CN115423212A - Demand forecasting method and device - Google Patents
Demand forecasting method and device Download PDFInfo
- Publication number
- CN115423212A CN115423212A CN202211196283.2A CN202211196283A CN115423212A CN 115423212 A CN115423212 A CN 115423212A CN 202211196283 A CN202211196283 A CN 202211196283A CN 115423212 A CN115423212 A CN 115423212A
- Authority
- CN
- China
- Prior art keywords
- determining
- correlation
- target
- area
- time period
- 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
- 238000013277 forecasting method Methods 0.000 title description 2
- 238000009826 distribution Methods 0.000 claims abstract description 90
- 238000000034 method Methods 0.000 claims abstract description 59
- 238000003860 storage Methods 0.000 claims abstract description 12
- 238000004891 communication Methods 0.000 claims description 24
- 230000007246 mechanism Effects 0.000 claims description 23
- 238000012545 processing Methods 0.000 claims description 18
- 230000002123 temporal effect Effects 0.000 claims description 10
- 238000004590 computer program Methods 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 21
- 238000010276 construction Methods 0.000 description 19
- 230000008569 process Effects 0.000 description 14
- 239000013598 vector Substances 0.000 description 14
- 230000006870 function Effects 0.000 description 9
- 239000000126 substance Substances 0.000 description 5
- 238000012512 characterization method Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 101000822695 Clostridium perfringens (strain 13 / Type A) Small, acid-soluble spore protein C1 Proteins 0.000 description 1
- 101000655262 Clostridium perfringens (strain 13 / Type A) Small, acid-soluble spore protein C2 Proteins 0.000 description 1
- 101000655256 Paraclostridium bifermentans Small, acid-soluble spore protein alpha Proteins 0.000 description 1
- 101000655264 Paraclostridium bifermentans Small, acid-soluble spore protein beta Proteins 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000029052 metamorphosis Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- 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
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Educational Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
According to embodiments of the present disclosure, a method, an apparatus, an electronic device, a computer storage medium, and a computer program product for demand forecasting are provided. The method described herein comprises: determining description information of the target area in at least one historical time period, wherein the description information at least indicates the correlation between the service requirement of the target area and the service requirement of the associated area in the corresponding historical time period, and the correlation at least comprises at least one of the following items: location relevance, point of interest relevance, and distribution relevance; and determining the service requirement of the target area in the target time period based on the description information. Based on the above manner, the embodiment of the present disclosure can more accurately determine the service requirement of a specific area by considering the spatio-temporal correlation between regions.
Description
Technical Field
Example embodiments of the present disclosure relate generally to the field of computers, and more particularly, relate to methods, apparatuses, electronic devices, computer storage media, and computer program products for demand forecasting.
Background
Demand forecasting is an important task in today's society. People often rely on predictions of future demand to schedule corresponding resources so that the supply and demand for resources can be as balanced as possible.
For example, for transportation services, demand is often a tremendous concern with space-time. Different regions often generate different service requirements at different times. For example, for travel services such as net appointments, greater service demands may be generated at weekday peak evening hours, locations such as industrial parks.
Disclosure of Invention
In a first aspect of the disclosure, a method of demand forecasting is provided. The method comprises the following steps: determining description information of the target area in at least one historical time period, wherein the description information at least indicates the correlation between the service requirement of the target area and the service requirement of the associated area in the corresponding historical time period, and the correlation at least comprises at least one of the following items: the method comprises the steps of obtaining a target area and an associated area, wherein the target area and the associated area are distributed in a time unit according to the position correlation, the interest point correlation and the distribution correlation, the position correlation is used for indicating the position relation between the target area and the associated area, the interest point correlation indicates the similarity of interest point distribution in the target area and the associated area, and the distribution correlation is used for indicating the similarity of service demand distribution of the target area and the associated area in the time unit; and determining the service requirement of the target area in the target time period based on the description information.
In a second aspect of the present disclosure, an apparatus for demand forecasting is provided. The device includes: a description module configured to determine description information of the target area for at least one historical time period, the description information at least indicating a correlation between service requirements of the target area and service requirements of the associated area within the respective historical time period, the correlation including at least one of: the method comprises the steps of obtaining a target area and an associated area, wherein the target area and the associated area are distributed in a time unit according to the position correlation, the interest point correlation and the distribution correlation, the position correlation is used for indicating the position relation between the target area and the associated area, the interest point correlation indicates the similarity of interest point distribution in the target area and the associated area, and the distribution correlation is used for indicating the similarity of service demand distribution of the target area and the associated area in the time unit; and a prediction module configured to determine a service demand of the target area for the target time period based on the description information.
In a third aspect of the present disclosure, there is provided an electronic device comprising: a memory and a processor; wherein the memory is for storing one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method according to the first aspect of the disclosure.
In a fourth aspect of the present disclosure, there is provided a computer readable storage medium having stored thereon one or more computer instructions, wherein the one or more computer instructions are executed by a processor to implement a method according to the first aspect of the present disclosure.
In a fifth aspect of the present disclosure, a computer program product is provided comprising computer executable instructions, wherein the computer executable instructions, when executed by a processor, implement the method according to the first aspect of the present disclosure.
The embodiment of the disclosure can simultaneously consider the space-time characteristics of the region in the process of demand prediction, thereby improving the accuracy of the demand prediction.
Drawings
The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. In the drawings, like or similar reference characters designate like or similar elements, and wherein:
FIG. 1 illustrates a schematic diagram of an example environment in which embodiments in accordance with the present disclosure may be implemented;
FIG. 2 illustrates a schematic diagram of an example architecture of a demand forecasting apparatus, in accordance with some embodiments of the present disclosure;
FIG. 3 illustrates a schematic diagram of an example implementation of a demand forecasting apparatus, in accordance with some embodiments of the present disclosure;
FIG. 4 illustrates a schematic diagram of an example process of demand forecasting, in accordance with some embodiments of the present disclosure;
FIG. 5 illustrates a schematic block diagram of an apparatus for demand forecasting, in accordance with some embodiments of the present disclosure;
FIG. 6 illustrates a block diagram of an electronic device capable of implementing various embodiments of the present disclosure.
