Knowledge graph-based traffic travel analysis method and device, equipment and medium

Abstract

The application discloses a traffic travel analysis method and device based on a knowledge graph, equipment and a medium. The method comprises the following steps: acquiring a pre-constructed traffic service knowledge graph, wherein the traffic service knowledge graph is constructed based on multi-source traffic data; acquiring a travel service demand of a user, querying a travel service model corresponding to the travel service demand; querying entity information associated with the travel service model based on the traffic service knowledge graph, and calculating a travel service analysis result according to the travel service model and the associated entity information. According to the traffic travel analysis method provided in the application, the public travel data related to urban traffic travel is processed, the urban traffic travel knowledge graph is established, and the graph operation theory is used for mining and calculation, so that data support is provided for the upper research and decision model in the form of a data interface, and the calculation efficiency of subsequent traffic travel analysis services is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent transportation, and in particular to a traffic travel analysis method and device based on a knowledge graph, equipment and a medium. BACKGROUND

[0002] With the continuous development of urban traffic travel, the data generated by the urban traffic system is also increasing, and the data types are also becoming more and more complex. For example, different track devices will generate a large number of data with complex content in different life cycles. At the same time, the data format and data structure are also various, which can include structured data, semi-structured data and unstructured data, etc. Obviously, in the face of the above situation, the traditional processing method based on artificial experience, business rules and operations research is difficult to process these differentiated and large amount of urban traffic data. Because the business involves traffic, traffic control, urban management, social enterprise and other multi-source units, the data surface relationship is low, which is not conducive to the correlation application of overall business data.

[0003] In the prior art, when analyzing travel services according to differentiated urban traffic data and analysis and judgment decision models, each data analyst mainly selects data tables and fields according to his understanding of the business, and stacks analysis logic codes according to his personal judgment. This approach has several problems. First, the learning cost of business understanding is high, and the understanding level is uneven, which is prone to understanding errors. Second, because the data is too complex, the data analysis logic is too complex, and the model construction cost is too high. Third, it is difficult to drill down the data, and each layer of drilling down requires corresponding data retrieval, data calculation and other logic. In addition, the existing data in the field of urban traffic travel mainly focuses on the core business of each business unit, and is not effectively bound with the spatial characteristics and aggregation characteristics of actual public travel. Therefore, it cannot reflect the actual travel characteristics, and is not conducive to the research and decision-making of urban traffic travel. SUMMARY

[0004] The embodiments of the present application provide a traffic travel analysis method and device based on a knowledge graph, equipment and a medium. In order to have a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This part is not a general review, nor is it intended to determine the key / important elements or describe the protection scope of these embodiments. The only purpose is to present some concepts in a simple form as a preface to the detailed description below.

[0005] In a first aspect, the embodiments of the present application provide a traffic travel analysis method based on a knowledge graph, comprising:

[0006] obtaining a pre-constructed traffic service knowledge graph, wherein the traffic service knowledge graph is constructed based on multi-source traffic data;

[0007] Obtaining a travel service demand of a user, querying a travel service model corresponding to the travel service demand;

[0008] Querying entity information associated with the travel service model based on the traffic service knowledge graph, and calculating a travel service analysis result based on the travel service model and the associated entity information.

[0009] In an optional embodiment, the traffic service knowledge graph is constructed in advance, including:

[0010] Constructing a traffic service knowledge graph ontology according to the association between people, vehicles, roads, environments, and travel factors;

[0011] Obtaining an entity data requirement list required for constructing the traffic service knowledge graph according to the traffic service knowledge graph ontology information;

[0012] Extracting entity data according to the entity data requirement list, and associating the entity data with the ontology to obtain a correspondence relationship of entities, attributes, and relationships;

[0013] Performing semi-supervised graph construction based on the correspondence relationship of entities, attributes, and relationships and a semi-supervised machine learning algorithm to obtain the constructed traffic service knowledge graph.

[0014] In an optional embodiment, after obtaining the constructed traffic service knowledge graph, the method further includes:

[0015] Calculating distances between road entities in the graph based on the constructed traffic service knowledge graph and a preset shortest path algorithm, and storing the calculated distance data in a database.

[0016] In an optional embodiment, the travel service demand includes one or more of a departure location and destination analysis service demand, a traffic volume analysis service demand, an accident-prone location analysis service demand, a traffic control analysis service demand, and a path planning analysis service demand.

[0017] The travel service model includes one or more of a departure location and destination analysis model, a traffic volume analysis model, an accident-prone location analysis model, a traffic control analysis model, and a path planning analysis model.

