A user travel data processing method, device and electronic equipment

Abstract

The user travel data processing scheme provided in the application, when processing user travel data, inputs the original travel data of the user into a data sub-model to obtain travel detail data in a preset format; constructs a travel knowledge graph corresponding to the user, the travel knowledge graph including event nodes corresponding to the original travel data, data nodes corresponding to the travel detail data, space-time label nodes corresponding to space-time data of travel, characteristic label nodes corresponding to travel characteristics, and relationship edges between the nodes; calls an index sub-model to calculate travel index values of the user according to the event nodes, data nodes, space-time nodes and relationship edges in the travel knowledge graph of the user; and determines the travel characteristics of the label characteristic nodes of the user according to the travel index values. Through the scheme of the present disclosure, the user travel knowledge graph is continuously constructed, and the correlation analysis and algorithm mining are performed based on the graph, so that the correlation efficiency and analysis efficiency are higher, and the potential travel characteristics of the passenger can be more effectively and accurately mined.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a method, apparatus, and electronic device for processing user travel data. Background Technology

[0002] With the accelerating pace of urbanization, my country's urban rail transit industry has experienced rapid development in recent years. The massive passenger flow has generated a wealth of related data, such as passenger flow and ticketing information. Without systems like passenger profiling, this data remains fragmented and undirected, resulting in low data value and a significant waste of data resources. To analyze the travel characteristics of each passenger and specific passenger groups from this data, and thus provide more precise and personalized services, subway companies have begun building passenger profiling systems.

[0003] Analyzing passenger travel characteristics is the most important, complex, and difficult step in building a passenger profile system. Currently, there are some clustering algorithms for analyzing passenger travel characteristics, but these methods often face challenges such as high computational complexity and difficulty in implementation.

[0004] Some implementations use the k-means algorithm from the unsupervised learning algorithm K-MEANS to cluster passengers and analyze their behavioral characteristics. The typical approach is as follows:

[0005] 1> Data preprocessing: The raw data was merged and organized to obtain passenger travel records. Each passenger travel record includes: passenger's entry station, entry card swipe time, exit station, and exit card swipe time.

[0006] 2> Feature Extraction: Based on passenger travel records, cluster attributes such as pedestrian stations and time to obtain passenger travel features;

[0007] 3> Passenger Clustering: Based on passenger travel characteristics, passenger clustering is performed to obtain the travel characteristics of passenger groups.

[0008] Existing solutions for analyzing passenger travel characteristics use clustering algorithms, which have high computational complexity; the parameter adjustment of the algorithm model is complex, and there are technical problems that are difficult to implement. Summary of the Invention

[0009] In view of this, the present disclosure provides a user travel data processing method, apparatus, and electronic device, which at least partially solves the problems existing in the prior art.

[0010] In a first aspect, embodiments of this disclosure provide a user travel data processing method, including:

[0011] The travel data processing model is invoked, wherein the travel data processing model includes a data sub-model, an indicator sub-model, and a graph sub-model;

[0012] Input the user's original travel data into the data sub-model to obtain travel detail data in a preset format;

[0013] Construct a travel knowledge graph corresponding to the user, wherein the travel knowledge graph includes event nodes corresponding to the original travel data, data nodes corresponding to the travel details data, spatiotemporal label nodes corresponding to the travel spatiotemporal data, feature label nodes corresponding to the travel features, and the relationship edges between the nodes;

[0014] The indicator sub-model is invoked to calculate the user's travel indicator value based on the event nodes, data nodes, spatiotemporal nodes, and relation edges in the user's travel knowledge graph;

[0015] The user's travel characteristics are determined based on the travel index value, which represents the user's tag feature node.

[0016] According to a specific implementation of this disclosure, before the step of invoking the travel data processing model, the method further includes:

[0017] Construct data sub-models, indicator sub-models, and graph sub-models;

[0018] Define the field recognition rules and field format processing rules for the data sub-model;

[0019] Define the indicator types, hierarchical relationships, and calculation rules for the indicator sub-models;

[0020] Define the entity type corresponding to each node of the graph sub-model and the edge type between the nodes.