Detailed Description
Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided for a more complete and thorough understanding of the present disclosure. It should be understood that the drawings and embodiments of the disclosure are for illustration purposes only and are not intended to limit the scope of the disclosure.
In describing embodiments of the present disclosure, the terms "include" and its derivatives should be interpreted as being inclusive, i.e., "including but not limited to. The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first," "second," and the like may refer to different or the same objects. Other explicit and implicit definitions are also possible below.
As discussed above, demand forecasting of services is an important task in people's daily life. In transportation, demand prediction of traffic service is an important basis for coordinating traffic resources by a platform.
In transportation, the demand of transportation services exhibits a strong temporal and spatial correlation. That is, the demand of different regions at different times may exhibit great variability. Taking travel services such as network taxi appointment and the like as an example, the demand of a residential area may be relatively large at the early peak stage of a working day; in contrast, the demand on an industrial park may be relatively greater during the evening peak hours of the work day.
Conventional demand forecasting typically takes into account the timing characteristics of the demand to make the forecast. For example, one might predict the future in terms of service demand for an area over time. However, such predictions ignore the correlation of travel services with space, resulting in a prediction that may be difficult to accurately make.
In view of this, embodiments of the present disclosure provide a solution for demand prediction. According to the solution, first, description information of the target area for at least one historical time period is determined, wherein the description information at least indicates a correlation between the service requirements of the target area and the service requirements of the associated area within the corresponding historical time period. In particular, the correlation may comprise at least one of: the method comprises the steps of obtaining a target area and an associated area, obtaining location correlation, interest point correlation and distribution correlation, wherein the location correlation is used for indicating the location relation between the target area and the associated area, the interest point correlation is used for indicating the similarity of interest point distribution in the target area and the associated area, and the distribution correlation is used for indicating the similarity of service demand distribution of the target area and the associated area in a plurality of time units.
Further, a service requirement of the target area for the target time period may be determined based on the description information.
It can be seen that in making the service requirement determination, embodiments of the present disclosure not only utilize temporal characteristics, but also take into account the correlation between the geographic region and other associated regions. Thus, embodiments of the present disclosure enable more accurate service demand prediction based on spatiotemporal characteristics of the service.
Various example implementations of this scheme will be described in detail below with reference to the accompanying drawings.
Example Environment
Referring initially to FIG. 1, a schematic diagram of an example environment 100 is schematically illustrated in which example implementations according to the present disclosure may be used. As shown in fig. 1, environment 100 may include a predictive device 120. Such a prediction device 120 may be a suitable type of computing device, including but not limited to: cloud computing devices, edge computing devices, terminal computing devices, combinations thereof, and the like.
As shown in fig. 1, the prediction device 120 may determine the service demand 130 for the geographic area 110 based on the descriptive information for each geographic area 110. In some embodiments, such geographic areas 110 may be partitioned based on latitude and longitude information.
As an example, as shown in fig. 1, each of the geographic areas 110 may have, for example, a rectangular shape of the same size. In some embodiments, the geographic area 110 may have a shape other than rectangular, for example, or the size or shape of the different geographic areas may be different. For example, different geographical areas may also correspond to different business circles within a city, and may even correspond to different cities, provinces or regions, etc.
As will be described in detail below, the forecasting device 120 may forecast service requirements for particular geographic areas 110 based on characteristics of the service requirements for the different geographic areas. In some embodiments, such service requirements may include requirements for transportation services, such as requirements for travel services like internet appointments, requirements for goods delivery services like courier, requirements for designated drive services, and the like. Alternatively, such services may also include other types of services having relatively strong spatio-temporal characteristics, such as power usage services, telecommunication services, take-away services, and the like.
It should be understood that the aspects of the present disclosure will be described below with a traffic service as an example for convenience of description only.
Example architecture
FIG. 2 illustrates a schematic diagram of an example architecture of a demand forecasting apparatus, in accordance with some embodiments of the present disclosure. As shown in fig. 2, the demand forecasting apparatus may include three functional parts, namely: a graph construction unit 210, an attention unit 220, and a timing prediction unit 230.
In some embodiments, the graph construction unit 210 may generate spatio-temporal dissimilarity graphs to describe spatio-temporal distribution characteristics of service needs for different geographic regions.
In some embodiments, the attention unit 220 may be based on a spatiotemporal metamorphosis map to generate a feature representation, also referred to as descriptive information, for a target geographic region (also referred to as target region) utilizing an attention mechanism.
In some embodiments, the timing prediction unit 230 may then determine the service requirements of the target geographic area for a particular time period based on the characterization of the area.
A specific implementation of each unit will be described in detail below with reference to fig. 3.
Construction of spatio-temporal metamerism
In some embodiments, the graph construction unit 210 may construct the spatio-temporal anomaly graph based on historical service requirements of different geographic regions over at least one historical time period.
First, the graph building unit 210 may divide a plurality of geographic regions, e.g., the geographic regions 110 shown in fig. 1, based on longitude and latitude, for example. Taking service demand prediction for different areas within a city as an example, the graph building unit 210 may divide a city space into a plurality of geographic areas (also referred to as geographic grids) according to a fixed area size, for example.
Illustratively, a city, for example, may be divided intoH×WA geographical area and can usev i (i=(h-1)×W+w; i∈[1,H×W]; h∈[1,H]; w∈[1,W]) To representHDimension ofh th And is andWdimension ofw th The geographic area of (a). Thus, each geographic region may correspond to a node in the graph, and a set of nodes may be represented as 。
Further, the spatio-temporal heteromorphic graph to be constructed can be expressed as WhereinA set of nodes is represented that is,representing a set of edges (or a set of edge connections).Then representing a node type map and an edge type map respectively,a set of node types is represented that,representing a set of edge types.
As shown in fig. 3, for the spatio-temporal heteromorphic graph to be constructed, it may include a plurality of layers corresponding to different historical time periods, each layer having nodes corresponding to different regions therein, and edge connections of the nodes within each layer may be the same. Such edge connections may be used, for example, to indicate a correlation between service requirements for different geographic regions to which different nodes correspond.
The types of edge connections in the spatio-temporal profile graph will be described below.
nbrPosition correlation ()
In some embodiments, graph construction unit 210 may determine a location correlation based on a location relationship between different geographic regions, thereby constructing a first type of edge connection that indicates the location correlation.
In some embodiments, if two geographic regions are adjacent, graph building unit 210 may establish a first type of edge connection between two nodes corresponding to the two geographic regions, forming a subgraph indicating a positional relationship(s) ((s))nbr). Conversely, if two geographic regions are not adjacent, their corresponding nodes will not have edges connected in the subgraph.
Alternatively, the graph construction unit 210 may also construct a fully-connected subgraph and determine the distances between different geographic regions as weights for edges between corresponding nodes in the subgraph.
In this way, sub-graphs (nbr) It is possible to represent the position correlation between the nodes.
poiPoint of interest relevance ()
In some embodiments, the graph construction unit 210 may determine the point of interest relevance based on similarity of point of interest distributions within different geographic regions, thereby constructing a second type of edge connection for indicating the point of interest relevance.
In particular, graph construction unit 210 may determine a distribution of points of interest within each geographic region, where the distribution may indicate the number of points of interest within the corresponding region that contain the specified type. As an example, the graph construction unit 210 may construct a multi-dimensional feature vector, where each dimension corresponds to a particular type of point of interest, and the value of the vector corresponds to the number of points of interest of that type contained within the geographic region.
Further, the graph building unit 210 may determine the interest point relevance between two regions based on the similarity of interest point distribution between each geographic region. Illustratively, the graph construction unit 210 may determine the similarity of its points of interest based on the distance or included angle between the multi-dimensional feature vectors.
In some embodiments, if the point of interest similarity is greater than a predetermined threshold, graph building unit 210 may establish a second type of edge connection between two nodes corresponding to the two geographic areas, forming a subgraph indicating the point of interest similarity(s) ((s))poi). Conversely, if the point of interest similarity of two geographic regions is less than or equal to a predetermined threshold, then their corresponding nodes will have no edge connections in the subgraph.
In some embodiments, such a threshold may be a dynamic or static threshold. For example, the graph construction unit 210 may always keep a predetermined number of nodes having the greatest similarity to the point of interest of a specific node to construct the second-type edge connection, for example, regardless of the absolute magnitude of the similarity thereof. In this case, the threshold may be understood as a dynamic threshold based on similarity ranking.
Alternatively, the graph construction unit 210 may also construct a fully connected subgraph and directly determine the point of interest similarity between different geographic regions as weights for edges between corresponding nodes in the subgraph.
In this way, sub-graphs (poi) Can represent the similarity of the distribution of interest points in each node. Such similarity can help make more accurate demand forecasts. For example, if both areas belong to an office building business, they may exhibit closer service demand characteristics.
tpsDistribution dependence ()
In some embodiments, the graph construction unit 210 may determine the distribution correlations based on similarities in service demand distributions of different geographic regions over multiple time units, thereby constructing a third type of edge connection for indicating the distribution correlations.
In particular, the graph building unit 210 may determine the service demand distribution based on the service demands of each geographic region over a plurality of time units (e.g., 24 hours). Such a service demand distribution may, for example, indicate hourly service demand (e.g., average service demand over each hour of the day for the past month) for the corresponding geographic region over 24 hours, such that a service demand distribution may be formed.
As an example, graph building unit 210 may build a multidimensional feature vector, where each dimension corresponds to a particular time unit and the value of the vector corresponds to the service demand of the geographic area in the corresponding time unit.
Further, the graph building unit 210 may determine the distribution correlation between two regions based on the similarity of the service demand distribution between each geographic region and each other. Illustratively, the graph construction unit 210 may determine the distribution similarity thereof based on the distance or the included angle between the multi-dimensional feature vectors.
In some embodiments, if the distribution similarity is greater than a predetermined threshold, graph building unit 210 may establish a third type of edge connection between two nodes corresponding to the two geographic regions, forming a subgraph for indicating the distribution similarity(s) ((s))tps). On the contrary, ifIf the distribution similarity of two geographic regions is less than or equal to a predetermined threshold, then their corresponding nodes will not have edges connected in the subgraph.
In some embodiments, such a threshold may be a dynamic or static threshold. For example, the graph construction unit 210 may always keep a predetermined number of nodes having the greatest similarity to the distribution of the specific nodes to construct the second-type edge connection, for example, regardless of the absolute magnitude of the similarity thereof. In this case, the threshold may be understood as a dynamic threshold based on similarity ranking.
Alternatively, the graph construction unit 210 may also construct a fully-connected subgraph and directly determine the distribution similarity between different geographic regions as weights for edges between corresponding nodes in the subgraph.
In this way, subgraphs (tps) Can represent the similarity of service distribution within each node. Such distribution similarity can help make more accurate demand predictions.
timTemporal adjacency ()
In some embodiments, graph building unit 210 may establish edge connections across layers. It should be understood that such edge connections are intended to represent temporal characteristics of service needs of the same geographic area.
