CN112382089A - Traffic junction node flow prediction method based on road network directed graph and parallel long-time memory network - Google Patents
Traffic junction node flow prediction method based on road network directed graph and parallel long-time memory network Download PDFInfo
- Publication number
- CN112382089A CN112382089A CN202011251804.0A CN202011251804A CN112382089A CN 112382089 A CN112382089 A CN 112382089A CN 202011251804 A CN202011251804 A CN 202011251804A CN 112382089 A CN112382089 A CN 112382089A
- Authority
- CN
- China
- Prior art keywords
- traffic
- node
- network
- road
- junction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 230000015654 memory Effects 0.000 title claims abstract description 19
- 238000012850 discrimination method Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 14
- 238000011144 upstream manufacturing Methods 0.000 claims description 12
- 239000006185 dispersion Substances 0.000 claims description 4
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000013136 deep learning model Methods 0.000 claims 1
- 238000012549 training Methods 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000010276 construction Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 241001123248 Arma Species 0.000 description 1
- 241000728173 Sarima Species 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- 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"
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Human Resources & Organizations (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a traffic junction node flow prediction method based on a road network directed graph and a parallel long-time memory network. The invention mainly comprises the following steps: (1) a hub node discrimination method based on a traffic network directed graph; (2) traffic hub node flow prediction method based on parallel long-time memory network. The traffic junction node flow high-precision prediction is realized by identifying the traffic network junction nodes and training a junction node traffic flow prediction model.
Description
Technical Field
The invention relates to the technical field of deep learning and intelligent traffic, in particular to a traffic junction node flow prediction method based on a road network directed graph and a parallel long-time and short-time memory network.
Background
In recent years, with rapid economic development and accelerated urbanization progress, the number of motor vehicles kept in the country has increased significantly. By 6 months in 2020, the nationwide quantity of motor vehicles is up to 3.6 hundred million, wherein 2.7 hundred million vehicles are in the automobile, and 4.4 hundred million drivers are in the motor vehicle. With the rapid increase of the reserved quantity of motor vehicles, the problem of traffic congestion becomes more serious, and the problem becomes one of the problems to be solved urgently in developed countries and developing countries. The traffic junction node is used as a key node for traffic conversion in a traffic network structure, and the strength of the traffic capacity of the traffic junction node is directly related to the service level of a road network in a traffic area. Therefore, if the traffic flow of the traffic junction node can be accurately predicted, a favorable reference can be provided for relieving the traffic jam problem, and effective help can be provided for travel planning of related personnel. Therefore, how to accurately predict the traffic junction node flow by combining the deep learning technology so as to improve the traffic jam condition is very urgent and has practical significance.
Traffic hub node flow prediction is widely applied, and is mainly divided into three types of prediction methods at present: a statistical-based prediction method, a machine learning-based prediction method, and a deep learning-based prediction method. The statistical prediction method is a typical time series method such as ARIMA or SARIMA of ARMA and its variants. The method is simple to realize, can capture the time dependency in the time sequence data, but cannot model the spatial dependency of a traffic network, and has poor accuracy of a prediction result; in addition, such methods are inadequate for complex nonlinear data processing because they require data to meet certain preconditions, but real-life traffic flow data often have difficulty meeting these assumptions. The prediction method based on machine learning refers to nonparametric models such as K neighbors, support vector machines and Bayesian networks. The method can effectively model the complex nonlinear traffic data, but cannot consider the space-time correlation of the traffic data at the same time, and is not comprehensive enough in time dimension, for example, various inherent characteristics such as periodicity, tendency and the like are ignored; and such algorithms generally require extensive feature engineering. In contrast, the feature selection process of the deep learning prediction method is performed automatically through a general learning process without any human intervention. The method can efficiently process complex nonlinear traffic flow data, and can simultaneously capture the time dependence and the spatial dependence of traffic flow data, and the prediction performance of the method can be further improved.
Therefore, the traffic hub node flow can be predicted by using a deep learning method to obtain a relatively ideal effect. The long-term memory network LSTM is a time-cycling neural network that is commonly used to process and predict time series data. Because the traffic hub node flow prediction needs to judge the hub nodes in the road network and the flow of the hub nodes has a confluence/diversion condition, which is more complex compared with the general node flow prediction, the traffic hub node flow prediction is realized by constructing a road network directed graph and combining a parallel long-time and short-time memory network method.
Disclosure of Invention
The invention aims to predict the traffic hub node flow and improve the traffic jam problem.
