Data processing method, device, equipment, computer storage medium and vehicle
By constructing origin-destination graph structures and multi-layer graph models, and using LSTM to train user travel pattern feature vectors, the problem that existing path prediction methods cannot reflect the overall travel patterns of users is solved, achieving higher path prediction accuracy and user experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING CO WHEELS TECH CO LTD
- Filing Date
- 2022-10-08
- Publication Date
- 2026-07-24
AI Technical Summary
Existing route prediction methods mainly rely on user trajectory data, which cannot accurately reflect the user's overall travel patterns, resulting in low route prediction accuracy.
By extracting historical origin and destination information from users' historical travel history, an origin-destination graph structure is constructed. A multi-layer graph model is then established using a Long Short-Term Memory (LSTM) network model and trained using an autoencoder to obtain user travel pattern feature vectors, reflecting the user's overall travel patterns.
It improves the accuracy of route prediction, better follows users' travel habits, and enhances the user travel experience.
Smart Images
Figure CN117852680B_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of intelligent transportation technology, and in particular relates to a data processing method, apparatus, equipment, computer storage medium, and vehicle. Background Technology
[0002] Intelligent Transportation Systems (ITS) represent the future direction of transportation systems. They are comprehensive transportation management systems that effectively integrate advanced information technology, data communication and transmission technology, electronic sensing technology, control technology, and computer technology into the entire ground transportation management system. These systems function in a wide range, in all aspects, and are real-time, accurate, and efficient.
[0003] Existing route prediction methods primarily learn user travel patterns from trajectory data and then predict future routes based on these patterns. However, trajectory data only reflects a specific part of a user's journey and does not represent their overall travel patterns; therefore, current technologies have relatively low accuracy in predicting routes. Summary of the Invention
[0004] This application provides a data processing method, apparatus, device, computer storage medium, and vehicle that can determine the user's travel pattern feature vector based on the user's historical travel information, thereby improving the accuracy of path prediction.
[0005] On one hand, embodiments of this application provide a data processing method, the method comprising:
[0006] Obtain the user's historical trip information and extract the historical origin and destination from the historical trip information;
[0007] Based on the historical itinerary information and the historical origin and destination, construct the origin and destination map structure;
[0008] Based on the origin-endpoint graph structure, an initial graph model with a multi-layered structure is established for each of the historical origins and ends; the nodes of the initial graph model are the historical origins and ends.
[0009] Based on the historical travel parameter information corresponding to the historical start and end points of each layer of the initial graph model, the user's historical travel parameter summary information is obtained;
[0010] Based on the summarized information of the historical travel parameters, the initial graph model is trained to obtain the target graph model;
[0011] Historical travel parameter information corresponding to the historical origin and destination of the intermediate layer of the target graph model is extracted to obtain the user travel pattern feature vector; the user travel pattern feature vector is used to predict the user's path information.
[0012] In one possible implementation, training the initial graph model based on the historical travel parameter summary information to obtain the target graph model includes:
[0013] The historical travel parameter summary information is used as the input data and label data of the initial graph model, and autoencoder training is performed to obtain the transition graph model;
[0014] Based on the historical journey parameter information corresponding to the historical start and end points of each layer of the transition graph model, the user's transition historical journey parameter summary information is obtained;
[0015] The summary information of the transition history parameters is used as the input data and label data of the transition graph model, and autoencoder training is performed to obtain the target graph model.
[0016] In one possible implementation, after obtaining the user travel pattern feature vector, the method further includes:
[0017] Obtain the user's target origin and destination information;
[0018] Based on the user's historical travel information, destination information, and user travel pattern feature vector, predict path information that matches the destination information.
[0019] In one possible implementation, extracting the historical origin and destination from the historical itinerary information includes:
[0020] Extract n trip origins and destinations from the historical trip information;
[0021] Clustering is performed on the n trip start and end points, and the center point of each cluster after clustering is determined as the historical start and end point, resulting in m historical start and end points; where n≥m.
[0022] In one possible implementation, constructing the origin-destination map structure based on the historical travel information and the historical origin-destination points includes:
[0023] Based on the historical origin and destination information and historical travel information, determine the relationship between any two historical origin and destination information;
[0024] The relationships are summarized to obtain the relationship matrix of the historical start and end points;
[0025] Based on the relationship matrix of the historical start and end points, construct the start and end point graph structure.