In particular, the graph construction unit 210 may establish edge connections for corresponding nodes (corresponding to the same geographical area) in layers corresponding to two adjacent time periods based on adjacency between the time periods, thereby forming a subgraph: (a subgraph)tim)。
tpmTime periodicity ()
In some embodiments, graph building unit 210 may also establish edge connections for corresponding nodes (corresponding to the same geographic area) in a layer corresponding to a particular time period based on periodicity between the time periods, thereby forming a subgraph: (sub-graph: (i)tpm)。
The periodicity of the time period may for example represent: the two time periods correspond to the same hour in different days, to the same day in different weeks, to the same day in different months, to the same month in different years, etc.
In reality, service demands are also typically time-periodic. For example, taking travel service as an example, the service needs of the same geographic area during morning peak hours of the week are generally similar.
In this way, sub-graphs (tpm) The periodicity of the corresponding time can be represented. Such periodicity can help make more accurate demand predictions.
Based on the above discussion, node type setsFor example, only one node type, a set of edge types, may be includedE.g. may comprisenbr, poi, tps, tim, tpm}。
Meta path (meta-path)
In some embodiments, the graph construction unit 210 may consider not only the edge connection types discussed above, but also the communication paths in the spatio-temporal metamorphic graph. Such a communication path may for example represent a combination of edge connections, e.g. a tonenbr-nbr, poi-nbr, poi-tpsTherein ofnbr-nbrCan indicate that two nodes exist between the target node and the target nodenbrThe communication paths formed by the edge connections may represent, for example, adjacent regions of adjacent regions;poi-nbrcan indicate that the target node has the routingpoiEdge connection andnbrthe communication paths formed by the edge connections, for example, the adjacent areas of the areas with similar interest point distribution;poi-tpscan indicate that the target node has the routingpoiEdge connection andtpsthe communication paths formed by the edge connections, for example, the regions with similar service demand distribution types of the regions with the interest point distribution.
It will be appreciated that the above definitions of specific communication paths are merely exemplary and that communication paths of any suitable combination of greater lengths may be defined as desired.
Based on this also manner, the graph construction unit 210 can construct a union of the edge connection type and the communication path type in the graph as a meta path (also referred to as meta-path) as basic data for predicting a service demand below. For example, meta-pathsCan be represented asnbr, poi, tps, tim, tpm, nbr-nbr, poi-nbr, poi-tp}。
Attention mechanism
In some embodiments, the attention module 220 may update the feature representation of each node according to the meta-path using an attention mechanism.
As shown in FIG. 3, attention module 220 may implement two levels of attention mechanisms: a node level attention mechanism and a meta-path level attention mechanism.
In some embodiments, the node level attention mechanism processing procedure represents a procedure for determining a plurality of feature representations of the target node corresponding to different meta-paths using the associated feature representations of the associated nodes. Such associated nodes may include nodes connected to the target node by meta-paths. This process may be represented, for example, as:
wherein the content of the first and second substances,representing nodesIn thattA feature vector (e.g., service demand) over a period of time, andis indicated by way oftMeta-paths within a time periodResulting nodeIs performed on the target node.
It should be appreciated that any suitable attention mechanism algorithm may be employed to implement the process as represented by equation (1), e.g., a multi-head attention mechanism, etc.
In one example implementation, the process of this attention mechanism may be represented as:
wherein the content of the first and second substances,representation and target nodeThrough meta pathThe set of all the nodes that are connected,is a target nodeAbout meta pathThe query vector of (a) is,respectively representing target nodesAbout meta pathThe key vector (key vector),representing meta pathThe projection matrix in the node level attention,is shown intTime period associated nodeRelative to the target nodeAbout meta pathThe attention score of (a) is given,representing a target nodeThe vector of values of the associated nodes of (a),representing meta pathsReLU, which represents the modified linear unit activation function.
In some embodiments, the meta-path level attention mechanism processing procedure represents a procedure for generating an updated feature representation of the target node using corresponding feature representations of a plurality of meta-paths, and the meta-path level attention mechanism processing procedure may be represented as:
it should be appreciated that any suitable attention mechanism algorithm may be employed to implement the process as represented by equation (8), e.g., a multi-head attention mechanism, etc.
In one example implementation, the process of this attention mechanism may be represented as:
wherein the content of the first and second substances,respectively representing target nodes in meta-path attention mechanismQuery vector, key vector and value vector, W andrepresenting parameters in the attention mechanism, reLU representing a modified linear unit activation function,is shown intTime period target nodeAnd (4) representing the target characteristics after the meta-path attention mechanism processing.
Timing prediction
As shown in FIG. 3, the timing prediction module 230 may utilize a target nodeCharacterization at different time periodsTo determine the service requirements of the target geographic area for the target time period.
It should be appreciated that any suitable temporal prediction model may be employed to process the target nodeCharacterization at different time periodsTo determine the service requirements.
Illustratively, as shown in fig. 3, the timing prediction module may employ a gated round-robin unit structure (GRU) for processing, which may be represented as:
wherein the content of the first and second substances,parameters representing a Fully Connected Network (FCN),respectively representing target nodesIn a period of timetAndt-a candidate hidden state in 1,to representThe operation of the cascade of (a) and (b),representation for target nodeParameters of the GRU model of (1).
The above introduces a process of predicting service demand using a time-space diversity scheme. It should be appreciated that the model as shown in FIG. 3 may be trained based on huber losses, for example, which may be expressed as:
wherein the content of the first and second substances,are respectively shown intTrue and predicted values within a time period.
Example procedure
Fig. 4 shows a flow diagram of a process 400 for trip planning in accordance with multiple embodiments of the present disclosure. Process 400 may be implemented by prediction device 120 as shown in fig. 1.