[0018] In an optional embodiment, the traffic service knowledge graph is queried based on the entity information associated with the travel service model, and the travel service analysis result is calculated based on the travel service model and the associated entity information, including:

[0019] When the travel service demand is a departure location and destination analysis demand, a departure location and destination analysis model corresponding to the demand is determined, and entity information associated with the departure location and destination analysis model is queried from the traffic service knowledge graph;

[0020] The departure and destination analysis result is calculated according to the departure and destination analysis model and the associated entity information, and the departure and destination analysis result is a POI-granularity accurate result.

[0021] In an optional embodiment, the entity information associated with the departure and destination analysis model is queried from the traffic service knowledge graph, and the departure and destination analysis result is calculated according to the departure and destination analysis model and the associated entity information, including:

[0022] The starting and ending gate information of the vehicle is obtained according to the departure and destination analysis requirement;

[0023] The POI entity within the preset range of the starting gate is queried according to the traffic service knowledge graph, and the queried POI entity is associated with the starting gate entity;

[0024] The POI entity within the preset range of the ending gate is queried according to the traffic service knowledge graph, and the queried POI entity is associated with the ending gate entity;

[0025] The query vector is constructed based on the time dimension, graph depth, data metric and link flow in the traffic service knowledge graph;

[0026] The departure and destination analysis result is queried from the traffic service knowledge graph according to the query vector.

[0027] In an optional embodiment, the departure and destination analysis result is queried from the traffic service knowledge graph according to the query vector, including:

[0028] The numerical values of the time dimension factor, the graph depth factor, the data metric factor and the link flow factor in the query vector are calculated;

[0029] The weight of each factor is calculated by using the analytic hierarchy process;

[0030] The value of the query vector of each path in the traffic service knowledge graph is calculated, and the value of the query vector is the sum of the product of each factor and the corresponding weight;

[0031] The starting point POI corresponding to the path with the maximum value of the query vector is taken as the departure of the vehicle, and the ending point POI of the path with the maximum value of the query vector is taken as the destination of the vehicle.

[0032] In a second aspect, the embodiments of the present application provide a traffic travel analysis device based on a knowledge graph, including:

[0033] A knowledge graph acquisition module is configured to acquire a pre-constructed traffic service knowledge graph, wherein the traffic service knowledge graph is constructed based on multi-source traffic data.

[0034] The travel service query module is configured to obtain a travel service demand of a user, and query a travel service model corresponding to the travel service demand.

[0035] The analysis module is configured to query entity information associated with the travel service model based on the traffic service knowledge graph, and calculate a travel service analysis result based on the travel service model and the associated entity information.

[0036] In a third aspect, an electronic device is provided, including a processor and a memory storing program instructions, the processor being configured to execute the program instructions to implement the traffic travel analysis method based on the knowledge graph provided in the above embodiments.

[0037] In a fourth aspect, a computer readable medium is provided, storing computer readable instructions, the computer readable instructions being executed by a processor to implement the traffic travel analysis method based on the knowledge graph provided in the above embodiments.

[0038] The technical solutions provided in the embodiments of the present application can include the following beneficial effects:

[0039] According to the traffic travel analysis method based on the knowledge graph provided in the embodiments of the present application, a pre-constructed traffic service knowledge graph can be obtained, the knowledge graph being constructed based on multi-source traffic data, and providing standardized data association query services for upper-layer travel service models in the form of data interfaces, obtaining a travel service analysis result corresponding to a travel service demand of a user, which greatly reduces the traffic service learning cost and data association calculation difficulty, and improves the calculation efficiency of subsequent traffic travel analysis services.

[0040] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0041] The accompanying drawings incorporated in the specification and forming a part thereof illustrate embodiments consistent with the present application and together with the description serve to explain the principles of the application.

[0042] Figure 1 is a flowchart of a traffic travel analysis method based on a knowledge graph according to an exemplary embodiment;

[0043] Figure 2 is a schematic diagram of a constructed traffic service knowledge graph according to an exemplary embodiment;

[0044] Figure 3 is a schematic diagram of a part of the knowledge graph in the vehicle dimension according to an exemplary embodiment;

[0045] Figure 4 FIG. 10 is a schematic diagram of recorded information of a vehicle start-stop clamp according to an example embodiment;

[0046] Figure 5 FIG. 11 is a schematic diagram of a POI entity around a clamp according to an example embodiment;

[0047] Figure 6 FIG. 12 is a schematic diagram of a knowledge graph of vehicle morning and evening travel according to an example embodiment;

[0048] Figure 7 FIG. 13 is a schematic diagram of an OD analysis result according to an example embodiment;

[0049] Figure 8 FIG. 14 is a schematic diagram of a structure of a traffic travel analysis device based on a knowledge graph according to an example embodiment;

[0050] Figure 9 FIG. 15 is a schematic diagram of a structure of an electronic device according to an example embodiment;

[0051] Figure 10 FIG. 16 is a schematic diagram of a computer storage medium according to an example embodiment. DETAILED DESCRIPTION

[0052] The following description and drawings are illustrative of the specific embodiments of the present application and are not intended to be limiting thereof.

