding formula of every section of vehicle change is as follows：
Here,For i-th section lending vehicle number,For i-th section of vehicle number also entered；At station, vehicle number encoder needs two to come
Represent, because there are positive and negative values, wherein first use 1 of negative is represented, first use 0 of positive number is represented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410616030.5A CN104376327B (en) | 2014-11-05 | 2014-11-05 | A kind of clustering method of public bicycles lease point |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410616030.5A CN104376327B (en) | 2014-11-05 | 2014-11-05 | A kind of clustering method of public bicycles lease point |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104376327A CN104376327A (en) | 2015-02-25 |
| CN104376327B true CN104376327B (en) | 2017-10-10 |
Family
ID=52555223
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410616030.5A Expired - Fee Related CN104376327B (en) | 2014-11-05 | 2014-11-05 | A kind of clustering method of public bicycles lease point |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104376327B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104850900B (en) * | 2015-04-27 | 2018-08-28 | 北京工业大学 | A kind of complete method of city-bike system net point layout optimization |
| CN105719083A (en) * | 2016-01-21 | 2016-06-29 | 华南理工大学 | Public bicycle peak time scheduling method based on multilevel partition |
| CN106910103B (en) * | 2017-01-09 | 2021-06-01 | 杭州电子科技大学 | Public bicycle system leasing point function clustering method |
| CN107045673B (en) * | 2017-03-31 | 2020-09-29 | 杭州电子科技大学 | Public bicycle flow variation prediction method based on stack model fusion |
| CN107292798A (en) * | 2017-06-29 | 2017-10-24 | 国信优易数据有限公司 | A kind of shared bicycle parks determination method and device a little |
| CN107784115B (en) * | 2017-11-09 | 2019-01-22 | 福建省特种设备检验研究院 | A kind of special equipment failure analysis methods and system based on editing distance algorithm |
| CN108256969B (en) * | 2018-01-12 | 2021-07-16 | 杭州电子科技大学 | Public bicycle leasing point dispatching area dividing method |
| CN108960476B (en) * | 2018-03-30 | 2022-04-15 | 山东师范大学 | AP-TI clustering-based shared bicycle flow prediction method and device |
| CN108877202B (en) * | 2018-08-24 | 2020-12-15 | 北京轻享科技有限公司 | Monitoring method and device for taxi, vehicle and computing equipment |
| CN109583491A (en) * | 2018-11-23 | 2019-04-05 | 温州职业技术学院 | A kind of shared bicycle intelligent dispatching method |
| CN109543752A (en) * | 2018-11-23 | 2019-03-29 | 温州职业技术学院 | A kind of shared bicycle website clustering method |
| CN110929783B (en) * | 2019-11-21 | 2023-04-07 | 同济大学 | Land use attribute classification method of rented and returned data based on shared object |
| CN113537549A (en) * | 2020-04-22 | 2021-10-22 | 富士通株式会社 | Information processing apparatus, information processing method, and computer program |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102982395A (en) * | 2012-11-28 | 2013-03-20 | 浙江工业大学 | Rapid bus transfer method based on space node clustering method |
| CN103198548A (en) * | 2013-04-03 | 2013-07-10 | 深圳职业技术学院 | Regional bus on-off passenger number distinguishing algorithm based on switch sensors |
| CN103593974A (en) * | 2013-11-06 | 2014-02-19 | 福建工程学院 | Bus passenger capacity collection method based on locating information |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005081158A2 (en) * | 2004-02-23 | 2005-09-01 | Novartis Ag | Use of feature point pharmacophores (fepops) |
-
2014
- 2014-11-05 CN CN201410616030.5A patent/CN104376327B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102982395A (en) * | 2012-11-28 | 2013-03-20 | 浙江工业大学 | Rapid bus transfer method based on space node clustering method |
| CN103198548A (en) * | 2013-04-03 | 2013-07-10 | 深圳职业技术学院 | Regional bus on-off passenger number distinguishing algorithm based on switch sensors |
| CN103593974A (en) * | 2013-11-06 | 2014-02-19 | 福建工程学院 | Bus passenger capacity collection method based on locating information |
Non-Patent Citations (1)
| Title |
|---|
| "城市公共自行车服务系统运行状况和效率分析";张彪， 等;《工程数学学报》;20131231;第30卷;第165-180页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104376327A (en) | 2015-02-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104376327B (en) | A kind of clustering method of public bicycles lease point | |
| CN103198104B (en) | A kind of public transport station OD acquisition methods based on city intelligent public transit system | |
| CN106779163A (en) | A kind of customization transit network planning method based on intelligent search | |
| CN105719019A (en) | Public bicycle peak time demand prediction method considering user reservation data | |
| CN107038168A (en) | A kind of user's commuting track management method, apparatus and system | |
| CN106816008A (en) | A kind of congestion in road early warning and congestion form time forecasting methods | |
| CN1734238A (en) | Two-step multi-path optimization method for central controlled vehicle information system | |
| CN111190942B (en) | Urban road parking spot overall analysis method based on data mining technology | |
| CN114021883A (en) | Dispatching method for subway transfer shared bicycle in peak period | |
| CN109886468A (en) | Charging station planing method based on improved self-adapted genetic algorithm | |
| CN109145989B (en) | Bus stop layout method and device and computer terminal | |
| CN104318081A (en) | Method for allocating bicycles at public bicycle rental stations with urgent demand in city | |
| CN111768638A (en) | Lane distribution method for single-point signalized intersection | |
| CN116187591A (en) | Method for predicting number of remaining parking spaces in commercial parking lot based on dynamic space-time trend | |
| CN107665583B (en) | Method for calculating lane saturation flow rate under different weather conditions | |
| CN113344240A (en) | Shared bicycle flow prediction method and system | |
| CN114358386A (en) | Double-trip-mode ride-sharing site generation method based on reserved trip demand | |
| CN116703132B (en) | Management method and device for dynamic scheduling of shared vehicles and computer equipment | |
| CN110097757B (en) | Intersection group critical path identification method based on depth-first search | |
| Sunitha et al. | Cluster‐Based Pavement Deterioration Models for Low‐Volume Rural Roads | |
| CN116993391A (en) | Site type shared bicycle system use demand prediction method | |
| CN109615727B (en) | Riding endpoint extraction method and system based on shared bicycle static GPS data | |
| CN115565376B (en) | Vehicle journey time prediction method and system integrating graph2vec and double-layer LSTM | |
| CN115858708A (en) | Traffic cell division method based on travel gravitation model | |
| CN114266316B (en) | Hierarchical graph convolutional network-based carbon footprint-user classification method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171010 Termination date: 20201105 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |