CN116307275B - Bicycle flow prediction method based on airspace causal transmission - Google Patents
Bicycle flow prediction method based on airspace causal transmission Download PDFInfo
- Publication number
- CN116307275B CN116307275B CN202310558821.6A CN202310558821A CN116307275B CN 116307275 B CN116307275 B CN 116307275B CN 202310558821 A CN202310558821 A CN 202310558821A CN 116307275 B CN116307275 B CN 116307275B
- Authority
- CN
- China
- Prior art keywords
- causal
- matrix
- bicycle
- airspace
- liquid crystal
- 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.)
- Active
Links
- 230000001364 causal effect Effects 0.000 title claims abstract description 61
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000005540 biological transmission Effects 0.000 title claims abstract description 22
- 239000011159 matrix material Substances 0.000 claims abstract description 45
- 238000012546 transfer Methods 0.000 claims abstract description 16
- 230000003993 interaction Effects 0.000 claims abstract description 9
- 230000008569 process Effects 0.000 claims abstract description 8
- 238000013135 deep learning Methods 0.000 claims description 28
- 238000013528 artificial neural network Methods 0.000 claims description 26
- 239000004973 liquid crystal related substance Substances 0.000 claims description 26
- 230000006870 function Effects 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 11
- 238000012384 transportation and delivery Methods 0.000 claims description 11
- 238000012360 testing method Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 238000003064 k means clustering Methods 0.000 claims description 3
- 238000005192 partition Methods 0.000 claims description 2
- 238000005096 rolling process Methods 0.000 claims description 2
- 238000013461 design Methods 0.000 abstract description 2
- 230000006698 induction Effects 0.000 abstract 1
- 125000002015 acyclic group Chemical group 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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"
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention belongs to the technical field of intelligent transportation, and discloses a bicycle flow prediction method based on airspace causal transmission. The method utilizes a gating circulation unit to extract time domain features, designs a learnable causal embedding vector, constructs a causal transfer matrix among areas, constructs a corresponding graph structure as induction bias by utilizing priori knowledge such as inter-area distance, interaction quantity and the like, and further obtains space-time causal features through an airspace causal transfer process; and finally, outputting a prediction result through a prediction module. The prediction method eliminates the false correlation of the airspace between the non-causal-related areas, and effectively improves the accuracy of the bicycle flow prediction.
Description
Technical Field
The invention belongs to the technical field of intelligent transportation, and particularly relates to a bicycle flow prediction method based on airspace causal transmission.
Background
The shared bicycle is an important component for urban intelligent traffic development, and can effectively relieve urban traffic jam and reduce energy consumption. The use condition of the shared bicycle, namely the prediction of the flow of the shared bicycle on the urban road, is a key link in the research of solving the urban public transportation problem. The effective shared bicycle flow prediction can provide a road network integral high-dimensional view for the shared bicycle platform, so that the site selection of the leasing site and the distribution of the number of shared bicycles are scientifically planned, and road traffic jam and resource waste are reduced. Meanwhile, the shared bicycle flow state projects urban population flow conditions, and important reference data is provided for urban traffic resource allocation.
Because the traditional method has poor effect on the problem of bicycle flow prediction, more researchers use a deep learning method for prediction. The Chinese patent publication No. CN107045673B discloses a method for predicting the flow variation of a public bicycle based on pile model fusion, which utilizes multi-source data fusion to construct multiple characteristic types in a classified manner, trains multiple basic models, constructs multiple models from different angles, ensures certain difference between the basic models, and finally constructs the pile models in a cross-validation mode. The Chinese patent publication No. CN115204477A discloses a bicycle flow prediction method of a context-aware graph recursion network, which uses a mode constraint clustering method to cluster independent stations into groups, each station group is a functional area, calculates bicycle inlet and outlet flow of stations in each area, and acquires weather data as external information; the bicycle flow prediction model based on the area is constructed by adopting a context-aware graph recursion network, and comprises an encoder and a decoder which are composed of context-aware graph recursion units, wherein each unit is formed by integrating a context embedding layer and an adaptive graph generator into a gating recursion unit GRU, and replacing a linear transformation layer in the GRU with an adaptive graph convolution; and deploying the trained prediction model on the shared bicycle management system.
The above approach lacks the mining of potential causal relationships between regions. In the message transmission process of graph convolution, the influence of space domain false correlation is introduced between non-causal association areas, so that the accuracy of model prediction is greatly reduced.
Disclosure of Invention
Based on the defects in the prior art, the invention provides the method for improving the prediction accuracy from the causal view angle. Because the bicycle flow has the characteristics of complexity and high dimension, the causal theory is applied to the prediction task. The invention provides a bicycle flow prediction method based on airspace causal transmission, which utilizes a gating circulation unit to extract time domain features, designs a learnable causal embedded vector, constructs a causal transmission matrix among nodes according to the time domain features, constructs a corresponding graph structure as a generalized bias according to priori knowledge such as inter-area distance, interaction quantity and the like, and further obtains space-time causal features through an airspace causal transmission process; and finally, outputting a prediction result through a prediction module.
The technical scheme of the invention is as follows:
the bicycle flow prediction method based on airspace causal transmission is characterized by establishing a deep learning neural network comprising an airspace causal transmission process, and realizing bicycle flow prediction based on the deep learning neural network; the deep learning neural network comprises a plurality of stacked gating circulation units, an airspace causality transmission unit and a prediction unit.
Preferably, the bicycle flow prediction method comprises the following steps:
step one, acquiring historical shared bicycle order information from a shared bicycle platform, and dividing a range to be predicted into sections according to urban partitionsNObtaining historical bicycle flow in different time periods in each area;
step two, establishing a deep learning neural network comprising an airspace causality transmission process;
training and testing the deep learning neural network by adopting the bicycle flow in different time periods in each region obtained in the step one;
and fourthly, predicting the bicycle flow in a future period based on the observation data to be predicted by using the trained deep learning neural network.
Preferably, the saidThe gating circulation unit introduces a POI data set according to known confounding factors, counts the quantity of various POIs in each area and forms a POI feature matrixWhereinqRepresenting the total number of POI categories used, and using a K-means clustering algorithm to performNThe individual regions are divided into POI categoriesqClass, extracting time domain features of various areas, and splicing to obtain the firsttTime domain features of all regions within each period +.>。
Preferably, the airspace causal delivery unit includes a causal delivery network constructed based on a causal delivery matrix:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the firsttThe reason for all areas within a period is embedded in matrix, < >>Is the firsttResults for all areas within each period are embedded in matrix, < >>Is the firsttEmbedding vector dimensions in regions within a period;
first, thetThe causal transfer process within each period is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,is a weight matrix which can be learned;
first, thetSpatiotemporal causal features within a time periodThe extraction is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,，/>representing the distance relation among the areas for the adjacency matrix corresponding to the distance graph constructed based on priori knowledge, wherein the firstiLine (1)jThe elements of the columns being regionsiAnd region ofjDistance of->Representing interaction relation among all areas for adjacency matrix corresponding to interaction graph constructed based on priori knowledge, wherein the firstiLine (1)jThe elements of the columns being regionsiTo the areajBicycle flow and areaiRatio of inner bicycle flow->，And->，/>Is the graph rolling network weight.
Preferably, the causal transfer matrix has the following loop-free constraints:
wherein the method comprises the steps ofIIs a matrix of units which is a matrix of units,representing the trace of the matrix.
Preferably, the bicycle order information includes: bicycle number, departure longitude, departure latitude, arrival longitude, arrival latitude, departure time, arrival time, and travel time.
Preferably, the prediction unit generates the prediction result using the stacked full connection layer and the Relu activation function, as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing the predicted outcome->Is the firstQSpatiotemporal causal features extracted over a period of time,is a parameter matrix which can be learned.
Preferably, the deep learning neural network training uses a multi-objective loss function, as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,to control the hyper-parameters of the balance +.>Representing the total loss function>Representing the predicted loss.
Compared with the prior art, the invention has the beneficial effects that:
according to the bicycle flow prediction method based on airspace causal transmission, which is provided by the invention, based on historical bicycle flow data in urban areas, a single gating circulation unit is utilized to extract time domain features, a learnable spatial causal transmission matrix is established, meanwhile, a corresponding graph structure is constructed as a generalized bias by utilizing the distance between areas and the priori knowledge of interaction quantity, the time domain features are converted into the spatio-temporal causal features, the airspace false correlation between non-causal association areas is eliminated, and the accuracy of bicycle flow prediction is effectively improved.
Drawings
So that the manner in which the above recited embodiments of the present invention and the manner in which the same are attained and can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments thereof which are illustrated in the appended drawings, which drawings are intended to be illustrative, and which drawings, however, are not to be construed as limiting the invention in any way, and in which other drawings may be obtained by those skilled in the art without the benefit of the appended claims.
FIG. 1 is a flow chart of a bicycle flow prediction method based on airspace causal transmission.
Fig. 2 is a graph of the prediction result in example 1 of the present invention.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, without conflict, the embodiments of the present invention and features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.
The invention provides a bicycle flow prediction method based on airspace causal transmission, which is shown in fig. 1, and specifically comprises the following steps:
step 1: acquiring shared bicycle order information in a certain period of time through a shared bicycle platform, wherein the order information comprises the following eight types of data: bicycle numbering and departureLongitude, latitude of departure, longitude of arrival, latitude of arrival, time of departure, time of arrival, and time of trip. Then, the order information is standardized and divided into areas according to the cityNRemoving the riding time and the riding distance abnormal order data in each area, and counting the areasQBicycle traffic for different periods of time.
Step 2: a deep learning neural network comprising an airspace causal delivery process is established, the deep learning neural network comprising a plurality of stacked gating loop units and airspace causal delivery units, and a prediction unit.
The gating circulation unit introduces a POI data set according to known confounding factors (such as the functional attribute of each region), counts the number of various POIs of each region and forms a POI feature matrixWhereinqRepresenting the total number of POI categories used, and using a K-means clustering algorithm to performNThe individual regions are divided into POI categoriesqClass, extracting time domain features of various areas, and splicing to obtain the firsttTime domain features of all regions within each period +.>。
The airspace causality transfer unit provides causality transfer processes among causality transfer network modeling areas on the basis of constructing a causality transfer matrix, so that unbiased space-time causality features can be extracted. To mine the potential causal structure between regions, a causal transfer matrix is generated end-to-end in the back propagation process, and the dynamic matrix is learned in different time steps in consideration of the dynamics of causal links between regions.
Because of the large number of categories of regions, fitting the potential causal transfer matrix by gradient descent can result in over-parameterization of the deep learning neural network and a large computational load. In order to solve the problems, the invention decomposes the potential causal transfer matrix as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the firsttThe reason for all areas within a period is embedded in matrix, < >>Is the firsttResults for all areas within each period are embedded in matrix, < >>Is the firsttThe regions within each period embed the vector dimensions.
Through the processing, the learnable parameter of the deep learning neural network is represented byDown to，/>Meanwhile, the calculation load is effectively reduced, and the model is prevented from being overfitted by reducing the parameter quantity.
In addition, because the causal graph has the property of directed acyclic, the invention adds the following directed acyclic constraints to the learned causal transfer matrix:
wherein the method comprises the steps ofIIs a matrix of units which is a matrix of units,representing the trace of the matrix.
The present invention defines a causal delivery network based on a learnable causal delivery matrix, which follows two principles: (1) The result areas in all causal links aggregate the information of all cause areas. (2) Non-causal spurious correlations cannot be introduced during causal delivery.
First, thetThe causal transfer process within each period is as follows:
wherein the method comprises the steps ofTranspose of causal transfer matrix +.>Is a weight matrix that can be learned.
In addition, a distance adjacency matrix and an interaction adjacency matrix are constructed based on priori knowledge to serve as inductive bias of the deep learning neural network, so that convergence speed is increased. First, thetThe spatiotemporal causal feature extraction process in each time period is as follows:
the prediction unit generates a prediction result by using the stacked full connection layer and the Relu activation function, wherein the prediction result is represented by the following formula:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing the predicted outcome->Is the firstQSpatiotemporal causal features extracted in the period (i.e. last period), +.>Is a parameter matrix which can be learned.
Using L2-norm as predictive lossAnd is combined withThe multi-objective loss function is designed to train the deep learning neural network, and the following formula is shown:
wherein, the liquid crystal display device comprises a liquid crystal display device,to control the hyper-parameters of the balance +.>Representing the total loss function.
Step 3: training and testing the deep learning neural network established in the step 2, inputting observation data to be predicted into the trained deep learning neural network, and outputting bicycle flow in future time periods of each region.
Example 1
The present invention uses shared bicycle order data for western urban areas of beijing to train a deep learning network. The training data is adopted to intercept order data of 4 th month 1 th year of 2016 to 30 th year of 2016, and 54 areas are divided in total.
Each order data includes the following eight data: bicycle number, departure longitude, departure latitude, arrival longitude, arrival latitude, departure time, arrival time, and travel time. The order data is divided into a plurality of data sets according to different areas and time periods, and the time period interval is 30 minutes. The dataset was divided in the time dimension into training set (60%), validation set (20%) and test set (20%). The historical 5 hour zone bicycle flow is used in this example to predict future 30 minute zone demand.
Training the established deep learning neural network comprising the airspace causal transmission process.
The data was Z-score normalized and all parameters in the deep learning neural network were randomly initialized.
