A method, device and apparatus for recommending vehicle services based on user behavior

Abstract

The embodiment of the present invention relates to the field of automobile sales technology, and discloses a method, device and equipment for recommending vehicle services based on user behavior. The method for recommending vehicle services based on user behavior establishes a user behavior knowledge graph through the historical behavior data of historical store users in the target area, and then uses the user behavior knowledge graph to predict the target behavior prediction value corresponding to the target user's behavior data, thereby generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user. Applying the technical solution of the present invention, it is possible to generate a recommendation strategy corresponding to the target customer to help sales staff accurately identify users with stronger intentions to implement the target behavior, and adaptively serve users based on the recommendation strategy, thereby improving the overall service quality of the store and increasing the transaction volume of products or services.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of automobile sales technology, and specifically to a method, device and equipment for recommending vehicle services based on user behavior. Background Art

[0002] The car experience center (sales store) plays a vital role in the entire product closed loop. It is very important to improve the service quality and efficiency of the store and bring higher sales conversion rate to the store. This requires not only good service from the store staff, but also the completeness of various software and hardware facilities of the store.

[0003] In addition to these two services visible to the naked eye, the sales strategy of sales staff is also an important factor affecting transaction volume. However, the sales strategy of sales staff is generally based on personal experience, and this experience, as a personal ability, is difficult to popularize among sales staff. Although certain experiences can be summarized for sales staff to learn, when the sales staff themselves lack observation skills or there are many customers in the car experience center, it is easy for sales staff to fail to accurately identify customers with stronger intentions to implement target behaviors (such as consumption intentions), thereby affecting the overall car transaction volume of the car experience center. Summary of the invention

[0004] In view of the above problems, the embodiments of the present invention provide a method, device and equipment for recommending vehicle services based on user behavior, which is used to solve the problem in the prior art that due to the sales staff's own lack of observation or the large number of customers in the car experience center, the sales staff are unable to accurately identify customers with a stronger intention to implement the target behavior, thereby affecting the overall car transaction volume of the car experience center.

[0005] According to one aspect of an embodiment of the present invention, a method for recommending vehicle services based on user behavior is provided, and the method for recommending vehicle services based on user behavior includes:

[0006] Acquire historical behavior data, and establish a user behavior knowledge graph based on the historical behavior data; the historical behavior data includes multiple behavior events of historical store users in a target area, and behavior objects corresponding to the behavior events, wherein the behavior event includes at least one behavior object, the user behavior knowledge graph includes a master node and at least one slave node, the master node corresponds to the historical store user, the slave node corresponds to the behavior object, and the edge attributes between the master node and the slave node, and between the slave nodes and the slave nodes are the behavior events and the edge weights corresponding to the behavior events;

[0007] Obtaining the behavior data of the target user in the target area, and generating a target behavior prediction value of the target user based on the behavior data of the target user and the user behavior knowledge graph;

[0008] Based on the target behavior prediction value of the target user, a recommendation strategy corresponding to the target customer is generated.

[0009] In an optional manner, the historical behavior data also includes the duration of the behavior corresponding to the behavior event;

[0010] In the step of establishing a user behavior knowledge graph based on the historical behavior data, it also includes:

[0011] Obtain the total store-entry time of the historical store-entering users;

[0012] Based on the total duration of entering the store and the duration of the behavior corresponding to the behavior event, the edge weight corresponding to the behavior event is determined.

[0013] In an optional manner, the step of establishing a user behavior knowledge graph based on the historical behavior data further includes:

[0014] If the behavior event includes a plurality of behavior objects, the edge weight corresponding to the behavior event is allocated to the edge attributes corresponding to the plurality of behavior objects according to the number of the plurality of behavior objects.

[0015] In an optional manner, after the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes:

[0016] Based on the user behavior knowledge graph, the behavior objects in the historical behavior data of multiple historical users entering the store, the frequencies corresponding to the behavior objects, and the layout environment in the target area are analyzed to determine the behavior objects with abnormal frequencies and the corresponding abnormal types in the historical behavior data of the multiple historical users entering the store.

[0017] In an optional manner, after the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes:

[0018] Historical behavior data including a first behavior object is obtained, and based on the target behavior record in the historical behavior data including the first behavior object, a correlation between the first behavior object and the consumption behavior is determined.

[0019] In an optional manner, the method for recommending vehicle services based on user behavior further includes:

[0020] Acquire a location level of the first behavior object in the target area, where the location level is used to indicate the importance of the behavior object in the target area;

[0021] Based on the location level of the first behavior object in the target area and the correlation between the first behavior object and the target behavior, a first instruction for adjusting the location level of the first behavior object is generated, or a second instruction for maintaining the location level of the first behavior object is generated.

[0022] In an optional manner, after the step of generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user, the method further includes:

[0023] Pushing service information corresponding to the recommendation strategy to the target user;

[0024] Obtain feedback information of the target user corresponding to the service information, and modify the user behavior knowledge graph based on the feedback information.

[0025] In an optional manner, the step of generating a target behavior prediction value of the target user based on the target user's behavior data and the user behavior knowledge graph includes:

[0026] In the behavior data of the target user, determining all behavior events and corresponding behavior objects of the target user;

[0027] According to all the behavior events and the corresponding behavior objects, in the user behavior knowledge graph, determining the edge weights corresponding to all the behavior events in the behavior data of the target user;

[0028] A target behavior prediction value of the target user is generated according to the edge weights corresponding to all behavior events in the behavior data of the target user.

[0029] According to another aspect of an embodiment of the present invention, a device for recommending vehicle services based on user behavior is provided, comprising: a knowledge graph establishment module, a target behavior prediction module, a recommendation strategy generation module, a behavior object anomaly analysis module, a behavior relevance generation module, a location level acquisition module instruction generation module and a recommendation strategy execution module.

[0030] The knowledge graph establishment module is used to obtain historical behavior data and establish a user behavior knowledge graph based on the historical behavior data; the historical behavior data includes behavior events of multiple historical store users in a target area, and behavior objects corresponding to the behavior events, wherein the behavior events include at least one behavior object, and the user behavior knowledge graph includes a master node and at least one slave node, the master node corresponds to the historical store users, the slave node corresponds to the behavior object, and the edge attributes between the master node and the slave nodes, and between the slave nodes are behavior events and edge weights corresponding to the behavior events.

[0031] The target behavior prediction module is used to obtain the behavior data of the target user in the target area, and generate the target behavior prediction value of the target user based on the behavior data of the target user and the user behavior knowledge graph.

[0032] The recommendation strategy generation module is used to generate a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user.

[0033] In an optional manner, the knowledge graph building module is also used to obtain the total time spent in the store by the historical users who have entered the store.

[0034] And, it is used to determine the edge weight corresponding to the behavior event based on the total time of entering the store and the duration of the behavior corresponding to the behavior event.

[0035] In an optional manner, the knowledge graph establishment module is also used to allocate the edge weights corresponding to the behavior event to the edge attributes corresponding to the multiple behavior objects according to the number of the multiple behavior objects if the behavior event includes multiple behavior objects.

[0036] In an optional manner, the behavior object anomaly analysis module is also used to analyze the behavior objects in the historical behavior data of multiple historical users entering the store, the frequencies corresponding to the behavior objects, and the layout environment in the target area based on the user behavior knowledge graph, and determine the behavior objects with abnormal frequencies and the corresponding anomaly types in the historical behavior data of the multiple historical users entering the store.

[0037] In an optional manner, the behavior relevance generation module is used to obtain historical behavior data containing a first behavior object, and determine the relevance between the first behavior object and the target behavior based on the target behavior record in the historical behavior data containing the first behavior object.

[0038] The location level acquisition module is used to acquire the location level of the first behavior object in the target area, and the location level is used to indicate the importance of the behavior object in the target area.