[0026] On the other hand, embodiments of this application provide a data processing apparatus, the apparatus comprising:
[0027] The acquisition module is used to acquire the user's historical trip information and extract the historical origin and destination from the historical trip information;
[0028] The graph structure module is used to construct a start-end point graph structure based on the historical itinerary information and the historical start-end point.
[0029] The graph model module is used to establish an initial graph model with a multi-layered structure for each of the historical start and end points based on the start and end point graph structure; the nodes of the initial graph model are the historical start and end points.
[0030] The summary module is used to obtain the user's historical travel parameter summary information based on the historical travel parameter information corresponding to the historical start and end points of each layer of the initial graph model;
[0031] The training module is used to train the initial graph model based on the information summarized from the historical travel parameters to obtain the target graph model;
[0032] The user travel pattern module is used to extract historical travel parameter information corresponding to the historical origin and destination of the intermediate layer of the target graph model to obtain the user travel pattern feature vector; the user travel pattern feature vector is used to predict the user's path information.
[0033] In another aspect, embodiments of this application provide a data processing device, the device comprising: a processor and a memory storing computer program instructions;
[0034] When the processor executes the computer program instructions, it implements the data processing method as described in one aspect.
[0035] In another aspect, embodiments of this application provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the data processing method as described in one aspect.
[0036] In another aspect, embodiments of this application provide a computer program product in which instructions, when executed by a processor of an electronic device, cause the electronic device to perform the data processing method as described in one aspect.
[0037] Furthermore, embodiments of this application provide a vehicle, which includes at least one of the following:
[0038] Such as the data processing device described above;
[0039] Such as the data processing equipment mentioned above;
[0040] Such as the computer-readable storage media described above;
[0041] Such as the computer program products mentioned above.
[0042] This application discloses a data processing method, apparatus, device, computer storage medium, and vehicle that extracts historical start and end points from a user's historical travel information. A graph structure is constructed based on the historical travel information and historical start and end points. The graph model is then trained using this graph structure to obtain a user travel pattern feature vector that indicates the user's overall travel patterns. This user travel pattern feature vector can be used for user route prediction. Thus, by adding this user travel pattern feature vector during route prediction, the user's travel patterns can be reflected macroscopically, thereby improving the accuracy of route prediction. Attached Figure Description
[0043] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0044] Figure 1 This is a flowchart illustrating the data processing method provided in the first embodiment of this application;
[0045] Figure 2 This is a schematic diagram of the relationship matrix provided in the first embodiment of this application;
[0046] Figure 3 This is a schematic diagram of the diagram structure provided in the first embodiment of this application;
[0047] Figure 4 This is a schematic diagram of the graphic model provided in the first embodiment of this application;
[0048] Figure 5 This is a schematic diagram of the data processing apparatus provided in the second embodiment of this application;
[0049] Figure 6 This is a schematic diagram of the structure of the data processing device provided in the third embodiment of this application. Detailed Implementation
[0050] To better understand the above-mentioned objectives, features, and advantages of this application, the solution of this application will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.
[0051] Many specific details are set forth in the following description in order to provide a full understanding of this application, but this application may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some embodiments of this application, and not all embodiments.
[0052] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the term "comprising" or any other variations thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0053] Existing route prediction models primarily take user trajectory data as input, learning from this data to deduce user travel patterns and then predicting future routes based on these patterns. However, trajectory data only represents a specific segment of a user's journey and cannot represent their overall travel patterns. In other words, existing route prediction models cannot learn from users' overall travel patterns, resulting in low accuracy in predicted route information.
[0054] Based on this, the inventors extracted the historical start and end points from the user's historical travel information, constructed a graph structure based on the historical travel information and historical start and end points, and then trained the graph model based on the graph structure to obtain a user travel pattern feature vector that can represent the user's overall travel pattern. This increases the input samples of the existing path prediction model and improves the accuracy of path prediction.
[0055] The first embodiment of this application provides a data processing method. Figure 1 This is a flowchart illustrating the data processing method provided in the first embodiment of this application, as shown below. Figure 1 As shown, the method includes:
[0056] Step S101: Obtain the user's historical trip information and extract the historical origin and destination from the historical trip information;
[0057] Specifically, n trip start and end points are extracted from the historical trip information, and clustering is performed on the n trip start and end points. The center point of each cluster after clustering is determined as the historical start and end point, resulting in m historical start and end points, where n≥m.