As shown in fig. 4, at block 410, the forecasting apparatus 120 determines description information of the target area for at least one historical time period, the description information indicating at least a correlation between service requirements of the target area and service requirements of the associated area within the corresponding historical time period, the correlation including at least one of: the method comprises the steps of obtaining a target area and an associated area, obtaining location correlation, interest point correlation and distribution correlation, wherein the location correlation is used for indicating the location relation between the target area and the associated area, the interest point correlation is used for indicating the similarity of interest point distribution in the target area and the associated area, and the distribution correlation is used for indicating the similarity of service demand distribution of the target area and the associated area in a plurality of time units.
At block 420, the prediction device 120 determines a service requirement for the target area for the target time period based on the description information.
In some embodiments, determining the descriptive information of the target area for the at least one historical time period comprises: constructing a graph structure based on historical service demands of a plurality of geographic areas in at least one historical time period, wherein the graph structure comprises at least one layer, the at least one layer corresponds to each time period in the at least one time period respectively, each layer in the at least one layer has the same nodes and edge connections in the layer, the nodes are used for representing the corresponding areas, and the edge connections are used for representing the correlation among different areas; and determining the description information based on the graph structure.
In some embodiments, the edge connections comprise a first type of edge connection corresponding to a position dependency, wherein the first type of edge connection is used to indicate two area neighboring regions corresponding to the connected two nodes, or a weight of the first type of edge connection is used to indicate a distance between the two areas corresponding to the connected two nodes.
In some embodiments, the edge connections comprise a second type of edge connection corresponding to a point of interest relevance, the method further comprising: determining a first distribution of interest points within the first region, the first distribution of interest points being indicative of a number of interest points of at least one type contained within the first region; determining a second point of interest distribution within a second region; and determining a point of interest correlation between the first region and the second region based on the first point of interest distribution and the second point of interest distribution.
In some embodiments, the method further comprises: in response to determining that the interest point correlation is greater than the first threshold correlation, constructing a second type edge connection between a first node corresponding to the first region and a second node corresponding to the second region; or determining a weight of a second type of edge connection between the first node and the second node based on the point of interest relevance.
In some embodiments, the edge connections comprise a third type of edge connection corresponding to a distribution dependency, the method further comprising: determining a first service demand profile associated with the third area based on service demands of the third area over a plurality of time units; determining a second service demand distribution associated with the fourth area based on service demands of the fourth area over a plurality of time units; and determining a distribution correlation between the third area and the fourth area based on the first service demand distribution and the second service demand distribution.
In some embodiments, the method further comprises: responsive to determining that the distribution correlation is greater than the second threshold correlation, constructing a third-type edge connection between a third node corresponding to the third region and a fourth node corresponding to the fourth region; or determining a weight of the third-type edge connection between the third node and the fourth node based on the distribution correlation.
In some embodiments, the at least one layer comprises a plurality of layers, and edge connections between a plurality of nodes corresponding to the same region in the plurality of layers are used to represent temporal dependencies of service requirements of the corresponding region over different time periods, the temporal dependencies being used to indicate: adjacency between time periods corresponding to different nodes, and/or periodicity between time periods corresponding to different nodes.
In some embodiments, determining the description information based on the graph structure comprises: based on the graph structure, at least one feature representation associated with the target area is determined, the at least one feature corresponding to at least one historical time period.
In some embodiments, determining the at least one feature representation associated with the target region further comprises: determining a first characteristic representation corresponding to a first historical time period of the at least one historical time period based on at least one of: a target feature representation of a target node corresponding to the first historical time period and the target area; an association characteristic representation of at least one association node having an edge connection with the target node; and an association signature representative of at least one associated node for which a particular communication path exists with the target node, the particular communication path comprising a plurality of edge connections having a particular edge connection type.
In some embodiments, determining the first feature representation comprises: determining a plurality of updated feature representations associated with the target node using a first attention mechanism based on the target feature representation and associated feature representations of the associated nodes, the updated feature representations corresponding to a particular edge connection type or communication path type; and determining the first feature representation using the second attention mechanism based on the plurality of updated feature representations.
In some embodiments, determining the service requirements of the target area for the target time period comprises: and processing the description information by using the time sequence model, and determining the service requirement of the target area in the target time period.
In some embodiments, the service requirements include requirements for a transportation service.
Example apparatus and devices
Embodiments of the present disclosure also provide corresponding apparatuses for implementing the above methods or processes. FIG. 5 illustrates a schematic block diagram of an apparatus 500 for demand forecasting, according to some embodiments of the present disclosure.
As shown in fig. 5, the apparatus 500 includes a description module 510 configured to determine description information of the target area in at least one historical time period, the description information indicating at least a correlation between service requirements of the target area and service requirements of the associated area in the corresponding historical time period, the correlation including at least one of: the method comprises the steps of obtaining a target area and an associated area, obtaining location correlation, interest point correlation and distribution correlation, wherein the location correlation is used for indicating the location relation between the target area and the associated area, the interest point correlation is used for indicating the similarity of interest point distribution in the target area and the associated area, and the distribution correlation is used for indicating the similarity of service demand distribution of the target area and the associated area in a plurality of time units.
Furthermore, the apparatus 500 further comprises a prediction module 520 configured to determine a service requirement of the target area for the target time period based on the description information.
In some embodiments, description module 510 is further configured to: constructing a graph structure based on historical service demands of a plurality of geographic areas in at least one historical time period, wherein the graph structure comprises at least one layer, the at least one layer corresponds to each time period in the at least one time period respectively, each layer in the at least one layer has the same nodes and edge connections in the layer, the nodes are used for representing the corresponding areas, and the edge connections are used for representing the correlation among different areas; and determining the description information based on the graph structure.