Therefore, the invention provides a traffic junction node flow prediction method based on a road network directed graph and a parallel long-time memory network, which mainly comprises the following two contents:
(1) a hub node discrimination method based on a traffic network directed graph;
(2) traffic hub node flow prediction method based on parallel long-time memory network.
The specific contents are as follows:
constructing a traffic network directed graph by adopting the method (1) to realize the judgment of traffic junction nodes; and (3) mining the time-space characteristics of the junction nodes and constructing a traffic junction node flow prediction model by adopting the method (2), so that the traffic junction node flow in a certain period of time in the future is predicted. The specific algorithm is as follows:
(1) hub node discrimination method based on traffic network directed graph
Constructing a traffic network directed graph: the traffic junction node flow prediction needs to predict the flow of all junction nodes of the whole traffic network in a certain future period according to historical time sequence data, and the spatial dependence of the junction nodes can be clearly reflected by constructing a traffic network directed graph.
The traffic network directed graph is defined as a directed graph G ═ (V, E), V is a node set, and denotes a number of each link, V ═ V (V, E)1,v2,…,vn) N is the number of nodes; e is an edge set, representing connectivity between nodes; a is an element of Rn×nIs the adjacency matrix of fig. G.
And (3) judging the traffic junction nodes: the traffic junction nodes are key road sections causing traffic congestion, and are divided into junction nodes and non-junction nodes according to the concentration degree of convergence and dispersion of upstream and downstream road sections of each road section. The concentration degree of convergence and dispersion of each road section is judged by the sum of the out-degree number out _ sum and the in-degree number in _ sum of each node calculated by traversing the directed graph, and when the set threshold Thr is more than or equal to 3, the road section corresponding to the node is considered as a junction road section, otherwise, the road section is a non-junction road section. The calculation formula is as follows:
Thr＝out_sum+in_sum
(2) traffic junction node flow prediction method based on parallel long-time and short-time memory network
And extracting node travel time data of all upstream and downstream parallel road sections of the hub nodes in the road network directed graph to construct a hub node space-time characteristic matrix, and preprocessing the data to be used as the input of the model. The extracted data of each parallel node of the junction node are two-dimensional data, the transverse dimension represents the change of a time domain, and the longitudinal dimension represents the observation index of the node, namely the average travel time of the node road section. Combining the data of all parallel nodes together and converting the data into a time sequence matrix, setting the F ∈ (1, …, F) time sequence of each node in a road network as a flow sequence, and setting the flow in a certain time range in the future as a prediction target;
represents the average travel time of the ith node within the t-th time period (where the time periods are each 2 minutes); and (3) taking data from t-n to t-1 of all upstream and downstream parallel nodes of the junction node to predict the average travel time of the t-th time period of the junction node, wherein n is 10, namely training a traffic junction node flow prediction model by adopting a long-time memory network with the step length of 10.
Represents the average travel time of the t-n to t-1 time period of the ith parallel node of the hub node,
is the number from t-n to t-1 of all upstream and downstream parallel nodes of the ith pivot nodeAccording to the matrix, the data matrix is normalized and then used as the input of single LSTM model training. The normalization process can be realized by the following formula:
the method determines all traffic junction nodes in the traffic network, and then excavates the time sequence dependency relationship between the junction nodes and the nodes connected in parallel upstream and downstream of the junction nodes to construct a junction node space-time characteristic matrix so as to facilitate modeling.
The invention adopts a parallel long-time memory network LSTM for training and constructs a traffic hub node flow prediction model. Firstly, constructing a pivot node space-time characteristic matrix, extracting pivot node and travel time data of nodes connected in parallel at the upstream and the downstream of the pivot node to construct the pivot node space-time characteristic matrix, and preprocessing the data to form a model input data matrix. And then, constructing a traffic junction node flow prediction model, taking the space-time characteristic matrix of each junction node as input data of an LSTM unit, analyzing through the LSTM, and further modeling the traffic flow space-time dependency relationship of the junction nodes. The LSTM network introduces a new internal state ct∈RDExclusively linear, cyclic information transfer, at each instant t, internal state c of the LSTM networktHistory information up to the current time is recorded; then outputting information to the external state h of the hidden layer through a gating mechanismt∈RD. The network controls the path of information transfer by introducing a gating mechanism, and the three gates are input gates itForgetting door ftAnd an output gate ot. Internal state ctAnd an external state htThe calculation of (c) can be achieved by the following formula:
it＝σ(Wixt+Uiht-1+bi)
ft＝σ(Wfxt+Ufht-1+bf)
ot＝σ(Woxt+Uoht-1+bo)
ht＝ot⊙tanh(ct)
wherein, the input gate it∈[0,1]DControlling candidate states at the current time
How much information needs to be saved; forget door ft∈[0,1]DControlling the internal state c of the previous momentt-1How much information needs to be forgotten; and an output gate ot∈[0,1]DControlling the internal state c at the present momenttHow much information needs to be output to the external state ht；xtAn input that is a current state; h ist-1The external state at the last moment; σ (-) is a Logistic function; tan h (·) is a Tanh function; as a vector element product; c. Ct-1The memory unit at the previous moment;
are candidate states obtained by a non-linear function. Here, the activation function of the gate is set to be a sigmoid function, and the activation function of the output is a tanh function. And finally, predicting the traffic hub node flow, and predicting the flow of the hub node to be predicted in a certain future time period through the trained model.