[0058] Because user stop behavior is random, extracting the origin and destination points from each historical trip record does not necessarily represent the accurate starting point. Clustering is a method that uses morphological operators to cluster and merge neighboring similar classification regions. By using clustering algorithms to cluster the origin and destination points extracted from historical trip information, we can obtain historical origin and destination points that accurately represent the user's origin and destination.
[0059] As can be seen, this application can obtain more accurate historical origins and destinations by clustering the origins and destinations extracted from historical travel information.
[0060] Existing user travel information is mostly recorded on a per-trip basis. However, on some older devices, it is still recorded on a per-travel-track basis. For this type of record based on travel tracks, the user's track information is obtained before obtaining the user's historical trip information, and the track information is segmented to obtain the historical trip information.
[0061] In one example, if a user travels from point A to point C and stops at point B for more than a preset threshold, then point B is considered as the origin and destination. Based on the arrival times of the origin and destination, the trajectory information is divided into A to B and B to C.
[0062] In another example, if a user starts from point A and travels to point C, and turns off the engine at point D, then point D is considered to be both the origin and destination. Based on the arrival times of the origin and destination, the trajectory information is divided into A to D and D to C.
[0063] In another example, if a user travels from point A to point C and stays in the area of point E for more than a preset time, then point E is considered to be the origin and destination. Based on the arrival time of the origin and destination, the trajectory information is divided into A to E and E to C.
[0064] As can be seen, this application can obtain historical trip information in units of individual trips by segmenting trajectory information, which facilitates subsequent analysis of users' overall travel patterns based on historical trip information.
[0065] Step S102: Construct the origin-destination map structure based on historical itinerary information and historical origin-destination points;
[0066] Specifically, based on historical origin-endpoint and historical itinerary information, the relationship between any two historical origin-endpoints is determined. These relationships are then aggregated to obtain a relationship matrix of historical origin-endpoints. For example, if the historical itinerary information is from historical origin-endpoint A to historical origin-endpoint B, then there is a relationship between historical origin-endpoint A and historical origin-endpoint B, and this relationship is recorded as 1. If there is no historical itinerary information from historical origin-endpoint A to historical origin-endpoint E, then there is no relationship between historical origin-endpoint A and historical origin-endpoint E, and this relationship is recorded as 0. This process is repeated for all historical itinerary information, using 0 or 1 to mark the relationship between the two historical origin-endpoints corresponding to each historical itinerary information.
[0067] Figure 2 This is a schematic diagram of the relation matrix provided in the first embodiment of this application, as shown below. Figure 2As shown, the final result is a relation matrix regarding historical start and end points. A, B, C, D, and E represent the user's historical start and end points. The 0s and 1s at the intersection of rows and columns represent the absence of a relationship between two historical start and end points, and the existence of a relationship between two historical start and end points, respectively. Columns represent historical start and end points as starting points, and rows represent historical start and end points as ending points. Figure 2 It can be concluded that there are historical travel information from historical origin and destination A to historical origin and destination B and historical origin and destination F; there are historical travel information from historical origin and destination B to historical origin and destination A and historical origin and destination C; there are historical travel information from historical origin and destination C to historical origin and destination D and historical origin and destination F; there are historical travel information from historical origin and destination D to historical origin and destination C; there are historical travel information from historical origin and destination E to historical origin and destination D and historical origin and destination F; and there are historical travel information from historical origin and destination F to historical origin and destination A, historical origin and destination C, and historical origin and destination E.
[0068] Based on the relationship matrix of historical start and end points, we can determine whether there is a relationship between any two historical start and end points, and construct a matrix as follows: Figure 3 The starting and ending point diagram structure is as follows: historical starting and ending point A and historical starting and ending point B are each other's starting and ending points; historical starting and ending point A and historical starting and ending point F are each other's starting and ending points; historical starting and ending point C and historical starting and ending point F are each other's starting and ending points; historical starting and ending point E and historical starting and ending point F are each other's starting and ending points; historical starting and ending point C and historical starting and ending point D are each other's starting and ending points; historical starting and ending point E is the starting point of historical starting and ending point D; and historical starting and ending point B is the starting point of historical starting and ending point C.
[0069] Step S103: Based on the origin and destination graph structure, establish an initial graph model for each historical origin and destination, including a multi-layered structure.