[0021] According to a specific implementation of this disclosure, the indicator types include basic indicators and composite indicators. The indicator calculation rules include a first type of calculation rule from basic indicators to composite indicators and a second type of calculation rule from composite indicators to travel features. The step of calling the indicator sub-model to calculate the user's travel indicator value based on the event nodes, data nodes, spatiotemporal nodes, and relation edges in the user's travel knowledge graph includes:

[0022] The indicator sub-model is invoked to determine the basic indicator values ​​of each basic indicator based on the event nodes, data nodes, spatiotemporal nodes, and relation edges in the user's travel knowledge graph;

[0023] Based on the values ​​of each basic indicator, the hierarchical relationship of indicators, and the first type of calculation rules for each composite indicator, calculate the composite indicator value of each composite indicator;

[0024] The user's travel index value is calculated based on the composite index values ​​of each composite index and the second type of calculation rules.

[0025] According to a specific implementation of this disclosure, after the step of calculating the user's travel indicator value based on the composite indicator values ​​of each composite indicator and the second type of calculation rules, the method further includes:

[0026] The basic indicator values, composite indicator values, and travel indicator values ​​are stored according to their time-series characteristics.

[0027] According to a specific implementation of this disclosure, the travel details data includes at least the user's user identifier, origin and destination station identifiers, and travel time; and / or,

[0028] Data nodes include username data nodes corresponding to user identifiers, station data nodes corresponding to start or end stations, route data nodes corresponding to travel routes, or ticket data nodes corresponding to travel tickets; and / or,

[0029] Spatiotemporal tag nodes include time tag nodes and spatial tag nodes; and / or,

[0030] Travel characteristics include travel stability characteristics.

[0031] Secondly, embodiments of this disclosure provide a user travel data processing device, comprising:

[0032] The calling module is used to call the travel data processing model, wherein the travel data processing model includes a data sub-model, an indicator sub-model, and a graph sub-model;

[0033] The processing module is used to input the user's original travel data into the data sub-model to obtain travel detail data in a preset format;

[0034] The construction module is used to construct a travel knowledge graph corresponding to the user, wherein the travel knowledge graph includes event nodes corresponding to the original travel data, data nodes corresponding to the travel details data, spatiotemporal label nodes corresponding to the travel spatiotemporal data, feature label nodes corresponding to the travel features, and the relationship edges between the nodes;

[0035] The calculation module is used to call the indicator sub-model to calculate the user's travel indicator value based on the event nodes, data nodes, spatiotemporal nodes and relation edges in the user's travel knowledge graph;

[0036] The determination module is used to determine the travel characteristics of the user's tag feature nodes based on the travel index value.

[0037] According to a specific implementation of an embodiment of this disclosure, the construction module is further configured to:

[0038] Construct data sub-models, indicator sub-models, and graph sub-models;

[0039] Define the field recognition rules and field format processing rules for the data sub-model;

[0040] Define the indicator types, hierarchical relationships, and calculation rules for the indicator sub-models;

[0041] Define the entity type corresponding to each node of the graph sub-model and the edge type between the nodes.

[0042] According to a specific implementation of this disclosure, the indicator types include basic indicators and composite indicators, and the indicator calculation rules include a first type of calculation rule from basic indicators to composite indicators and a second type of calculation rule from composite indicators to travel characteristics. The calculation module is used for:

[0043] The indicator sub-model is invoked to determine the basic indicator values ​​of each basic indicator based on the event nodes, data nodes, spatiotemporal nodes, and relation edges in the user's travel knowledge graph;

[0044] Based on the values ​​of each basic indicator, the hierarchical relationship of indicators, and the first type of calculation rules for each composite indicator, calculate the composite indicator value of each composite indicator;

[0045] The user's travel index value is calculated based on the composite index values ​​of each composite index and the second type of calculation rules.