[0039] An instruction generation module is used to generate a first instruction for adjusting the location level of the first behavior object based on the location level of the first behavior object in the target area and the correlation between the first behavior object and the target behavior, or to generate a second instruction for maintaining the location level of the first behavior object.

[0040] In an optional manner, the recommendation strategy execution module is used to push service information corresponding to the recommendation strategy to the target user;

[0041] Obtain feedback information of the target user corresponding to the service information, and modify the user behavior knowledge graph based on the feedback information.

[0042] In an optional manner, the target behavior prediction module is further used to determine all behavior events and corresponding behavior objects of the target user in the behavior data of the target user;

[0043] According to all the behavior events and the corresponding behavior objects, in the user behavior knowledge graph, determining the edge weights corresponding to all the behavior events in the behavior data of the target user;

[0044] A target behavior prediction value of the target user is generated according to the edge weights corresponding to all behavior events in the behavior data of the target user.

[0045] The embodiment of the present invention provides a vehicle service recommendation device based on user behavior. The knowledge graph establishment module establishes a user behavior knowledge graph. Then, the target behavior prediction module uses the user behavior knowledge graph to predict the target behavior prediction value corresponding to the target user's behavior data. Finally, the recommendation strategy generation module generates a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user. The technical solution of the embodiment of the present invention can generate a recommendation strategy corresponding to the target customer to help sales staff accurately identify users with stronger intentions to implement target behaviors, and adaptively serve users according to the recommendation strategy, thereby improving the overall service quality of the store and increasing the transaction volume of products or services.

[0046] According to another aspect of an embodiment of the present invention, a device for recommending vehicle services based on user behavior is provided, including: a processor, a communications interface, a memory, and a communication bus.

[0047] Specifically, the program may include program code including computer executable instructions.

[0048] The processor may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention. The one or more processors included in the recommendation device for vehicle services based on user behavior may be processors of the same type, such as one or more CPUs; or may be processors of different types, such as one or more CPUs and one or more ASICs.

[0049] The memory is used to store programs. The memory may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory.

[0050] The program can be specifically called by the processor to enable the vehicle service recommendation device based on user behavior to perform the following operations:

[0051] Acquire historical behavior data, and establish a user behavior knowledge graph based on the historical behavior data; the historical behavior data includes multiple behavior events of historical store users in a target area, and behavior objects corresponding to the behavior events, wherein the behavior events include at least one behavior object, and the user behavior knowledge graph includes a master node and at least one slave node, the master node corresponds to the historical store users, the slave node corresponds to the behavior object, and the edge attributes between the master node and the slave nodes, and between the slave nodes and the slave nodes are behavior events and edge weights corresponding to the behavior events.

[0052] The behavior data of the target user in the target area is obtained, and based on the behavior data of the target user and the user behavior knowledge graph, a target behavior prediction value of the target user is generated.

[0053] Based on the target behavior prediction value of the target user, a recommendation strategy corresponding to the target customer is generated.

[0054] In an optional manner, the historical behavior data also includes the duration of the behavior corresponding to the behavior event;

[0055] In the step of establishing a user behavior knowledge graph based on the historical behavior data, it also includes:

[0056] Get the total in-store time of the historical in-store users.

[0057] Based on the total duration of entering the store and the duration of the behavior corresponding to the behavior event, the edge weight corresponding to the behavior event is determined.

[0058] In an optional manner, the step of establishing a user behavior knowledge graph based on the historical behavior data further includes:

[0059] If the behavior event includes a plurality of behavior objects, the edge weight corresponding to the behavior event is allocated to the edge attributes corresponding to the plurality of behavior objects according to the number of the plurality of behavior objects.

[0060] In an optional manner, after the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes:

[0061] Based on the user behavior knowledge graph, the behavior objects in the historical behavior data of multiple historical users entering the store, the frequencies corresponding to the behavior objects, and the layout environment in the target area are analyzed to determine the behavior objects with abnormal frequencies and the corresponding abnormal types in the historical behavior data of the multiple historical users entering the store.