[0070] Specifically, the nodes of the initial graph model are the historical start and end points. The initial graph model uses an LSTM (Long Short-Term Memory) model. There is no limit to the number of layers in the model, but the number of layers in the initial graph model corresponding to each historical start and end point must be the same. LSTM is a temporal recurrent neural network, suitable for processing and predicting important events with relatively long intervals and delays in time series.
[0071] Figure 4 This is a schematic diagram of the graphical model provided in the first embodiment of this application, such as... Figure 4 As shown, this is the initial graph model corresponding to the historical start and end point A, with 3 layers.
[0072] Step S104: Based on the historical travel parameter information corresponding to the historical start and end points of each layer of the initial graph model, obtain the user's historical travel parameter summary information;
[0073] Specifically, the historical travel parameters corresponding to the historical origins and destinations of each layer of the initial graph model are averaged. The historical travel information consists of the attribute information of the historical origin and destination information, including: arrival time, departure time, location, number of arrivals, etc.
[0074] Step S105: Based on the information summarized from historical travel parameters, train the initial graph model to obtain the target graph model;
[0075] Specifically, the historical travel parameter summary information is used as input data and label data for the initial graph model, and autoencoder training is performed to obtain the transition graph model. Based on the historical travel parameter information corresponding to the historical start and end points of each layer of the transition graph model, the user's transition historical travel parameter summary information is obtained. This transition historical travel parameter summary information is then used as input data and label data for the transition graph model, and autoencoder training is performed to obtain the target graph model.
[0076] An autoencoder is a neural network with three or more layers, consisting of an encoder, hidden layers, and a decoder. An autoencoder encodes an input representation X into a new representation Y, and then decodes Y back into X. It is an unsupervised learning algorithm that uses backpropagation to train the network so that the output equals the input.
[0077] The initial graph model is an untrained graph model, the transition graph model is a graph model trained once, and the target graph model is a graph model trained twice. The historical journey parameter summary information represents the characteristics of the attributes of the historical start and end points.
[0078] As can be seen, this application constructs a graph model and graph structure for each historical origin and destination, and trains the graph model based on the graph structure to obtain the feature vector of the user's overall travel pattern.
[0079] Step S106: Extract the historical travel parameter information corresponding to the historical origin and destination of the intermediate layer of the target graph model to obtain the user travel pattern feature vector.
[0080] Specifically, user travel pattern feature vectors are used to predict user route information. Each historical origin and destination has a corresponding user travel feature vector, thus obtaining a user travel feature vector representing the overall travel pattern of the user starting from each historical origin and destination.
[0081] In one example, after obtaining the user's travel pattern feature vector, the user's destination information is acquired. Based on the user's historical travel information, destination information, and travel pattern feature vector, path information matching the destination information is predicted.
[0082] The methods for obtaining a user's origin and destination information include one or a combination of the following:
[0083] 1) Obtain the text entered by the user on the user terminal;
[0084] 2) Acquire the voice input by the user on the user terminal;
[0085] 3) Obtain the selection operation entered by the user on the user terminal.
[0086] In one specific embodiment, after obtaining the user's travel pattern feature vector, the user's historical travel information is searched based on the target origin and destination and the user's travel pattern feature vector. If historical travel information matching the target origin and destination and the user's travel pattern feature vector is found, the historical path corresponding to that historical travel information is used as the predicted path information. If no historical travel information matching the target origin and destination and the user's travel pattern feature vector is found, the path information planned based on the target origin and destination is used.
[0087] This approach maximizes adherence to historical travel habits corresponding to the origin and destination, reducing user discomfort caused by new routes and improving the user travel experience.
[0088] In another specific embodiment, after obtaining the user's travel pattern feature vector, the user's route selection habits are analyzed based on their historical travel information. For example, the user avoids highways and prefers routes with fewer traffic lights. Based on the user's route selection habits, destination, and travel pattern feature vector, the resulting route information is planned.
[0089] This allows us to predict the route information that best matches users' travel habits, thereby improving their travel experience.
[0090] Specifically, the origin and destination information refers to the origin and destination of a user-inputted trip, along with the corresponding trip parameters. The user's historical trip information and travel pattern feature vectors are used as samples for the existing path prediction model, serving as training samples. By using the origin and destination information as input to the path prediction model after the training, more accurate path information can be obtained by considering the user's overall travel patterns.