[0046] According to a specific implementation of an embodiment of this disclosure, the apparatus further includes a storage module, the storage module being used for:

[0047] The basic indicator values, composite indicator values, and travel indicator values ​​are stored according to their time-series characteristics.

[0048] Thirdly, embodiments of this disclosure also provide an electronic device, the electronic device comprising:

[0049] The electronic device includes:

[0050] At least one processor; and,

[0051] A memory that is communicatively connected to at least one processor; wherein,

[0052] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the user travel data processing method in the first aspect or any implementation thereof.

[0053] Fourthly, embodiments of this disclosure also provide a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the user travel data processing method in the first aspect or any implementation thereof.

[0054] Fifthly, embodiments of this disclosure also provide a computer program product, the computer program product including a computing program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, which, when executed by a computer, cause the computer to perform the user travel data processing method in the first aspect or any implementation thereof.

[0055] The user travel data processing scheme in this embodiment involves inputting the user's original travel data into the data sub-model to obtain travel detail data in a preset format. A travel knowledge graph corresponding to the user is constructed, wherein the travel knowledge graph includes event nodes corresponding to the original travel data, data nodes corresponding to the travel detail data, spatiotemporal label nodes corresponding to the travel spatiotemporal data, feature label nodes corresponding to travel characteristics, and relationship edges between nodes. The index sub-model is then invoked to calculate the user's travel index value based on the event nodes, data nodes, spatiotemporal nodes, and relationship edges in the user's travel knowledge graph. The travel characteristics of the user's label feature nodes are determined based on the travel index value. This scheme continuously constructs a user travel knowledge graph, and then performs association analysis and algorithm mining based on the graph, resulting in higher association and analysis efficiency, and enabling more effective and accurate mining of passengers' potential travel characteristics. Attached Figure Description

[0056] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 A flowchart illustrating a user travel data processing method provided in an embodiment of this disclosure;

[0058] Figure 2 This is a schematic diagram illustrating the implementation process of the user travel data processing method provided in the embodiments of this disclosure;

[0059] Figures 3 to 10 This is an example schematic diagram illustrating the user travel data processing method provided in the embodiments of this disclosure;

[0060] Figure 11 This is a block diagram of a user travel data processing device provided in an embodiment of the present disclosure. Detailed Implementation

[0061] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0062] The following specific examples illustrate the implementation of this disclosure. Those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. This disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0063] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this disclosure, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using structures and / or functionalities other than one or more of the aspects set forth herein.

[0064] It should also be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this disclosure. The drawings only show the components related to this disclosure and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0065] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0066] See Figure 1 This is a flowchart illustrating a user travel data processing method provided in an embodiment of this disclosure. Figure 1 As shown, the provided user travel data processing method mainly includes the following steps:

[0067] Step S101: Invoke the travel data processing model, wherein the travel data processing model includes a data sub-model, an indicator sub-model, and a graph sub-model;

[0068] The user travel data processing method provided in this embodiment is applied to an electronic device with data processing capabilities. A travel data processing model is loaded into the electronic device, and this model is invoked when data processing is required. Specifically, the data processing model includes a data sub-model, an indicator sub-model, and a graph sub-model. The data sub-model performs functions such as data field identification, data preprocessing, and format processing for OD (Origin to Destination) data. The indicator sub-model performs functions such as indicator classification, indicator assignment, and indicator value calculation. The graph sub-model performs functions such as graph construction and node assignment. In specific implementations, these multiple sub-models can be integrated into a single data processing model and loaded into the electronic device. Alternatively, the sub-models for each function can be loaded separately into one or more electronic devices. Furthermore, the sub-models for each function can be further divided into lightweight models with refined functions; no limitation is imposed.

[0069] In one specific implementation, such as Figure 2 As shown, considering that the main functions of the indicator sub-model include the maintenance of the indicator system and the storage of indicator data, the indicator sub-model is further subdivided into the indicator system model and the indicator data model.