[0062] In an optional manner, after the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes:

[0063] Historical behavior data including a first behavior object is obtained, and a correlation between the first behavior object and the target behavior is determined based on the target behavior record in the historical behavior data including the first behavior object.

[0064] In an optional manner, the method further includes:

[0065] A location level of the first behavior object in the target area is obtained, where the location level is used to indicate the importance of the behavior object in the target area.

[0066] Based on the location level of the first behavior object in the target area and the correlation between the first behavior object and the target behavior, a first instruction for adjusting the location level of the first behavior object is generated, or a second instruction for maintaining the location level of the first behavior object is generated.

[0067] In an optional manner, after the step of generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user, the method further includes:

[0068] Pushing service information corresponding to the recommendation strategy to the target user;

[0069] Obtain feedback information of the target user corresponding to the service information, and modify the user behavior knowledge graph based on the feedback information.

[0070] In an optional manner, the step of generating a target behavior prediction value of the target user based on the target user's behavior data and the user behavior knowledge graph includes:

[0071] In the behavior data of the target user, determining all behavior events and corresponding behavior objects of the target user;

[0072] According to all the behavior events and the corresponding behavior objects, in the user behavior knowledge graph, determining the edge weights corresponding to all the behavior events in the behavior data of the target user;

[0073] A target behavior prediction value of the target user is generated according to the edge weights corresponding to all behavior events in the behavior data of the target user.

[0074] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a vehicle service recommendation device / apparatus based on user behavior to perform the following operations:

[0075] Acquire historical behavior data, and establish a user behavior knowledge graph based on the historical behavior data; the historical behavior data includes multiple behavior events of historical store users in a target area, and behavior objects corresponding to the behavior events, wherein the behavior events include at least one behavior object, and the user behavior knowledge graph includes a master node and at least one slave node, the master node corresponds to the historical store users, the slave node corresponds to the behavior object, and the edge attributes between the master node and the slave nodes, and between the slave nodes and the slave nodes are behavior events and edge weights corresponding to the behavior events.

[0076] The behavior data of the target user in the target area is obtained, and based on the behavior data of the target user and the user behavior knowledge graph, a target behavior prediction value of the target user is generated.

[0077] Based on the target behavior prediction value of the target user, a recommendation strategy corresponding to the target customer is generated.

[0078] In an optional manner, the historical behavior data also includes the duration of the behavior corresponding to the behavior event;

[0079] In the step of establishing a user behavior knowledge graph based on the historical behavior data, it also includes:

[0080] Get the total in-store time of the historical in-store users.

[0081] Based on the total duration of entering the store and the duration of the behavior corresponding to the behavior event, the edge weight corresponding to the behavior event is determined.

[0082] In an optional manner, the step of establishing a user behavior knowledge graph based on the historical behavior data further includes:

[0083] If the behavior event includes a plurality of behavior objects, the edge weight corresponding to the behavior event is allocated to the edge attributes corresponding to the plurality of behavior objects according to the number of the plurality of behavior objects.

[0084] In an optional manner, after the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes:

[0085] Based on the user behavior knowledge graph, the behavior objects in the historical behavior data of multiple historical users entering the store, the frequencies corresponding to the behavior objects, and the layout environment in the target area are analyzed to determine the behavior objects with abnormal frequencies and the corresponding abnormal types in the historical behavior data of the multiple historical users entering the store.

[0086] In an optional manner, after the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes:

[0087] Historical behavior data including a first behavior object is obtained, and a correlation between the first behavior object and the target behavior is determined based on the target behavior record in the historical behavior data including the first behavior object.

[0088] In an optional manner, it also includes: obtaining a location level of the first behavior object in the target area, wherein the location level is used to indicate the importance of the behavior object in the target area.

[0089] Based on the location level of the first behavior object in the target area and the correlation between the first behavior object and the target behavior, a first instruction for adjusting the location level of the first behavior object is generated, or a second instruction for maintaining the location level of the first behavior object is generated.