[0091] As can be seen, this application adds user travel pattern feature vectors that can reflect the overall travel patterns of users to the existing path prediction model samples, thereby improving the accuracy of the path prediction model in predicting paths.
[0092] As can be seen, the data processing method of this application extracts the historical start and end points from the user's historical travel information, constructs a graph structure based on the historical travel information and historical start and end points, and then trains the graph model based on the graph structure to obtain a user travel pattern feature vector that can indicate the user's overall travel pattern. This user travel pattern feature vector that can indicate the user's overall travel pattern can be used for user path prediction. Thus, when predicting paths, adding this user travel pattern feature vector can reflect the user's travel pattern from a macroscopic perspective, thereby improving the accuracy of path prediction.
[0093] A second embodiment of this application also provides a data processing apparatus, comprising:
[0094] Module 1 is used to obtain the user's historical trip information and extract the historical origin and destination from the historical trip information;
[0095] Graph structure module 2 is used to construct the origin-endpoint graph structure based on historical itinerary information and historical origin-endpoints;
[0096] Graph model module 3 is used to build an initial graph model with a multi-layered structure for each historical start and end point based on the start and end point graph structure; the nodes of the initial graph model are the historical start and end points.
[0097] Summary module 4 is used to obtain the user's historical travel parameter summary information based on the historical travel parameter information corresponding to the historical start and end points of each layer of the initial graph model;
[0098] Training module 5 is used to train the initial graph model based on the information summarized from historical travel parameters, so as to obtain the target graph model;
[0099] The user travel pattern module 6 is used to extract historical travel parameter information corresponding to the historical origin and destination of the intermediate layer of the target graph model to obtain the user travel pattern feature vector; the user travel pattern feature vector is used to predict the user's path information.
[0100] In one example, the graph model module 3 is used to: use the historical travel parameter summary information as input data and label data of the initial graph model for autoencode training to obtain the transition graph model; obtain the user's transition historical travel parameter summary information based on the historical travel parameter information corresponding to the historical start and end points of each layer of the transition graph model; and use the transition historical travel parameter summary information as input data and label data of the transition graph model for autoencode training to obtain the target graph model.
[0101] In one example, the data processing device further includes a path prediction module 7, used to: obtain the user's target origin and destination information; and predict path information that matches the target origin and destination information based on the user's historical travel information, target origin and destination information, and user travel pattern feature vector.
[0102] In one example, module 1 is used to: extract n trip start and end points from historical trip information; perform clustering processing on the n trip start and end points, determine the center point of each cluster after clustering as the historical start and end point, and obtain m historical start and end points; n≥m.
[0103] In one example, the graph structure module 2 is used to: determine the relationship between any two historical origins and destinations based on historical origin-endpoint and historical itinerary information; perform summary processing based on the relationship to obtain the relationship matrix of historical origins and destinations; and construct the origin-endpoint graph structure based on the relationship matrix of historical origins and destinations.
[0104] As can be seen, the data processing device executes the data processing method in this application embodiment based on the user's historical travel information to obtain user travel pattern characteristics that can indicate the user's overall travel patterns.
[0105] A third embodiment of this application also provides a data processing device. Figure 6 This is a schematic diagram of the structure of the data processing device provided in the third embodiment of this application, as shown below. Figure 6 As shown, the device includes a processor 301 and a memory 302 storing computer program instructions.
[0106] Specifically, the processor 301 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.
[0107] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 302 may include removable or non-removable (or fixed) media. Where appropriate, memory 302 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 302 is non-volatile solid-state memory.
[0108] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the methods according to one aspect of this disclosure.
[0109] The processor 301 implements the data processing method of the first embodiment described above by reading and executing computer program instructions stored in the memory 302.
[0110] In one example, the data processing device may further include a communication interface 303 and a bus 310. Wherein, as... Figure 6 As shown, the processor 301, memory 302, and communication interface 303 are connected through bus 310 and complete communication with each other.
[0111] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.
[0112] Bus 310 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 310 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.
[0113] It is understood that the data processing device executes the data processing method in this application embodiment based on the user's historical travel information, thereby achieving... Figure 1 and Figure 5 The data processing methods and apparatus described.
[0114] The fourth embodiment of this application also provides a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the data processing method as described in the first embodiment.
[0115] The fifth embodiment of this application also provides a computer program product, wherein the instructions in the computer program product, when executed by the processor of an electronic device, cause the electronic device to perform the data processing method as described in the first embodiment.