[0070] When processing user travel data for the first time, it is necessary to first construct a relevant model. According to a specific implementation of this disclosure, before the step of calling the travel data processing model, the method further includes:

[0071] Construct data sub-models, indicator sub-models, and graph sub-models;

[0072] Define the field recognition rules and field format processing rules for the data sub-model;

[0073] Define the indicator types, hierarchical relationships, and calculation rules for the indicator sub-models;

[0074] Define the entity type corresponding to each node of the graph sub-model and the edge type between the nodes.

[0075] Step S102: Input the user's original travel data into the data sub-model to obtain travel detail data in a preset format;

[0076] The data sub-model processes users' raw travel data, such as passenger travel records from the origin station to the destination station, to obtain travel detail data. The travel detail data is in a preset format and includes essential fields. Optionally, the travel detail data may include at least the user's identifier, origin and destination station identifiers, and travel time.

[0077] like Figure 3 The image shows an example of a data sub-model, which may include, but is not limited to, fields such as ticket serial number, entry station number, exit station number, entry time, and exit time.

[0078] Step S103: Construct a travel knowledge graph corresponding to the user, wherein the travel knowledge graph includes event nodes corresponding to the original travel data, data nodes corresponding to the travel details data, spatiotemporal label nodes corresponding to the travel spatiotemporal data, feature label nodes corresponding to the travel features, and the relationship edges between the nodes;

[0079] After obtaining detailed travel data, the graph model constructs a corresponding travel knowledge graph based on this data. In the constructed travel knowledge graph, nodes can be event nodes, data nodes, spatiotemporal label nodes, and feature label nodes, while edges represent relationships between entity nodes. Of course, in specific designs, nodes representing the attributes of entity nodes can be added, such as... Figure 4 As shown, rectangular nodes represent entity nodes, elliptical nodes represent attribute nodes, and the connecting lines represent the edges between entity nodes.

[0080] like Figure 4 As shown, event nodes primarily store OD (Original Destination) detail records. Data nodes include username data nodes corresponding to user identifiers, station data nodes corresponding to start or end stations, route data nodes corresponding to travel routes, or ticket data nodes corresponding to travel tickets. The following will provide examples of the stored entity data:

[0081] User identification mainly includes attributes such as ID card number, name, and gender; the relationship between the passenger and the physical ticket is that the passenger owns the ticket.

[0082] Ticket: The ticket entity mainly includes attributes such as ticket serial number and ticket type; the relationship between the ticket and OD detail record entity is detail, which represents the travel detail record corresponding to the ticket;

[0083] Site: The site record contains detailed field information for each site, including site number, site name, etc.

[0084] Route: The route record contains alley field information for each route, including route number, route name, and other information.

[0085] Spatiotemporal tag nodes include time tag nodes and spatial tag nodes. For example... Figure 5 As shown, spatial tag entities are entities divided according to spatial characteristics. They are mainly used to store entities such as bidirectional OD combinations, OD combinations, O stations, and D stations, which are used to describe the spatial characteristics of user travel.

[0086] like Figure 6 As shown, time tag entities are entities divided according to time characteristics. They are mainly used to store entities such as time intervals and time peaks to describe the time characteristics of user travel.

[0087] like Figure 7 As shown, the feature tag node is used to store business tag entities, such as travel stability features. Of course, more business tag entities can be continuously created according to business needs, without limitation.

[0088] Step S104: Call the indicator sub-model to calculate the user's travel indicator value based on the event nodes, data nodes, spatiotemporal nodes and relation edges in the user's travel knowledge graph;

[0089] After acquiring the user's travel details data, the electronic device initially constructs relevant entity nodes in the knowledge graph. Then, based on attributes such as time and station, it establishes associations with spatial tag entities and time tag entities to calculate the travel features corresponding to the feature tag nodes. Specifically, based on event nodes, data nodes, spatiotemporal nodes, and relationships, it calculates the user's travel index value.

[0090] Step S105: Determine the travel characteristics of the user's tag feature node based on the travel index value.