[0053] It should be noted that the described embodiments are merely a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0054] The following description refers to the accompanying drawings. Unless otherwise indicated, same numbers in different drawings indicate same or similar elements. The embodiments described in the following example embodiments are not representative of all embodiments consistent with the present application. Rather, they are merely examples of systems and methods consistent with some aspects of the present application as detailed in the appended claims.

[0055] The knowledge graph-based traffic travel analysis method provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings. Referring to FIG. 10, the method specifically includes the following steps. Figure 1 S101, a pre-constructed traffic service knowledge graph is acquired, wherein the traffic service knowledge graph is constructed based on multi-source data.

[0056]

[0057] ​Among them, the knowledge graph is an important branch technology of artificial intelligence, which is essentially a semantic knowledge base, describing the concepts and relationships in the physical world in symbolic form, and forming a networked knowledge structure. The embodiments of the application are mainly based on the in-depth research on the business data in the field of urban traffic travel, establishing the link between multiple entities, realizing the multi-element expansion association between data and data, and constructing the knowledge graph of the field of urban traffic travel.

[0058] In an optional embodiment, first, according to the correlation between people, vehicles, roads, environments and travel factors, a traffic service knowledge graph ontology is constructed.

[0059] Specifically, the business of traffic, traffic control, city management, social enterprise service and other departments / units related to the field of urban traffic travel is analyzed, and according to the five core factors of people, vehicles, roads, environment and travel, the core factors are split into multiple sub-factors according to the top-down method, the relationship between each factor is described, multiple triples are formed, the triples include labels, entities and relationships, the label is the classification of the entity, the entity is the specific practice of the label, and the relationship is used to describe the direct contact between entities, a complete knowledge graph ontology is constructed, the design of the urban traffic travel knowledge graph framework is completed, and stored in a graph database.

[0060] Specifically, the five factors of people, vehicles, roads, environment and travel are taken as root nodes, and the knowledge graph is established according to the top-down method. The knowledge graph ontology is constructed, and the ontology includes labels class, entities entity and relationships relatio. The label class is the classification of the entity.

[0061] Among them, the entity is the specific physical object of the label, which is composed of basic information and support information. The basic information is necessary, and the support information is not necessary. The basic information is used to describe the specific appearance of the entity; the support information is the unique identification information that distinguishes each entity. The support information is usually composed of one or more attributes, and the support information should have business significance. In the scene where the entity uniqueness cannot be identified or cannot be identified, the support information can be without support information.

[0062] The relationship is used to describe the direct contact between entities, which is composed of basic information and support information. The basic information is necessary, and the support information is not necessary. The basic information includes attributes, labels, relationship names, FROM entities and TO entities; the support information has two groups of rules: FROM key combination rules and TO key combination rules. Through the key combination rules, a group of entities is associated with another group of entities.

[0063] Taking a car as an example, the knowledge graph ontology of the car dimension is established by step-by-step disassembly and refinement, and the correlation between the car dimension and other entity factors is established, and the rules are as follows: constructing an entity entity(rootnode, instanceof, class)->entity, extracting entity attributes property(entiy, type)->property[], and creating a correlation relation relation(from_entity, [r: {property: {name: value}}], to_entity)->r. As shown in the following table:

[0064]

[0065]

[0066]

[0067] The part of the knowledge graph ontology of the car dimension constructed according to the above table is shown in Table 1, and the construction method of other entity factors is similar to that of the car dimension. The correlation between each entity factor is obtained to obtain the overall knowledge graph ontology. The overall knowledge graph ontology constructed is shown in Table 2, which includes people, cars, roads, environments, travel, and other factors. Each entity factor is connected to each other to complete the relationship binding between data, and the knowledge graph ontology is obtained. Figure 3 Figure 2

[0068] Further, according to the traffic service knowledge graph ontology information, the entity data requirement list required for constructing the traffic service knowledge graph is obtained.

[0069] According to the constructed knowledge graph ontology, the entity data requirement list of each city traffic travel field is analyzed and sorted, including data table, data field, data explanation, etc. The city traffic travel data bottom table is constructed in the database to complete the data framework construction. Based on the entity data requirement list, a data extraction model and an extraction interface are established to realize the object extraction of the original multi-source data and store it in the relational database. If the quality of the original data is poor, data cleaning and correction work is required to ensure the quality of the data.