Drawings
FIG. 1 is a technical scheme of the present invention.
Detailed Description
The invention discloses a traffic junction node flow prediction method based on a road network directed graph and a parallel long-time memory network, which combines a technical road graph shown in the attached figure 1 and comprises the following specific steps:
the method comprises the following steps: traffic network directed graph construction
According to the direction of the vehicles allowed to pass, the direct upstream road section and the direct downstream road section of each road section are given in the upstream and downstream information data of each road section, and the traffic network is constructed into a directed graph structure according to the invention content (1).
Step two: pivot node determination
The convergence and dispersion density degree of the upstream and downstream road sections of each road section is calculated by calculating the out-degree and the in-degree of each road section node in the road network directed graph, and each road section node in the traffic network is divided into a junction node (the in-degree is more than or equal to a threshold value) and a non-junction node (the in-degree is less than the threshold value).
Step three: spatio-temporal feature matrix construction
And extracting data of the upstream and downstream parallel nodes of the pivot node as the input of the model. Before that, min-max normalization processing needs to be carried out on the data according to the prediction target, and the data are converted into a space-time characteristic matrix.
Step four: traffic hub node flow prediction model construction
And (3) inputting the pivot node space-time characteristic matrix into a long-time memory network according to the invention content (2), and searching for optimal model parameters through iterative updating to construct a prediction model.
Step five: traffic hub node flow prediction
And predicting the flow of other traffic junction nodes to be predicted at a certain period of time in the future by using the trained flow prediction model.
Claims (3)
1. A traffic junction node flow prediction method based on a road network directed graph and a parallel long-time memory network is characterized by comprising the following steps:
(1) a hub node discrimination method based on a traffic network directed graph;
(2) traffic hub node flow prediction method based on parallel long-time memory network.
2. A method for discriminating hub nodes based on a traffic network directed graph according to claim 1, characterized in that: aiming at the traffic junction node flow prediction problem, all junction node road sections in a traffic road network need to be distinguished before modeling, a traffic road network directed weightless graph is constructed based on a traffic road network topological structure, the outgoing degree and the incoming degree of all nodes in a road network directed graph are calculated and used for reflecting the convergence and dispersion density degree of each road section node, a threshold value Thr is set, when Thr is larger than or equal to 3, the road section corresponding to the node is a junction road section, otherwise, the road section is a non-junction road section, and all the junction node road sections in the traffic road network are finally distinguished.
3. The traffic junction node flow prediction method based on the parallel long-time memory network according to claim 1, characterized in that: aiming at the problem of low accuracy of traffic flow prediction of a time sequence method and a machine learning method, a deep learning model based on a parallel long-time memory network is adopted, node travel time data of parallel connection road sections of upstream and downstream of a hub node in a road network directed graph are extracted, a hub node space-time characteristic matrix is constructed, the long-time memory network with the step length of 10 is adopted to train a traffic hub node flow prediction model, the time sequence dependency relationship between the hub node and the parallel nodes of the upstream and downstream of the hub node is mined, and the high-accuracy prediction of the flow of the traffic hub node in a certain period of the future is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011251804.0A CN112382089A (en) | 2020-11-11 | 2020-11-11 | Traffic junction node flow prediction method based on road network directed graph and parallel long-time memory network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011251804.0A CN112382089A (en) | 2020-11-11 | 2020-11-11 | Traffic junction node flow prediction method based on road network directed graph and parallel long-time memory network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112382089A true CN112382089A (en) | 2021-02-19 |
Family
ID=74578664