[0091] After calculating the indicator values ​​according to the above steps, the user's travel characteristics can be found or calculated based on the pre-stored mapping relationship between indicator values ​​and travel characteristics. This completes the entire process from the user's basic travel records to travel characteristics in the travel knowledge graph.

[0092] According to a specific implementation of this disclosure, the indicator types include basic indicators and composite indicators. The indicator calculation rules include a first type of calculation rule from basic indicators to composite indicators and a second type of calculation rule from composite indicators to travel features. The step of calling the indicator sub-model to calculate the user's travel indicator value based on the event nodes, data nodes, spatiotemporal nodes, and relation edges in the user's travel knowledge graph includes:

[0093] The indicator sub-model is invoked to determine the basic indicator values ​​of each basic indicator based on the event nodes, data nodes, spatiotemporal nodes, and relation edges in the user's travel knowledge graph;

[0094] Based on the values ​​of each basic indicator, the hierarchical relationship of indicators, and the first type of calculation rules for each composite indicator, calculate the composite indicator value of each composite indicator;

[0095] The user's travel index value is calculated based on the composite index values ​​of each composite index and the second type of calculation rules.

[0096] This embodiment specifically explains the indicator definition and calculation scheme in the process of constructing a travel knowledge graph, mainly distinguishing between the functions of the two main execution entities: the indicator system model and the indicator data model. The indicator system model describes a series of indicators related to passenger travel characteristics, as well as the rules and time range for indicator calculation. As shown in Table 1 below, the indicator system model includes fields for indicator category, indicator name, indicator identifier, indicator level, calculation engine, calculation rules, and time range.

[0097] Table 1

[0098]

[0099] The metric category is used to classify metrics at the business level and can be customized. The metric name corresponds to the business meaning of the metric. The metric identifier is globally unique, uniquely identifying each metric. The metric level is used for hierarchical management of metrics; different levels of metrics often have different calculation methods and orders, with lower-level metrics being calculated less frequently. The calculation engine describes the method of metric calculation, initially divided into two categories: database SQL and expression engine. New calculation engines can be added as needed. Database SQL means that the value of the relevant metric can be directly queried using SQL statements; expression calculation means that the metric value needs to be obtained based on calculations of lower-level metrics. The calculation rule describes the specific rule expression used in the metric calculation. The calculation rules corresponding to different calculation engines are usually different. The calculation rules for the database SQL calculation engine are SQL statements; while the calculation rules for the expression calculation engine are calculation expressions, which can also be built-in functions or user-defined functions. The time range describes the data range limited by the metric calculation. Of course, Table 1 above only shows some examples of OD metrics; in actual applications, there will be many more metrics, requiring continuous addition as needed.

[0100] In practice, as shown in Table 2, the indicator data model can adopt a wide table design model, with the index column including the ticket serial number and calculation time; the indicator column contains all indicators in the indicator system.

[0101] Table 2

[0102]

[0103] The rule engine is used to calculate metric data, and the basis for rule calculation is the metric calculation rule. For example... Figure 8 As shown, the indicator system can be designed hierarchically, consisting of basic indicators and composite indicators. Composite indicators can include one level or multiple levels. Figure 8 The diagram shows three levels of indicators: 0, 2, and so on. The calculation of higher-level indicators depends on the calculation of lower-level indicators, and the rule engine's calculation logic is automatically divided into stages. First, the basic indicators at level 0 are calculated; then, the composite indicators at level 1 are calculated; and finally, the composite indicators at level 2 are calculated. If there are more indicator levels, the rule engine will automatically divide the calculation into more stages.

[0104] The rules engine is data-driven and can simultaneously meet both real-time and offline calculation scenarios for metrics, adapting to the calculation needs of different environments and improving its adaptability.