[0070] In an example scenario, according to the constructed knowledge graph ontology, the entity data requirement list is sorted, and the bus related data is taken as an example to illustrate. The sorted bus data requirement list is shown in the following table: bus line:

[0071] Encoding Explanation Format Xlbh Line number String Xlmc Line name String Xlfx Line direction String Xh Station sequence number String Zdbh Station number String Ssdw Belonging unit String Scsj Onboard time datetime

[0072] Bus station:

[0073] Encoding Explanation Format Zdbh Station number String Zdmc Station name String Jd Longitude Double Wd Latitude Double Scsj Upload time datetime

[0074] ​​Bus scheduling:

[0075]

[0076]

[0077] Bus vehicle positioning:

[0078] Encoding Explanation Format Cph License plate number String Jd Longitude String Wd Latitude String Sd Speed String Fx Direction String Sbsj Equipment time String Scsj Upload time String

[0079] Bus card swiping:

[0080] Encoding Explanation Format Kh Card number String Klx Card type String Xlbh Line number String Zdbh Station number String Cph License plate number String Skrq Card swiping date String Sksj Card swiping time String Scsj Upload time String

[0081] Further, business entity data is extracted according to the entity data requirement list, and the entity data is associated with the ontology to obtain the corresponding relationship of the entity, attribute, and relationship.

[0082] Specifically, based on the perception data hub / platform, extraction of original multi-source data is realized, wherein the multi-source data includes the card mouth passing vehicle data, illegal data, municipal bus data, Internet floating car data, network car data, parking data, shared bicycle data, and meteorological data. After completing data extraction and storage, an entity, attribute, and attribute value triple is formed for each entity, and then the entity is associated with the constructed knowledge graph ontology to complete multi-dimensional and wide-area "attribute value-attribute-entity, entity-relation-entity" mode link information association between entities, to obtain the corresponding relationship of the entity, attribute, and relationship, to realize deep linking of information at the data level, and to complete construction of the knowledge graph of the city traffic travel field.

[0083] When the entity is associated with the constructed knowledge graph ontology, the entity is first determined, the entity is the basis for the existence of the attribute, and is free, independent, and does not depend on other things. Then the attribute and the relationship are determined, the attribute is a feature describing a certain aspect of the entity, such as the date of birth, height, weight, etc., and the attribute is the basis for cognition and description of the world. The relationship is a special kind of attribute, when the attribute value of the entity is also an entity, the attribute is essentially a relationship. Finally, the attribute value is determined, the attribute value includes numerical value type, date, and text type. The determined entity, attribute, and attribute value are combined to realize the "attribute value-attribute-entity, entity-relation-entity" abstraction, and the reasoning operation on the knowledge graph means that the reasoning ends once an attribute is encountered.

[0084] In an optional embodiment, the business data extraction to the ontology is divided into entity extraction and relationship extraction, and the association of the business entity data and the ontology is realized.

[0085] The entity extraction includes single entity extraction and associated entity extraction. The single entity extraction is to extract independent entities from the business data, and the extracted content includes attributes, labels, and primary keys. The associated entity extraction is to extract two or more entities having a correlation relationship from the business data, and to extract the relationship between the entities. The associated entity extraction includes FROM_entity, TO_entity, and relationship. The relationship extraction includes automatic construction of ontology relationship and supplementary relationship. The automatic construction of the ontology relationship is to extract the relationship according to the rules of the relationship applied in the ontology, wherein the relationship applied in the ontology refers to the relationship between ontologies, and the relationship between entities is realized. The supplementary relationship extraction provides a FROM_entity and a TO_entity list data or rules to realize the relationship extraction of the existing entity, wherein the existing entity refers to the entity stored in the knowledge graph, and the extracted content includes TO__[entity|rule], FROM__[entity|rule], attributes, labels, and relationship names.

[0086] Further, the semi-supervised graph construction is performed according to the correspondence relationship among the constructed entities, relationships, and attributes and a semi-supervised machine learning algorithm, to obtain the constructed traffic service knowledge graph.

[0087] In a possible implementation, a part of data is associated with an ontology to obtain a correspondence relationship among entities, relationships, and attributes. Based on the correspondence relationship among the entities, relationships, and attributes, a semi-supervised machine learning algorithm is used to learn the correspondence relationship among the entities, relationships, and attributes in the constructed graph based on the ontology of the knowledge graph, to implement semi-supervised graph construction on the remaining data, and to complete the overall knowledge graph construction.

[0088] In an optional implementation, the method further includes calculating distances between road entities in the graph according to the constructed traffic service knowledge graph and a preset shortest path algorithm, and storing the calculated distance data in a database.