In some embodiments, the edge connections comprise a first type of edge connection corresponding to a position dependency, wherein the first type of edge connection is used to indicate two area neighboring regions corresponding to the connected two nodes, or a weight of the first type of edge connection is used to indicate a distance between the two areas corresponding to the connected two nodes.
In some embodiments, the edge connections comprise a second type of edge connection corresponding to a point of interest relevance, the description module 510 is further configured to: determining a first distribution of interest points within the first region, the first distribution of interest points being indicative of a number of interest points of at least one type contained within the first region; determining a second distribution of interest points within a second region; and determining a point of interest correlation between the first region and the second region based on the first point of interest distribution and the second point of interest distribution.
In some embodiments, description module 510 is further configured to: responsive to determining that the point of interest relevance is greater than a first threshold relevance, constructing a second type of edge connection between a first node corresponding to the first region and a second node corresponding to the second region; or determining a weight of a second type of edge connection between the first node and the second node based on the point of interest relevance.
In some embodiments, the edge connections comprise a third type of edge connection corresponding to a distribution dependency, the description module 510 is further configured to: determining a first service demand distribution associated with the third area based on service demands of the third area over a plurality of time units; determining a second service demand distribution associated with the fourth area based on service demands of the fourth area over a plurality of time units; and determining a distribution correlation between the third area and the fourth area based on the first service demand distribution and the second service demand distribution.
In some embodiments, description module 510 is further configured to: responsive to determining that the distribution correlation is greater than the second threshold correlation, constructing a third-type edge connection between a third node corresponding to the third region and a fourth node corresponding to the fourth region; or determining a weight of the third-type edge connection between the third node and the fourth node based on the distribution correlation.
In some embodiments, the at least one layer comprises a plurality of layers, and edge connections between a plurality of nodes corresponding to the same region in the plurality of layers are used to represent temporal dependencies of service requirements of the corresponding region over different time periods, the temporal dependencies being used to indicate: adjacency between time periods corresponding to different nodes, and/or periodicity between time periods corresponding to different nodes.
In some embodiments, description module 510 is further configured to: based on the graph structure, at least one feature representation associated with the target area is determined, the at least one feature corresponding to at least one historical time period.
In some embodiments, description module 510 is further configured to: determining a first characteristic representation corresponding to a first historical time period of the at least one historical time period based on at least one of: a target feature representation of a target node corresponding to the first historical time period and the target area; an association characteristic representation of at least one association node having an edge connection with the target node; and an association signature representative of at least one associated node for which a particular communication path exists with the target node, the particular communication path comprising a plurality of edge connections having a particular edge connection type.
In some embodiments, description module 510 is further configured to: determining a plurality of updated feature representations associated with the target node using a first attention mechanism based on the target feature representation and associated feature representations of the associated nodes, the updated feature representations corresponding to a particular edge connection type or communication path type; and determining the first feature representation using the second attention mechanism based on the plurality of updated feature representations.
In some embodiments, the prediction module 520 is configured to: and processing the description information by using the time sequence model, and determining the service requirement of the target area in the target time period.
In some embodiments, the service requirements include requirements for a transportation service.
The elements included in apparatus 500 may be implemented in a variety of ways including software, hardware, firmware, or any combination thereof. In some embodiments, one or more of the units may be implemented using software and/or firmware, such as machine executable instructions stored on a storage medium. In addition to, or in the alternative to, machine-executable instructions, some or all of the elements in apparatus 500 may be implemented at least in part by one or more hardware logic components. By way of example, and not limitation, exemplary types of hardware logic components that may be used include Field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standards (ASSPs), systems on a chip (SOCs), complex Programmable Logic Devices (CPLDs), and so forth.
Fig. 6 illustrates a block diagram of an electronic device/server 600 in which one or more embodiments of the present disclosure may be implemented. It should be understood that the electronic device/server 600 illustrated in fig. 6 is merely exemplary and should not constitute any limitation as to the functionality and scope of the embodiments described herein.
As shown in fig. 6, the electronic device/server 600 is in the form of a general-purpose electronic device. The components of the electronic device/server 600 may include, but are not limited to, one or more processors or processing units 610, memory 620, storage 630, one or more communication units 640, one or more input devices 650, and one or more output devices 660. The processing unit 610 may be a real or virtual processor and can perform various processes according to programs stored in the memory 620. In a multi-processor system, multiple processing units execute computer-executable instructions in parallel to improve the parallel processing capabilities of the electronic device/server 600.
The electronic device/server 600 typically includes a number of computer storage media. Such media may be any available media that is accessible by electronic device/server 600 and includes, but is not limited to, volatile and non-volatile media, removable and non-removable media. The memory 620 may be volatile memory (e.g., registers, cache, random Access Memory (RAM)), non-volatile memory (e.g., read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory), or some combination thereof. Storage 630 may be a removable or non-removable medium and may include a machine-readable medium, such as a flash drive, a magnetic disk, or any other medium that may be capable of being used to store information and/or data (e.g., training data for training) and that may be accessed within electronic device/server 600.
The electronic device/server 600 may further include additional removable/non-removable, volatile/nonvolatile storage media. Although not shown in FIG. 6, a magnetic disk drive for reading from or writing to a removable, non-volatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, non-volatile optical disk may be provided. In these cases, each drive may be connected to a bus (not shown) by one or more data media interfaces. Memory 620 may include a computer program product 625 having one or more program modules configured to perform the various methods or acts of the various embodiments of the disclosure.
The communication unit 640 enables communication with other electronic devices through a communication medium. Additionally, the functionality of the components of the electronic device/server 600 may be implemented in a single computing cluster or multiple computing machines, which are capable of communicating over a communications connection. Thus, the electronic device/server 600 may operate in a networked environment using logical connections to one or more other servers, network Personal Computers (PCs), or another network node.