[0105] The following example will explain the calculation process from detailed travel data to indicator data and then to travel characteristics. Figure 9 As shown, the basic indicator "Total Number of Trips" is calculated. Assuming the current date is July 1, 2022, the total number of trips for ticket card (pk01) in the past 30 days is calculated, specifically the total number of trips from June 1, 2022 to June 30, 2022. The calculation method is a SQL query to the database. For graph databases, a graph query language can be used. The query returns a total of 3 trips in the past 30 days, with the corresponding query statement: "match(ticket card)-[:trip details]->(OD details record)where OD details record.entry time>='June 1, 2022'and OD details record.entry time<='June 30, 2022'with ticket card.serial number,count(OD details record)as total number of trips return ticket card.serial number,total number of trips". Similarly, other basic indicators can also be obtained using corresponding query statements.

[0106] After the metrics are calculated, the relationship between business tag entities and passengers is constructed, specifically as follows: Figure 10 As shown in the figure, the relevant data comes from Table 2. Based on the indicator values ​​of the tickets in Table 2, the graph is further constructed. For example, when the "bidirectional spatial instability" indicator value is 1, an edge from the user to the bidirectional spatial instability label is added to the graph. Similarly, when the "bidirectional spatial stability" indicator value is 1, an edge from the user to the bidirectional spatial stability label is added to the graph. Likewise, when the "bidirectional spatial gradually stabilizing" indicator value is 1, an edge from the user to the bidirectional spatial gradually stabilizing label is added to the graph. Ultimately, the construction of the travel knowledge graph is continuously improved to facilitate the analysis of user travel characteristics.

[0107] According to a specific implementation of this disclosure, after the step of calculating the user's travel indicator value based on the composite indicator values ​​of each composite indicator and the second type of calculation rules, the method further includes:

[0108] The basic indicator values, composite indicator values, and travel indicator values ​​are stored according to their time-series characteristics.

[0109] The indicator data has obvious time-series data characteristics. Storing it in a time-series database will provide a significant performance advantage for time-series analysis and query scenarios.

[0110] The user travel data processing scheme in this embodiment of the disclosure inputs the user's original travel data into the data sub-model to obtain travel detail data in a preset format; and constructs a travel knowledge graph corresponding to the user. Through this scheme, the user travel knowledge graph is continuously built, and then association analysis and algorithm mining are performed based on the graph, resulting in higher association and analysis efficiency, and enabling more effective and accurate mining of passengers' potential travel characteristics. A series of indicator data related to passenger travel characteristics are stored in a time-series format, resulting in higher query and analysis performance in scenarios analyzing historical travel data. The rule-based calculation of indicators is clearer in principle, easier to understand and implement than algorithms; and the use of a rule engine for indicator calculation facilitates flexible addition of new indicators and convenient adjustment of rule logic, allowing for faster adaptation to business changes. In the rail transit field, this can help enterprises quickly build business systems for analyzing passenger travel characteristics based on passenger travel data, greatly improving enterprise development and analysis efficiency and creating business value.

[0111] See Figure 11 This is a block diagram of a user travel data processing device provided in an embodiment of this disclosure. Corresponding to the method embodiments described above, such as... Figure 11 As shown, the provided user travel data processing device 1100 mainly includes:

[0112] The calling module 1101 is used to call the travel data processing model, wherein the travel data processing model includes a data sub-model, an indicator sub-model, and a graph sub-model;

[0113] Processing module 1102 is used to input the user's original travel data into the data sub-model to obtain travel detail data in a preset format;

[0114] The construction module 1103 is used to construct a travel knowledge graph corresponding to the user, wherein the travel knowledge graph includes event nodes corresponding to the original travel data, data nodes corresponding to the travel details data, spatiotemporal label nodes corresponding to the travel spatiotemporal data, feature label nodes corresponding to the travel features, and the relationship edges between the nodes;

[0115] The calculation module 1104 is used to call the indicator sub-model to calculate the user's travel indicator value based on the event nodes, data nodes, spatiotemporal nodes and relation edges in the user's travel knowledge graph;

[0116] The determination module 1105 is used to determine the travel characteristics of the user's tag feature nodes based on the travel index value.