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011251804.0A Pending CN112382089A (en) | 2020-11-11 | 2020-11-11 | Traffic junction node flow prediction method based on road network directed graph and parallel long-time memory network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112382089A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107464419A (en) * | 2017-08-28 | 2017-12-12 | 北京工业大学 | A kind of Short-time Traffic Flow Forecasting Methods for considering space-time characterisation |
| CN109410575A (en) * | 2018-10-29 | 2019-03-01 | 北京航空航天大学 | A kind of road network trend prediction method based on capsule network and the long Memory Neural Networks in short-term of nested type |
| CN110070713A (en) * | 2019-04-15 | 2019-07-30 | 浙江工业大学 | A kind of traffic flow forecasting method based on two-way nested-grid ocean LSTM neural network |
| CN111091712A (en) * | 2019-12-25 | 2020-05-01 | 浙江大学 | Traffic flow prediction method based on cyclic attention dual graph convolution network |
| CN111275971A (en) * | 2020-02-18 | 2020-06-12 | 山西交通控股集团有限公司 | Expressway traffic flow prediction method |
-
2020
- 2020-11-11 CN CN202011251804.0A patent/CN112382089A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107464419A (en) * | 2017-08-28 | 2017-12-12 | 北京工业大学 | A kind of Short-time Traffic Flow Forecasting Methods for considering space-time characterisation |
| CN109410575A (en) * | 2018-10-29 | 2019-03-01 | 北京航空航天大学 | A kind of road network trend prediction method based on capsule network and the long Memory Neural Networks in short-term of nested type |
| US20200135017A1 (en) * | 2018-10-29 | 2020-04-30 | Beihang University | Transportation network speed foreeasting method using deep capsule networks with nested lstm models |
| CN110070713A (en) * | 2019-04-15 | 2019-07-30 | 浙江工业大学 | A kind of traffic flow forecasting method based on two-way nested-grid ocean LSTM neural network |
| CN111091712A (en) * | 2019-12-25 | 2020-05-01 | 浙江大学 | Traffic flow prediction method based on cyclic attention dual graph convolution network |
| CN111275971A (en) * | 2020-02-18 | 2020-06-12 | 山西交通控股集团有限公司 | Expressway traffic flow prediction method |
Non-Patent Citations (2)
| Title |
|---|
| 满春涛 等: "考虑上下游的LSTM短时交通流量预测", 《哈尔滨理工大学学报》 * |
| 王伟: "城市道路交通网络关键节点和路段辨识方法研究", 《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅱ辑》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109754605B (en) | Traffic prediction method based on attention temporal graph convolution network | |
| CN111612243B (en) | Traffic speed prediction method, system and storage medium | |
| US6470261B1 (en) | Automatic freeway incident detection system and method using artificial neural network and genetic algorithms | |
| Smagulova et al. | Overview of long short-term memory neural networks | |
| CN112382082A (en) | Method and system for predicting traffic running state in congested area | |
| CN113268916A (en) | Traffic accident prediction method based on space-time graph convolutional network | |
| CN112270355B (en) | Active safety prediction method based on big data technology and SAE-GRU | |
| CN114519932B (en) | Regional traffic condition integrated prediction method based on space-time relation extraction | |
| CN113762595B (en) | Traffic time prediction model training method, traffic time prediction method and equipment | |
| Peeta et al. | Adaptability of a hybrid route choice model to incorporating driver behavior dynamics under information provision | |
| CN113704956A (en) | Urban road online microscopic simulation method and system based on digital twin technology | |
| CN113538910A (en) | Self-adaptive full-chain urban area network signal control optimization method | |
| Salehinejad et al. | Dynamic Fuzzy Logic‐Ant Colony System‐Based Route Selection System | |
| Zhou et al. | Grouptron: Dynamic multi-scale graph convolutional networks for group-aware dense crowd trajectory forecasting | |
| CN115206092A (en) | Traffic prediction method of BiLSTM and LightGBM model based on attention mechanism | |
| Srinivasan et al. | Computational intelligence-based congestion prediction for a dynamic urban street network | |
| Lu et al. | Leveraging graph neural network with lstm for traffic speed prediction | |
| CN117671992A (en) | Intelligent bus dispatching method and system | |
| CN115273466A (en) | Monitoring method and system based on flexible lane management and control algorithm | |
| CN112927507B (en) | Traffic flow prediction method based on LSTM-Attention | |
| CN112989539B (en) | Traffic prediction method based on intersection transfer calculation | |
| Dai et al. | Short-term traffic flow prediction: an ensemble machine learning approach | |
| CN115762147B (en) | Traffic flow prediction method based on self-adaptive graph meaning neural network | |
| CN112529294A (en) | Training method, medium and equipment for individual random trip destination prediction model | |
| CN112562312A (en) | GraphSAGE traffic network data prediction method based on fusion characteristics |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210219 |
|
| RJ01 | Rejection of invention patent application after publication |