The deep learning neural network was trained on the complete data set for 200 cycles by means of Adam optimization algorithm and exponential decay dynamic learning rate strategy. And verifying the deep learning neural network by using the loss function in each training period, and then storing the optimal parameters according to the loss function value. An Early stop strategy is used in the training process, and when the loss function value does not drop for 50 continuous periods, the training is terminated in advance.
The prediction results of this example were compared with the prior art, and regional bicycle flow predictions were performed on the same dataset, with the comparison results shown in table 1 and fig. 2. The prediction results were evaluated using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) and mean percent absolute error (MAPE), with lower errors indicating better prediction results. The present invention compares eight prior art prediction methods.
The first is GRU, the second is LSTM, the third is STGCN, the fourth is HGCN, the fifth is GraphWaveNet, the sixth is AGCRN, the seventh is DGCRN, and the eighth is DMSTGCN.
As can be clearly seen from comparison results, the prediction effect of the bicycle flow prediction method based on airspace causal transmission provided by the invention is better than that of the prior art.
Table 1 comparison of the results of bicycle flow predictions in the Western urban area of Beijing city
In the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," "secured," and the like are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally formed; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communicated with the inside of two elements or the interaction relationship of the two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.
In the present invention, unless expressly stated or limited otherwise, a first feature "above" or "below" a second feature may include both the first and second features being in direct contact, as well as the first and second features not being in direct contact but being in contact with each other through additional features therebetween. Moreover, a first feature being "above," "over" and "on" a second feature includes the first feature being directly above and obliquely above the second feature, or simply indicating that the first feature is higher in level than the second feature. The first feature being "under", "below" and "beneath" the second feature includes the first feature being directly under and obliquely below the second feature, or simply means that the first feature is less level than the second feature.
In the present invention, the terms "first," "second," "third," "fourth" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. The term "plurality" refers to two or more, unless explicitly defined otherwise.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (3)
1. The bicycle flow prediction method based on airspace causal transmission is characterized by establishing a deep learning neural network comprising an airspace causal transmission process and realizing bicycle flow prediction based on the deep learning neural network;
the deep learning neural network comprises a plurality of stacked gating circulation units, an airspace causality transmission unit and a prediction unit;
the gating circulation unit introduces a POI data set according to known confounding factors, counts the quantity of various POIs in each area and forms a POI feature matrixWhereinqRepresenting the total number of POI categories used, and using a K-means clustering algorithm to performNThe individual regions are divided into POI categoriesqClass, extracting time domain features of various areas, and splicing to obtain the firsttTime domain features of all regions within each period +.>；
The airspace causal delivery unit includes a causal delivery network constructed based on a causal delivery matrix:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the firsttThe reason for all areas within a period is embedded in matrix, < >>Is the firsttResults for all areas within each period are embedded in matrix, < >>Is the firsttEmbedding vector dimensions in regions within a period;
first, thetThe causal transfer process within each period is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,is a weight matrix which can be learned;
first, thetSpatiotemporal causal features over each period are extracted as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,middle->，/>Representing the distance relation among the areas for the adjacency matrix corresponding to the distance graph constructed based on priori knowledge, wherein the firstiLine (1)jThe elements of the columns being regionsiAnd region ofjDistance of->Representing interaction relation among all areas for adjacency matrix corresponding to interaction graph constructed based on priori knowledge, wherein the firstiLine (1)jThe elements of the columns being regionsiTo the areajBicycle flow and areaiRatio of inner bicycle flow->，/>And->，/>Is the graph rolling network weight;
the causal transfer matrix has the following constraint of no loop:
wherein the method comprises the steps ofIIs a matrix of units which is a matrix of units,representing the trace of the matrix;
the prediction unit generates a prediction result by using the stacked full connection layer and the Relu activation function, wherein the prediction result is represented by the following formula:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing the predicted outcome->Is the firstQSpatiotemporal causal features extracted over a period of time, +.>Is a parameter matrix which can be learned;
the deep learning neural network training uses a multi-objective loss function as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,to control the hyper-parameters of the balance +.>Representing the total loss function>Representing the predicted loss.
2. The bicycle flow prediction method of claim 1, characterized in that it comprises the steps of:
step one, acquiring historical shared bicycle order information from a shared bicycle platform, and dividing a range to be predicted into sections according to urban partitionsNIndividual areas, getHistorical bicycle flow at different time intervals in each region;
step two, establishing a deep learning neural network comprising an airspace causality transmission process;
training and testing the deep learning neural network by adopting the bicycle flow in different time periods in each region obtained in the step one;
and fourthly, predicting the bicycle flow in a future period based on the observation data to be predicted by using the trained deep learning neural network.