[0117] According to a specific implementation of an embodiment of this disclosure, the construction module is further configured to:

[0118] Construct data sub-models, indicator sub-models, and graph sub-models;

[0119] Define the field recognition rules and field format processing rules for the data sub-model;

[0120] Define the indicator types, hierarchical relationships, and calculation rules for the indicator sub-models;

[0121] Define the entity type corresponding to each node of the graph sub-model and the edge type between the nodes.

[0122] According to a specific implementation of this disclosure, the indicator types include basic indicators and composite indicators, and the indicator calculation rules include a first type of calculation rule from basic indicators to composite indicators and a second type of calculation rule from composite indicators to travel characteristics. The calculation module is used for:

[0123] The indicator sub-model is invoked to determine the basic indicator values ​​of each basic indicator based on the event nodes, data nodes, spatiotemporal nodes, and relation edges in the user's travel knowledge graph;

[0124] Based on the values ​​of each basic indicator, the hierarchical relationship of indicators, and the first type of calculation rules for each composite indicator, calculate the composite indicator value of each composite indicator;

[0125] The user's travel index value is calculated based on the composite index values ​​of each composite index and the second type of calculation rules.

[0126] According to a specific implementation of an embodiment of this disclosure, the apparatus further includes a storage module, the storage module being used for:

[0127] The basic indicator values, composite indicator values, and travel indicator values ​​are stored according to their time-series characteristics.

[0128] Furthermore, this disclosure also provides an electronic device, the electronic device comprising:

[0129] The electronic device includes:

[0130] At least one processor; and,

[0131] A memory that is communicatively connected to at least one processor; wherein,

[0132] The memory stores instructions that can be executed by the at least one processor, which enables the at least one processor to perform the user travel data processing method provided in the foregoing embodiments.

[0133] Furthermore, this disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the user travel data processing method provided in the foregoing embodiments.

[0134] In addition, this disclosure also provides a computer program product, which includes a computing program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, which, when executed by a computer, cause the computer to perform the user travel data processing method provided in the foregoing embodiments.

[0135] The user travel data processing device and electronic device in the above-described embodiments of this disclosure, when processing user travel data, input the user's original travel data into the data sub-model to obtain travel detail data in a preset format; construct a travel knowledge graph corresponding to the user, wherein the travel knowledge graph includes event nodes corresponding to the original travel data, data nodes corresponding to the travel detail data, spatiotemporal label nodes corresponding to the travel spatiotemporal data, feature label nodes corresponding to travel characteristics, and relationship edges between nodes; call the indicator sub-model to calculate the user's travel indicator value based on the event nodes, data nodes, spatiotemporal nodes, and relationship edges in the user's travel knowledge graph; and determine the travel characteristics of the user's label feature nodes based on the travel indicator value. Through the scheme of this disclosure, a user travel knowledge graph is continuously constructed, and then association analysis and algorithm mining are performed based on the graph, resulting in higher association efficiency and analysis efficiency, and enabling more effective and accurate mining of passengers' potential travel characteristics. The specific implementation process of the provided user travel data processing device and electronic device can be found in the specific implementation process of the foregoing method embodiments, and will not be repeated here.

[0136] The above description is merely a specific embodiment of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure. Therefore, the scope of protection of this disclosure should be determined by the scope of the claims.