The input device 650 may be one or more input devices such as a mouse, keyboard, trackball, or the like. Output device 660 may be one or more output devices such as a display, speakers, printer, or the like. The electronic device/server 600 may also communicate with one or more external devices (not shown), such as storage devices, display devices, etc., with one or more devices that enable a user to interact with the electronic device/server 600, or with any device (e.g., network card, modem, etc.) that enables the electronic device/server 600 to communicate with one or more other electronic devices, as desired, via the communication unit 640. Such communication may be performed via input/output (I/O) interfaces (not shown).
According to an exemplary implementation of the present disclosure, a computer-readable storage medium is provided, on which one or more computer instructions are stored, wherein the one or more computer instructions are executed by a processor to implement the above-described method.
Various aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products implemented in accordance with the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.
These computer-readable program instructions may be provided to a processing unit of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processing unit of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer-readable program instructions may also be stored in a computer-readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer-readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various implementations of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing has described implementations of the present disclosure, and the above description is illustrative, not exhaustive, and not limited to the implementations disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described implementations. The terminology used herein was chosen in order to best explain the principles of implementations, the practical application, or improvements to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the implementations disclosed herein.
Claims (16)
1. A method for demand forecasting, comprising:
determining description information of a target area in at least one historical time period, the description information at least indicating a correlation between service requirements of the target area and service requirements of associated areas in the corresponding historical time period, the correlation at least comprising at least one of: a location correlation indicating a location relationship between the target area and the associated area, a point of interest correlation indicating a similarity of point of interest distributions within the target area and within the associated area, and a distribution correlation indicating a similarity of service demand distributions of the target area and the associated area over a plurality of time units; and
determining a service requirement of the target area in a target time period based on the description information.
2. The method of claim 1, wherein determining descriptive information of the target area for at least one historical time period comprises:
constructing a graph structure based on historical service demands of a plurality of geographic regions in the at least one historical time period, wherein the graph structure comprises at least one layer, the at least one layer corresponds to each time period in the at least one time period, each layer in the at least one layer has the same nodes and edge connections in the layer, the nodes are used for representing the corresponding regions, and the edge connections are used for representing the correlation among different regions; and
determining the description information based on the graph structure.
3. The method of claim 2, wherein the edge connections comprise a first type of edge connection corresponding to the location correlation,
the first-type edge connection is used for indicating two area adjacent areas corresponding to the two connected nodes, or the weight of the first-type edge connection is used for indicating the distance between the two areas corresponding to the two connected nodes.
4. The method of claim 2, wherein the edge connection comprises a second type of edge connection corresponding to the point of interest relevance, the method further comprising:
determining a first distribution of interest points within a first region, the first distribution of interest points being indicative of a number of interest points of at least one type contained within the first region;
determining a second distribution of interest points within a second region; and
determining the point of interest relevance between the first region and the second region based on the first point of interest distribution and the second point of interest distribution.
5. The method of claim 4, further comprising:
responsive to determining that the point of interest relevance is greater than a first threshold relevance, constructing the second type of edge connection between a first node corresponding to the first region and a second node corresponding to the second region; or
Determining a weight of the second type of edge connection between the first node and the second node based on the point of interest relevance.
6. The method of claim 2, wherein the edge connection comprises a third type of edge connection corresponding to the distribution dependency, the method further comprising:
determining a first service demand profile associated with a third zone based on service demands of the third zone within the plurality of time units;
determining a second service demand distribution associated with a fourth area based on service demands of the fourth area over the plurality of time units; and
determining a distribution correlation between the third area and the fourth area based on the first service demand distribution and the second service demand distribution.
7. The method of claim 6, further comprising:
responsive to determining that the distribution correlation is greater than a second threshold correlation, constructing the third-type edge connection between a third node corresponding to the third region and a fourth node corresponding to the fourth region; or
Determining a weight of the third type of edge connection between the third node and the fourth node based on the distribution correlation.
8. The method of claim 2, wherein the at least one tier comprises a plurality of tiers, and edge connections between a plurality of nodes in the plurality of tiers corresponding to a same region are used to represent temporal dependencies of service requirements of the corresponding region over different time periods, the temporal dependencies being used to indicate: adjacency between time periods corresponding to different nodes, and/or periodicity between time periods corresponding to different nodes.
9. The method of claim 2, wherein determining the description information based on the graph structure comprises:
based on the graph structure, at least one feature representation associated with the target region is determined, the at least one feature corresponding to the at least one historical time period.
10. The method of claim 9, wherein determining at least one feature representation associated with the target region further comprises:
determining a first characteristic representation corresponding to a first historical time period of the at least one historical time period based on at least one of:
a target feature representation of a target node corresponding to the first historical time period and the target area;
an association signature representation of at least one association node having the edge connection with the target node; and
an association signature of at least one associated node for which a particular communication path exists with the target node, the particular communication path comprising a plurality of edge connections of a particular edge connection type.
11. The method of claim 10, wherein determining the first feature representation comprises:
determining a plurality of updated feature representations associated with the target node using a first attention mechanism based on the target feature representation and associated feature representations of associated nodes, the updated feature representations corresponding to a particular edge connection type or communication path type; and
determining the first feature representation using a second attention mechanism based on the plurality of updated feature representations.
12. The method of claim 1, wherein determining the service demand of the target area for a target time period comprises:
and processing the description information by utilizing a time sequence model, and determining the service requirement of the target area in the target time period.