[0089] The knowledge graph is only a data-level correlation relationship calibration, and can be further explored according to a relationship link. In specific data applications, especially in urban traffic travel research and calculation processes, it is necessary to complete the maximum and most accurate relationship data query and calculation in the shortest time. The shortest path algorithm Dijkstra can be used to calculate the distances between road entities, and the calculated distance data is stored in a database to facilitate fast query and call.

[0090] The structured data in the graph of the embodiment of the application is stored on a network, contains attributes in the form of key / value, is connected through relationships defined by Relationships, forms a relational network structure, directly provides calculation data for algorithms, reduces data conversion, improves query efficiency, achieves services such as query, association query, result extraction, and result storage of underlying data for research and development, and supports high-concurrency data query. The data in the prior art is stored in a table, and further data extraction and data conversion are required. Therefore, the method of the application can improve the data query efficiency.

[0091] S102 obtains a travel service demand of a user, and queries a travel service model corresponding to the travel service demand.

[0092] City traffic travel is a comprehensive body integrating traffic, traffic control, city management, social enterprises and other multi-source businesses. Based on public travel orders, bus, rail, taxi, online car-hailing, shared bicycle operation, parking operation, road traffic and other data, the laws and characteristics of public travel, travel service, parking operation and road traffic are researched to provide decision support for traffic travel management and service.

[0093] In a possible implementation, the travel service demand input by the user is obtained, and the travel service demand can include one or more of a departure and destination analysis service demand, a traffic volume analysis service demand, an accident-prone place analysis service demand, a traffic control analysis service demand, and a path planning analysis service demand. The embodiment of the application is not limited specifically.

[0094] After receiving the travel service demand input by the user, a travel service model corresponding to the travel service demand is called, including one or more of a departure and destination analysis model (OD analysis model), a traffic volume analysis model, an accident-prone place analysis model, a traffic control analysis model, and a path planning analysis model. Each analysis model can be pre-constructed according to modeling and analysis algorithms in the prior art, for example, a traffic volume analysis model is obtained according to a traffic volume calculation algorithm, and an OD analysis model is obtained according to an OD analysis algorithm. The knowledge graph provided by the embodiment of the application can be applied to various traffic travel analysis scenarios, and the embodiment of the application is not limited specifically.

[0095] S103 queries entity information associated with the travel service model based on the traffic service knowledge graph, and calculates a travel service analysis result according to the travel service model and the associated entity information.

[0096] After obtaining the constructed knowledge graph, data relationship encapsulation can be performed on the basis. Specifically, in the business application, according to the travel service business demand, the relationship of each entity network in the knowledge graph is applied, the business logic is combined according to the entity relationship, the data relationship result data is stored in advance, and direct business statistical data or travel service API is generated, which provides data support and computing service for subsequent travel service model building, that is, the encapsulated data relationship can be a certain type of business statistical data involved in the travel service model, or a travel service API, that is, a travel service model.

[0097] In a specific embodiment, the constructed traffic travel knowledge graph serves as a base, and the traffic travel service model library is connected with the knowledge graph through an API service interface. When there is a traffic travel service demand, the associated entity information corresponding to the target traffic travel service model in the knowledge graph can be called based on the determined traffic travel service model corresponding to the traffic travel service demand. The associated entity information includes business statistical data obtained based on the encapsulated data relationship, the corresponding relationship of entities, attributes and relationships, the pre-calculated distance between road entities, etc., and then the travel service analysis result corresponding to the traffic travel service demand is calculated according to the traffic service model and the corresponding associated entity information.

[0098] A key scenario in the field of urban traffic travel is private car travel analysis. In combination with the private car OD analysis business, based on the card mouth passing data, the travel trajectory of each vehicle is connected in series and in parallel, and in combination with the travel regularity, event interval and other factors, a private car OD analysis model is constructed. The model includes information such as the departure place (O) of the vehicle, the destination (D), the travel time, the travel distance, the travel frequency, etc. However, this model has a problem, that is, the track point is based on the card mouth position, as shown in FIG. 1, the essence of model analysis is the starting card mouth point and the terminal card mouth point, and the obtained data is irrelevant to the actual spatial position of the city (such as community, hospital, business, etc.), and cannot effectively reflect the accurate spatial properties of public travel. Figure 4

[0099] In an optional embodiment, when the travel service demand is the departure place and destination analysis, the departure place and destination analysis model corresponding to the departure place and destination analysis demand is determined first, and the entity information associated with the departure place and destination analysis model is queried from the traffic service knowledge graph; the departure place and destination analysis result is calculated according to the departure place and destination analysis model and the associated entity information, and the departure place and destination analysis result is the accurate result of POI granularity.