Claims

1. A method for processing user travel data, characterized in that, include: The travel data processing model is invoked, wherein the travel data processing model includes a data sub-model, an indicator sub-model, and a graph sub-model; Input the user's original travel data into the data sub-model to obtain travel detail data in a preset format; Construct a travel knowledge graph corresponding to the user, wherein the travel knowledge graph includes event nodes corresponding to the original travel data, data nodes corresponding to the travel details data, spatiotemporal label nodes corresponding to the travel spatiotemporal data, feature label nodes corresponding to the travel features, and the relationship edges between the nodes; The indicator sub-model is invoked to calculate the user's travel indicator value based on the event nodes, data nodes, spatiotemporal nodes, and relation edges in the user's travel knowledge graph; The user's travel characteristics are determined based on the travel index value; It also includes: constructing data sub-models, indicator sub-models, and graph sub-models; The indicator types include basic indicators and composite indicators. The indicator calculation rules include a first type of calculation rule from basic indicators to composite indicators and a second type of calculation rule from composite indicators to travel features. The step of calling the indicator sub-model to calculate the user's travel indicator value based on the event nodes, data nodes, spatiotemporal nodes, and relation edges in the user's travel knowledge graph includes: The indicator sub-model is invoked to determine the basic indicator values ​​of each basic indicator based on the event nodes, data nodes, spatiotemporal nodes, and relation edges in the user's travel knowledge graph; Based on the values ​​of each basic indicator, the hierarchical relationship of indicators, and the first type of calculation rules for each composite indicator, calculate the composite indicator value of each composite indicator; The user's travel index value is calculated based on the composite index values ​​of each composite index and the second type of calculation rules. Before the step of invoking the travel data processing model, the method further includes: Define the field recognition rules and field format processing rules for the data sub-model; Define the indicator types, hierarchical relationships, and calculation rules for the indicator sub-models; Define the entity type corresponding to each node of the graph sub-model and the edge type between the nodes.

2. The method according to claim 1, characterized in that, After the step of calculating the user's travel indicator value based on the composite indicator values ​​of each composite indicator and the second type of calculation rule, the method further includes: The basic indicator values, composite indicator values, and travel indicator values ​​are stored according to their time-series characteristics.

3. The method according to claim 2, characterized in that, Travel details data should include at least the user's identifier, origin and destination station identifiers, and travel time; and / or, Data nodes include username data nodes corresponding to user identifiers, station data nodes corresponding to start or end stations, route data nodes corresponding to travel routes, or ticket data nodes corresponding to travel tickets; and / or, Spatiotemporal tag nodes include time tag nodes and spatial tag nodes; and / or, Travel characteristics include travel stability characteristics.

4. A user travel data processing device, characterized in that, include: The calling module is used to call the travel data processing model, wherein the travel data processing model includes a data sub-model, an indicator sub-model, and a graph sub-model; The processing module is used to input the user's original travel data into the data sub-model to obtain travel detail data in a preset format; The construction module is used to construct a travel knowledge graph corresponding to the user, wherein the travel knowledge graph includes event nodes corresponding to the original travel data, data nodes corresponding to the travel details data, spatiotemporal label nodes corresponding to the travel spatiotemporal data, feature label nodes corresponding to the travel features, and the relationship edges between the nodes; The calculation module is used to call the indicator sub-model to calculate the user's travel indicator value based on the event nodes, data nodes, spatiotemporal nodes and relation edges in the user's travel knowledge graph; The determination module is used to determine the travel characteristics of the user's tag feature nodes based on the travel index value; It also includes constructing data sub-models, indicator sub-models, and graph sub-models; The indicator types include basic indicators and composite indicators. The indicator calculation rules include a first type of calculation rule from basic indicators to composite indicators and a second type of calculation rule from composite indicators to travel characteristics. The calculation module is used for: The indicator sub-model is invoked to determine the basic indicator values ​​of each basic indicator based on the event nodes, data nodes, spatiotemporal nodes, and relation edges in the user's travel knowledge graph; Based on the values ​​of each basic indicator, the hierarchical relationship of indicators, and the first type of calculation rules for each composite indicator, calculate the composite indicator value of each composite indicator; The user's travel index value is calculated based on the composite index values ​​of each composite index and the second type of calculation rules. The building module is also used for: Define the field recognition rules and field format processing rules for the data sub-model; Define the indicator types, hierarchical relationships, and calculation rules for the indicator sub-models; Define the entity type corresponding to each node of the graph sub-model and the edge type between the nodes.

5. The apparatus according to claim 4, characterized in that, The device further includes a storage module, the storage module being used for: The basic indicator values, composite indicator values, and travel indicator values ​​are stored according to their time-series characteristics.

6. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory that is communicatively connected to at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the user travel data processing method according to any one of claims 1 to 3.

Images