3. The bicycle flow prediction method according to claim 2, wherein the sharing bicycle order information includes: bicycle number, departure longitude, departure latitude, arrival longitude, arrival latitude, departure time, arrival time, and travel time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310558821.6A CN116307275B (en) | 2023-05-18 | 2023-05-18 | Bicycle flow prediction method based on airspace causal transmission |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310558821.6A CN116307275B (en) | 2023-05-18 | 2023-05-18 | Bicycle flow prediction method based on airspace causal transmission |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116307275A CN116307275A (en) | 2023-06-23 |
| CN116307275B true CN116307275B (en) | 2023-07-28 |
Family
ID=86796364
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310558821.6A Active CN116307275B (en) | 2023-05-18 | 2023-05-18 | Bicycle flow prediction method based on airspace causal transmission |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116307275B (en) |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10733877B2 (en) * | 2017-11-30 | 2020-08-04 | Volkswagen Ag | System and method for predicting and maximizing traffic flow |
| CN112801404B (en) * | 2021-02-14 | 2024-03-22 | 北京工业大学 | Traffic prediction method based on self-adaptive space self-attention force diagram convolution |
| CN115204477A (en) * | 2022-06-30 | 2022-10-18 | 北京航空航天大学 | Bicycle flow prediction method of context awareness graph recursive network |
| CN115204478A (en) * | 2022-06-30 | 2022-10-18 | 北京航空航天大学 | Public traffic flow prediction method combining urban interest points and space-time causal relationship |
| CN115691137B (en) * | 2022-11-01 | 2024-04-30 | 北京航空航天大学 | Multi-modal data prediction method based on causal Markov model |
| CN116108984A (en) * | 2023-02-01 | 2023-05-12 | 南京航空航天大学 | Urban flow prediction method based on flow-POI causal relationship reasoning |
-
2023
- 2023-05-18 CN CN202310558821.6A patent/CN116307275B/en active Active
Non-Patent Citations (1)
| Title |
|---|
| Spatio-temporal Neural Structural Causal Models for Bike Flow Prediction;pan deng 等;htttps:/arxiv.org/pdf/2301.07843.pdf;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116307275A (en) | 2023-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pan et al. | Urban traffic prediction from spatio-temporal data using deep meta learning | |
| He et al. | Customer preference and station network in the London bike-share system | |
| Zhang et al. | Taxi demand prediction using parallel multi-task learning model | |
| CN112489426B (en) | Urban traffic flow space-time prediction scheme based on graph convolution neural network | |
| US20240054321A1 (en) | Traffic prediction | |
| Wang et al. | An improved robust principal component analysis model for anomalies detection of subway passenger flow | |
| US20240143999A1 (en) | Multi-modal data prediction method based on causal markov model | |
| Huang et al. | Multi-view dynamic graph convolution neural network for traffic flow prediction | |
| Deng et al. | The pulse of urban transport: Exploring the co-evolving pattern for spatio-temporal forecasting | |
| CN113051811A (en) | Multi-mode short-term traffic jam prediction method based on GRU network | |
| CN115204477A (en) | Bicycle flow prediction method of context awareness graph recursive network | |
| CN116307152A (en) | Traffic prediction method for space-time interactive dynamic graph attention network | |
| Wang et al. | Predicting electric vehicle charging demand using a heterogeneous spatio-temporal graph convolutional network | |
| Zhang et al. | Spatio-temporal digraph convolutional network-based taxi pickup location recommendation | |
| Liu et al. | Exploring the impact of spatiotemporal granularity on the demand prediction of dynamic ride-hailing | |
| CN116307293B (en) | Urban space-time data prediction method based on hybrid perception and causal depolarization | |
| Zhang et al. | Road section traffic flow prediction method based on the traffic factor state network | |
| CN116307275B (en) | Bicycle flow prediction method based on airspace causal transmission | |
| CN112382081A (en) | Traffic flow prediction method based on multiple tasks | |
| CN110490365B (en) | Method for predicting network car booking order quantity based on multi-source data fusion | |
| Li et al. | Multigraph aggregation spatiotemporal graph convolution network for ride-hailing pick-up region prediction | |
| Zou et al. | A feature extraction and deep learning approach for network traffic volume prediction considering detector reliability | |
| Yao et al. | Forecasting passenger flow distribution on holidays for urban rail transit based on destination choice behavior analysis | |
| Vankdoth et al. | Deep intelligent transportation system for travel time estimation on spatio-temporal data | |
| Yuan et al. | A novel learning approach for citywide crowd flow prediction |
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 |