[0100] ​Specifically, the starting and ending gate information of the vehicle is obtained according to the departure and destination analysis demand. The POI entities within the preset range of the starting gate are queried according to the traffic service knowledge graph, and the queried POI entities are associated with the starting gate entities. The POI entities within the preset range of the ending gate are queried according to the traffic service knowledge graph, and the queried POI entities are associated with the ending gate entities.

[0101] In one example scenario, the starting and ending gate information of the vehicle is analyzed according to a preset departure and destination analysis model. Based on the constructed traffic service knowledge graph, the entities such as roads, bus stations, communities, hospitals and schools within a radius of 500 meters with the gate coordinates as the center point are calculated and associated, and the gate position is given more spatial concept. As shown in Figure 5 The gate is associated with the nearby accurate POI entities, such as Yiyuan community, Provincial Second Hospital, Bikaiyuan community and Minsheng Bank.

[0102] As shown in Figure 6 The full graph of a certain vehicle in the morning and evening for half a year is shown. Usually in a period of time, the relationship between each vehicle and the starting gate is r{1, n}, that is, one vehicle may correspond to multiple gates. The relationship between the starting gate and the accurate POI starting point (such as a community) is r{n, n}, that is, each gate corresponds to multiple POI points, and each POI point may also correspond to multiple gates. Through recursive and simulation composite calculation of the path, the relationship r{1, 1} is finally obtained. That is, the spatial particles of the starting point and the ending point are continuously refined, and the relatively accurate starting point and destination of the vehicle are obtained, so that the private car travel rule and demand are more accurately studied, and decision support is provided for further bus line planning and parking lot planning.

[0103] Specifically, based on the constructed knowledge graph, the spatial particles of the starting point and the ending point are continuously refined to obtain accurate departure and destination analysis results. Including: based on the time dimension, graph depth, data metric and link flow in the traffic service knowledge graph, a query vector is constructed. The query vector is combined into odvector = {time && deap && num && join}, and the departure and destination analysis results are queried from the traffic service knowledge graph according to the query vector.

[0104] First, the values of the time dimension factor, the graph depth factor, the data metric factor and the link flow factor in the query vector are calculated. The time dimension is the time period contained in the data in the graph, the graph depth is the number of gates passed by the vehicle, the data metric is the number of times the vehicle passes through the gate, and the link flow is the number of vehicle times from a gate to a POI point.

[0105] Further, the analytic hierarchy process is used to calculate the weight of each factor. For example, referring to the analytic hierarchy process AHP, the weight of each factor is calculated, and the 1-5 scale method is applied as shown in the following table:

[0106] time deap num join time 1 0.333333 0.5 0.5 deap 3 1 2 5 num 2 0.5 1 5 join 2 0.5 0.5 1

[0107] Obtain the weight result:

[0108]

[0109] Further, calculate the value of the query vector for each path in the traffic service knowledge graph. The value of the query vector is the sum of the products of each factor and its corresponding weight. The starting point POI corresponding to the path with the largest query vector value is taken as the departure place of the vehicle, and the end point POI of the path with the largest query vector value is taken as the destination of the vehicle.

[0110] In one embodiment, the query condition Querypath = {Querypath = {MATCH(n:Car{name='Ji A9845Y'})-[c:pass:0..1]-(deep:kk:*)-[l:join*]>(a:untity{name:'Yiyuan'})wheremax(sum(time*0.12094,deep*0.41680,pass.num*0.26.94,join*0.19278))return a;}. As Figure 7 shown, according to the query method provided in the embodiment of the present application, by inputting the query vector, the starting point POI is "Yiyuan". Similarly, by inputting the query vector related to the destination checkpoint, the accurate destination POI can be obtained.

[0111] In an optional embodiment, other traffic travel business analysis calculations can also be performed based on the knowledge graph in the embodiment of the present application. For example, according to the congestion section analysis model, the checkpoint information of the congestion section is analyzed, and combined with the knowledge graph, the accurate POI location point is further determined to reflect the spatial characteristics of actual travel.

[0112] In an optional embodiment, other traffic travel business analysis calculations can also be performed based on the knowledge graph in the embodiment of the present application. For example, an accident-prone area analysis model is constructed based on the person, vehicle, road, passage, and environment association data in the knowledge graph, and combined with the accurate spatial characteristics in the graph, the accurate accident-prone section is analyzed, reducing the difficulty of data processing, and the calculation accuracy can be improved based on the association data, reflecting the spatial characteristics of the accurate accident-prone section.

[0113] The traffic travel analysis method provided by the embodiments of the present application is to research public travel data, vehicle (including private cars, buses, taxis, etc.) operation data, road traffic data, spatial geographic data and other information related to urban traffic travel, organize multi-dimensional data relationship network, construct the urban traffic travel knowledge graph through entity extraction and entity mapping. Based on the graph operation theory, the data interface is used to provide standardized data correlation query services for the upper research and decision-making model, which greatly reduces the business learning cost and data correlation calculation difficulty, and improves the calculation efficiency of subsequent traffic travel analysis services.