[0090] In an optional manner, after the step of generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user, the method further includes: pushing service information corresponding to the recommendation strategy to the target user;

[0091] Obtain feedback information of the target user corresponding to the service information, and modify the user behavior knowledge graph based on the feedback information.

[0092] In an optional manner, the step of generating a target behavior prediction value of the target user based on the target user's behavior data and the user behavior knowledge graph includes:

[0093] In the behavior data of the target user, determining all behavior events and corresponding behavior objects of the target user;

[0094] According to all the behavior events and the corresponding behavior objects, in the user behavior knowledge graph, determining the edge weights corresponding to all the behavior events in the behavior data of the target user;

[0095] A target behavior prediction value of the target user is generated according to the edge weights corresponding to all behavior events in the behavior data of the target user.

[0096] The present invention provides a method for recommending vehicle services based on user behavior. The method establishes a user behavior knowledge graph through the historical behavior data of historical store users in the target area, and then uses the user behavior knowledge graph to predict the target behavior prediction value corresponding to the target user's behavior data, thereby generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user. The technical solution of the embodiment of the present invention can generate a recommendation strategy corresponding to the target customer to help sales staff accurately identify users with stronger intentions to implement the target behavior, and adaptively serve users according to the recommendation strategy, thereby improving the overall service quality of the store and increasing the transaction volume of products or services.

[0097] The embodiment of the present invention establishes a user behavior knowledge graph through the historical behavior data of historical store users in the target area, and then uses the user behavior knowledge graph to predict the target behavior prediction value corresponding to the behavior data of the target user, thereby generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user. The technical solution of the embodiment of the present invention can generate a recommendation strategy corresponding to the target customer to help sales staff accurately identify users with stronger intentions to implement the target behavior, and adaptively serve users according to the recommendation strategy, thereby improving the overall service quality of the store and increasing the transaction volume of products or services.

[0098] The above description is only an overview of the technical solution of the embodiment of the present invention. In order to more clearly understand the technical means of the embodiment of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiment of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] The accompanying drawings are only used to illustrate the embodiments and are not to be considered as limiting the present invention. In addition, the same reference symbols are used to represent the same components throughout the accompanying drawings. In the accompanying drawings:

[0100] Figure 1 A schematic flow chart of a first embodiment of a method for recommending vehicle services based on user behavior provided by the present invention is shown;

[0101] Figure 2 A first structural schematic diagram of a user behavior knowledge graph provided by the present invention is shown;

[0102] Figure 3 A first structural schematic diagram of a user behavior knowledge graph provided by the present invention is shown;

[0103] Figure 4 A flow chart showing a second embodiment of a method for recommending vehicle services based on user behavior provided by the present invention;

[0104] Figure 5 A schematic flow chart of a third embodiment of a method for recommending vehicle services based on user behavior provided by the present invention is shown;

[0105] Figure 6 A schematic diagram showing the structure of an embodiment of a vehicle service recommendation device based on user behavior provided by the present invention;

[0106] Figure 7 A schematic structural diagram of an embodiment of a vehicle service recommendation device based on user behavior provided by the present invention is shown. DETAILED DESCRIPTION

[0107] Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0108] Figure 1 The flowchart of the first embodiment of a method for recommending vehicle services based on user behavior provided by the present invention is shown. The method for recommending vehicle services based on user behavior is executed by a device for recommending vehicle services based on user behavior. Figure 1 As shown, the method for recommending vehicle services based on user behavior includes the following steps:

[0109] Step 110: Obtain historical behavior data, and establish a user behavior knowledge graph based on the historical behavior data.

[0110] Among them, the historical behavior data includes behavior events of multiple historical users entering the store within the target area, and the behavior objects corresponding to the behavior events, and the duration of the behavior corresponding to the behavior events, wherein the behavior events include at least one behavior object, and the user behavior knowledge graph includes a master node and at least one slave node, the master node corresponds to the historical users entering the store, the slave node corresponds to the behavior object, and the edge attributes between the master node and the slave nodes, and between the slave nodes are behavior events and edge weights corresponding to the behavior events.