[0116] A sixth embodiment of this application also provides a vehicle, which includes at least one of the following: a data processing apparatus as described in the second embodiment; a data processing device as described in the third embodiment; a computer-readable storage medium as described in the fourth embodiment; and a computer program product as described in the fifth embodiment.
[0117] It should be understood that the specific embodiments described above are merely illustrative or explanatory of the principles of this application and do not constitute a limitation thereof. Therefore, any modifications, equivalent substitutions, improvements, etc., made without departing from the spirit and scope of this application should be included within the protection scope of this application. Furthermore, the appended claims are intended to cover all variations and modifications falling within the scope and boundaries of the appended claims, or equivalent forms of such scope and boundaries.
[0118] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.
[0119] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.
[0120] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.
Claims
1. A data processing method, characterized in that, The method includes: Obtain the user's historical trip information and extract the historical origin and destination from the historical trip information; Based on the historical itinerary information and the historical origin and destination, construct the origin and destination map structure; Based on the origin-endpoint graph structure, an initial graph model with a multi-layered structure is established for each of the historical origins and ends; the nodes of the initial graph model are the historical origins and ends. Based on the historical travel parameter information corresponding to the historical start and end points of each layer of the initial graph model, the user's historical travel parameter summary information is obtained; Based on the summarized information of the historical travel parameters, the initial graph model is trained to obtain the target graph model; Historical travel parameter information corresponding to the historical origin and destination of the intermediate layer of the target graph model is extracted to obtain the user travel pattern feature vector; the user travel pattern feature vector is used to predict the user's path information.
2. The data processing method according to claim 1, characterized in that, The step of training the initial graph model based on the historical travel parameter summary information to obtain the target graph model includes: The historical travel parameter summary information is used as the input data and label data of the initial graph model, and autoencoder training is performed to obtain the transition graph model; Based on the historical journey parameter information corresponding to the historical start and end points of each layer of the transition graph model, the user's transition historical journey parameter summary information is obtained; The summary information of the transition history parameters is used as the input data and label data of the transition graph model, and autoencoder training is performed to obtain the target graph model.
3. The data processing method according to claim 1, characterized in that, After obtaining the user travel pattern feature vector, the method further includes: Obtain the user's target origin and destination information; Based on the user's historical travel information, destination information, and user travel pattern feature vector, predict path information that matches the destination information.
4. The data processing method according to claim 1, characterized in that, The extraction of historical origin and destination points from the historical itinerary information includes: Extract n trip origins and destinations from the historical trip information; Clustering is performed on the n trip start and end points, and the center point of each cluster after clustering is determined as the historical start and end point, resulting in m historical start and end points; where n≥m.
5. The data processing method according to claim 1, characterized in that, The step of constructing a start-endpoint map structure based on the historical itinerary information and the historical start-endpoints includes: Based on the historical origin and destination information and historical travel information, determine the relationship between any two historical origin and destination information; The relationships are summarized to obtain the relationship matrix of the historical start and end points; Based on the relationship matrix of the historical start and end points, construct the start and end point graph structure.
6. A data processing apparatus, characterized in that, The device includes: The acquisition module is used to acquire the user's historical trip information and extract the historical origin and destination from the historical trip information; The graph structure module is used to construct a start-end point graph structure based on the historical itinerary information and the historical start-end point. The graph model module is used to establish an initial graph model with a multi-layered structure for each of the historical start and end points based on the start and end point graph structure; the nodes of the initial graph model are the historical start and end points. The summary module is used to obtain the user's historical travel parameter summary information based on the historical travel parameter information corresponding to the historical start and end points of each layer of the initial graph model; The training module is used to train the initial graph model based on the information summarized from the historical travel parameters to obtain the target graph model; The user travel pattern module is used to extract historical travel parameter information corresponding to the historical origin and destination of the intermediate layer of the target graph model to obtain the user travel pattern feature vector; the user travel pattern feature vector is used to predict the user's path information.
7. A data processing device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the data processing method as described in any one of claims 1-5.
8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the data processing method as described in any one of claims 1-5.
9. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device performs the data processing method as described in any one of claims 1-5.
10. A vehicle, characterized in that, The vehicle includes at least one of the following: The data processing apparatus as described in claim 6; The data processing device as described in claim 7; The computer-readable storage medium as described in claim 8; The computer program product as described in claim 9.