[0114] The embodiments of the present application also provide a traffic travel analysis device based on a knowledge graph, which is used to execute the traffic travel analysis method based on the knowledge graph of the above embodiments, as shown in Figure 8 The device comprises:

[0115] The knowledge graph acquisition module 801 is configured to acquire a pre-constructed traffic service knowledge graph, wherein the traffic service knowledge graph is constructed based on multi-source traffic data.

[0116] The travel service query module 802 is configured to acquire a travel service demand of a user and query a travel service model corresponding to the travel service demand.

[0117] The analysis module 803 is configured to query entity information associated with the travel service model based on the traffic service knowledge graph, and calculate a travel service analysis result according to the travel service model and the associated entity information.

[0118] It should be noted that the traffic travel analysis device based on the knowledge graph provided by the above embodiments is only used as an example to divide the above functional modules when the traffic travel analysis method based on the knowledge graph is executed. In actual application, the above functions can be completed by different functional modules according to the needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the traffic travel analysis device based on the knowledge graph provided by the above embodiments and the traffic travel analysis method based on the knowledge graph belong to the same concept, and the implementation process is described in detail in the method embodiments. Here, it is not repeated.

[0119] The embodiments of the present application also provide an electronic device corresponding to the traffic travel analysis method based on the knowledge graph provided by the above embodiments, to execute the traffic travel analysis method based on the knowledge graph.

[0120] Please refer to Figure 9 which shows a schematic diagram of an electronic device provided by some embodiments of the present application. As shown in Figure 9As shown, the electronic device includes a processor 900, a memory 901, a bus 902, and a communication interface 903, the processor 900, the communication interface 903, and the memory 901 are connected through the bus 902; the memory 901 stores a computer program capable of running on the processor 900, and the processor 900 executes the computer program to perform the knowledge graph-based traffic travel analysis method provided by any one of the preceding embodiments of the present application.

[0121] The memory 901 can include a high-speed random access memory (RAM: Random Access Memory) and can also include a non-volatile memory such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 903 (which can be wired or wireless), and the Internet, a wide area network, a local network, a metropolitan area network, etc. can be used.

[0122] The bus 902 can be an ISA bus, a PCI bus, or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory 901 is used to store programs, and the processor 900 executes the programs after receiving execution instructions. The knowledge graph-based traffic travel analysis method disclosed in any one of the preceding embodiments of the present application can be applied to the processor 900 or implemented by the processor 900.

[0123] The processor 900 can be an integrated circuit chip having a processing capability of signals. In the implementation process, each step of the above method can be completed by the integrated logic circuit of hardware in the processor 900 or the instruction in the form of software. The processor 900 described above can be a general processor, including a central processing unit (CPU), a network processor (NP), etc.; can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a ready programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. Each method, step and logic block diagram disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as a hardware code processor for execution, or a combination of hardware and software modules in the code processor for execution. The software module can be located in a random access memory, a flash memory, a read only memory, a programmable read only memory or an electrically erasable programmable memory, a register, etc. The storage medium in the art. The storage medium is located in the memory 901, and the processor 900 reads the information in the memory 901, and combines the hardware to complete the steps of the above method.

[0124] The electronic device provided by the embodiments of the present application and the traffic travel analysis method based on the knowledge graph provided by the embodiments of the present application have the same beneficial effects as the method adopted, run or implemented.

[0125] The embodiments of the present application also provide a computer readable storage medium corresponding to the traffic travel analysis method based on the knowledge graph provided by the preceding embodiments. Please refer to Figure 10 The computer readable storage medium shown in the figure is an optical disc 1000, and a computer program (i.e. program product) is stored on the optical disc 1000. When the computer program is run by the processor, the traffic travel analysis method based on the knowledge graph provided by any of the preceding embodiments will be executed.

[0126] It should be noted that examples of the computer readable storage medium can also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other optical, magnetic storage medium, which will not be described one by one here.

[0127] The computer readable storage medium provided by the above embodiments of the present application and the traffic travel analysis method based on a knowledge graph provided by the embodiments of the present application have the same beneficial effects as the method adopted, run or implemented by the application program stored therein.

[0128] The technical features of the above embodiments can be combined in any manner. To make the description concise, all possible combinations of the technical features in the above embodiments are not described, but as long as the combinations of the technical features do not exist, they should be considered as the scope of the present disclosure.

[0129] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation on the patent scope of the present application. It should be pointed out that, for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application patent should be subject to the appended claims.