13. The method of claim 1, wherein the service demand comprises demand for a transportation service.
14. An apparatus for demand forecasting, comprising:
a description module configured to determine description information of a target area for at least one historical time period, the description information at least indicating a correlation between service requirements of the target area and service requirements of associated areas over the respective historical time period, the correlation including at least one of: a location correlation indicating a location relationship between the target area and the associated area, a point of interest correlation indicating a similarity of point of interest distributions within the target area and within the associated area, and a distribution correlation indicating a similarity of service demand distributions of the target area and the associated area over a plurality of time units; and
a prediction module configured to determine a service requirement of the target area for a target time period based on the description information.
15. An electronic device, comprising:
a memory and a processor;
wherein the memory is to store one or more computer instructions, wherein the one or more computer instructions are to be executed by the processor to implement the method of any one of claims 1 to 13.
16. A computer readable storage medium having one or more computer instructions stored thereon, wherein the one or more computer instructions are executed by a processor to implement the method of any one of claims 1 to 13.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211196283.2A CN115423212A (en) | 2022-09-29 | 2022-09-29 | Demand forecasting method and device |
CN202211534274.XA CN115618986B (en) | 2022-09-29 | 2022-12-02 | Method and device for coordinating resources |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211196283.2A CN115423212A (en) | 2022-09-29 | 2022-09-29 | Demand forecasting method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115423212A true CN115423212A (en) | 2022-12-02 |
Family
ID=84206027
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211196283.2A Pending CN115423212A (en) | 2022-09-29 | 2022-09-29 | Demand forecasting method and device |
CN202211534274.XA Active CN115618986B (en) | 2022-09-29 | 2022-12-02 | Method and device for coordinating resources |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211534274.XA Active CN115618986B (en) | 2022-09-29 | 2022-12-02 | Method and device for coordinating resources |
Country Status (1)
Country | Link |
---|---|
CN (2) | CN115423212A (en) |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10817775B2 (en) * | 2017-01-12 | 2020-10-27 | International Business Machines Corporation | Neural network computing systems for predicting vehicle requests |
US11544535B2 (en) * | 2019-03-08 | 2023-01-03 | Adobe Inc. | Graph convolutional networks with motif-based attention |
CN111784018A (en) * | 2019-04-03 | 2020-10-16 | 北京嘀嘀无限科技发展有限公司 | Resource scheduling method and device, electronic equipment and storage medium |
US11842271B2 (en) * | 2019-08-29 | 2023-12-12 | Nec Corporation | Multi-scale multi-granularity spatial-temporal traffic volume prediction |
CN111915057B (en) * | 2020-06-28 | 2022-05-17 | 厦门大学 | Bicycle demand prediction and scheduling method based on deep learning and crowd sensing |
CN112101682B (en) * | 2020-09-25 | 2024-04-09 | 北京百度网讯科技有限公司 | Traffic pattern prediction method, traffic pattern prediction device, server and readable medium |
CN113159371B (en) * | 2021-01-27 | 2022-05-20 | 南京航空航天大学 | Unknown target feature modeling and demand prediction method based on cross-modal data fusion |
CN114117259A (en) * | 2021-11-30 | 2022-03-01 | 重庆七腾科技有限公司 | Trajectory prediction method and device based on double attention mechanism |
CN114372830A (en) * | 2022-01-13 | 2022-04-19 | 长安大学 | Network taxi booking demand prediction method based on space-time multi-graph neural network |
CN114529081B (en) * | 2022-02-18 | 2024-06-11 | 哈尔滨工程大学 | Space-time combined traffic flow prediction method and device |
-
2022
- 2022-09-29 CN CN202211196283.2A patent/CN115423212A/en active Pending
- 2022-12-02 CN CN202211534274.XA patent/CN115618986B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN115618986A (en) | 2023-01-17 |
CN115618986B (en) | 2023-04-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Xue et al. | Solving the data sparsity problem in destination prediction | |
Wang et al. | A quadtree approach to domain decomposition for spatial interpolation in grid computing environments | |
Wang et al. | Moving destination prediction using sparse dataset: A mobility gradient descent approach | |
Li et al. | Go slow to go fast: minimal on-road time route scheduling with parking facilities using historical trajectory | |
Baum et al. | Dynamic time-dependent route planning in road networks with user preferences | |
US20110251789A1 (en) | Method and system for time-dependent routing | |
Evans et al. | Enabling spatial big data via CyberGIS: Challenges and opportunities | |
Chen et al. | Gaussian process-based decentralized data fusion and active sensing for mobility-on-demand system | |
Ahmed et al. | Knowledge graph based trajectory outlier detection in sustainable smart cities | |
Murphy et al. | Risky planning on probabilistic costmaps for path planning in outdoor environments | |
Vander Aa et al. | Distributed Bayesian probabilistic matrix factorization | |
Guo et al. | Real-time ride-sharing framework with dynamic timeframe and anticipation-based migration | |
JP2023547451A (en) | Scalable modeling for large collections of time series | |
CN113656670A (en) | Flight data-oriented space-time trajectory data management analysis method and device | |
Al Jawarneh et al. | Spatial-aware approximate big data stream processing | |
CN115618986B (en) | Method and device for coordinating resources | |
Tesfaye et al. | Speeding up reachability queries in public transport networks using graph partitioning | |
CN112767032A (en) | Information processing method and device, electronic equipment and storage medium | |
Zhang et al. | Clustering with implicit constraints: A novel approach to housing market segmentation | |
Escalante et al. | Methodological issues in modern track analysis | |
Biswas et al. | Ripple: An approach to locate k nearest neighbours for location-based services | |
Mariescu-Istodor et al. | Fast travel-distance estimation using overhead graph | |
Ghosh et al. | An application of network lasso optimization for ride sharing prediction | |
Reijsbergen | Probabilistic modelling of station locations in bicycle-sharing systems | |
CN112801401A (en) | Method and device for determining time information of route |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20221202 |
|
WD01 | Invention patent application deemed withdrawn after publication |