Claims

1. A knowledge graph-based traffic travel analysis method, characterized in that, The method comprises the following steps: acquiring a pre-constructed traffic service knowledge graph, wherein the traffic service knowledge graph is constructed based on multi-source traffic data; acquiring a user's travel service demand, and querying a travel service model corresponding to the travel service demand; querying entity information associated with the travel service model based on the traffic service knowledge graph, and calculating a travel service analysis result based on the travel service model and the associated entity information; when the travel service demand is a departure location and destination analysis demand, a departure location and destination analysis model corresponding to the demand is determined, and starting and ending toll gate information of a vehicle is obtained according to the departure location and destination analysis demand; POI entities within a preset range of the starting toll gate are queried based on the traffic service knowledge graph, and the queried POI entities are associated with the starting toll gate entity; POI entities within a preset range of the ending toll gate are queried based on the traffic service knowledge graph, and the queried POI entities are associated with the ending toll gate entity; a query vector is constructed based on a time dimension, graph depth, data metric and link flow in the traffic service knowledge graph; and a departure location and destination analysis result is queried from the traffic service knowledge graph based on the query vector; the departure location and destination analysis result is a POI-granularity accurate result.

2. The method of claim 1, wherein, The method for pre-constructing a traffic service knowledge graph comprises the following steps: constructing a traffic service knowledge graph ontology according to the correlation between people, vehicles, roads, environment and travel factors; acquiring an entity data requirement list required for constructing the traffic service knowledge graph according to the traffic service knowledge graph ontology information; extracting entity data according to the entity data requirement list, and associating the entity data with the ontology to obtain a corresponding relationship among entities, attributes and relationships; performing semi-supervised graph construction according to the corresponding relationship among the entities, attributes and relationships and a semi-supervised machine learning algorithm to obtain the constructed traffic service knowledge graph.

3. The method of claim 2, wherein, After obtaining the constructed traffic service knowledge graph, the following steps are further included: calculating distances between road entities in the graph based on the constructed traffic service knowledge graph and a preset shortest path algorithm, and storing the calculated distance data in a database.

4. The method of claim 1, wherein, The travel service demand comprises one or more of a departure location and destination analysis service demand, a traffic volume analysis service demand, an accident-prone location analysis service demand, a traffic control analysis service demand and a path planning analysis service demand. The travel service model comprises one or more of a departure location and destination analysis model, a traffic volume analysis model, an accident-prone location analysis model, a traffic control analysis model and a path planning analysis model.

5. The method of claim 1, wherein, Querying a departure location and destination analysis result from the traffic service knowledge graph based on the query vector comprises the following steps: calculating numerical values of a time dimension factor, a graph depth factor, a data metric factor and a link flow factor in the query vector; calculating the weight of each factor by using an analytic hierarchy process; and calculating a value of a query vector of each path in the traffic service knowledge graph, the value of the query vector being a sum of products of each factor and its corresponding weight; taking a starting point POI corresponding to a path with the maximum value of the query vector as a departure place of the vehicle, and taking a terminal point POI of the path with the maximum value of the query vector as a destination of the vehicle. 6.A knowledge graph-based traffic travel analysis apparatus, characterized in that, The method comprises the following steps: a knowledge graph acquisition module configured to acquire a pre-constructed traffic service knowledge graph, wherein the traffic service knowledge graph is constructed based on multi-source traffic data; a travel service query module configured to acquire a travel service demand of a user and query a travel service model corresponding to the travel service demand; an analysis module configured to query entity information associated with the travel service model based on the traffic service knowledge graph, and calculate a travel service analysis result according to the travel service model and the associated entity information; when the travel service demand is a departure place and destination analysis demand, the analysis module is configured to determine a departure place and destination analysis model corresponding to the demand, obtain starting and terminal gate information of a vehicle according to the departure place and destination analysis demand, query POI entities within a preset range of the starting gate based on the traffic service knowledge graph, and associate the queried POI entities with the starting gate entity; query POI entities within a preset range of the terminal gate based on the traffic service knowledge graph, and associate the queried POI entities with the terminal gate entity; construct a query vector based on a time dimension, graph depth, data metric, and link flow in the traffic service knowledge graph; and query a departure place and destination analysis result from the traffic service knowledge graph according to the query vector; the departure place and destination analysis result is a POI-granularity accurate result.

7. An electronic device, comprising: The method comprises a processor and a memory storing program instructions, and the processor is configured to execute the program instructions to perform the knowledge graph-based traffic travel analysis method according to any one of claims 1 to 5.

8. A computer readable medium characterized by A computer readable instruction is stored thereon, and the computer readable instruction is executed by a processor to implement the knowledge graph-based traffic travel analysis method according to any one of claims 1 to 5.

Images