[0111] It should be noted that the target area in the embodiment of the present invention is a marketing area of ​​a store, for example, a product display area, a reception area, a rest area, a product experience area and / or a transaction area and other places related to product marketing.

[0112] Among them, the behavior object is the object corresponding to the user behavior. For example, if the user is in the meeting area and communicating with the salesperson, the behavior corresponds to the salesperson, and the behavior event is communicating with the salesperson. For another example, if the user is inside the vehicle experiencing the vehicle (the interior of the vehicle can also be divided into the front driving seat and the rear seat), the behavior object is the interior of the vehicle, and the behavior event is experiencing the vehicle inside the vehicle. For another example, if the user is at the rear of the vehicle viewing the vehicle, the behavior object is the rear of the vehicle, and the behavior event is viewing the rear of the vehicle.

[0113] It should be noted that in the actual application process, vehicles can be divided into more detailed categories. For example, vehicles can be divided into different behavior objects according to information such as vehicle model, price, and place of origin.

[0114] Among them, in the embodiment of the present invention, the behavioral event also includes the user's consumption behavior. Specifically, the consumption behavior may include the purchased products or services. In this process, the behavior object can be a salesperson or a service person. For example, in the case of vehicle maintenance service, the corresponding behavior object is the vehicle maintenance service person.

[0115] Among them, in the embodiment of the present invention, the user behavior knowledge graph includes a master node and at least one slave node, the master node corresponds to the historical store-entering user, and the slave node corresponds to the behavior object. In the actual application process, the layout of the master and slave nodes can be adopted in two ways, such as Figure 2 As shown in , the first layout mode is that all slave nodes are connected to the master node (forming edges between nodes); Figure 3 As shown, the second layout method is to connect multiple slave nodes in sequence according to the chronological order of the behavior events. This method is conducive to reflecting the relationship between the behavior events.

[0116] The embodiment of the present invention adopts the second layout mode. It should be noted that, Figure 3 As shown, in the actual application process, since a behavior event can involve multiple behavior objects at the same time, there is a situation where a master node is connected to multiple slave nodes, and a slave node is connected to multiple slave nodes. In addition, because the event-based layout method is adopted, in a user's knowledge graph, there will also be a situation where a behavior object appears multiple times, and a behavior object can appear in multiple different behavior events. Among them, the edge attributes between the master node and the slave node, and between the slave nodes are behavior events and the edge weights corresponding to the behavior events.

[0117] Among them, when determining the edge weight corresponding to the behavior event, the total time the historical users have spent in the store and the total time the users have stayed in the target area are obtained, and then the edge weight corresponding to the behavior event is determined based on the proportion of the behavior duration corresponding to the behavior event in the total duration.

[0118] It should be noted that in actual application, one behavior event may correspond to multiple behavior objects. In this case, the edge weights corresponding to the behavior event may be evenly distributed to the multiple behavior objects, or, according to a pre-set distribution ratio, the edge weights corresponding to the behavior event may be distributed to the multiple behavior objects according to the distribution ratio.

[0119] It should be noted that in actual application, after a user has visited the target area of ​​a store, he or she may not have any consumption behavior in the target area, but may consume online after leaving the target area. Alternatively, within a certain period of time, the user visits the target area again and completes the consumption behavior. In this case, the user's consumption behavior can be merged into the historical behavior data of the previous visit to the target area.

[0120] Step 120: Obtain the behavior data of the target user in the target area, and generate a target behavior prediction value of the target user based on the behavior data of the target user and the user behavior knowledge graph.

[0121] Among them, the target user may be a user who is currently in the target area, or a user who has left the target area after visiting the target area. For the user who is currently in the target area, it is only necessary to obtain the behavior data that has occurred at the current time node. For the user who has left the target area, it is necessary to obtain all the behavior data of the target user in the target area, and use the user behavior knowledge graph to analyze the obtained behavior data of the target user, so as to predict the possibility of the target user implementing the target behavior, for example, the target behavior is to purchase a vehicle, or purchase services for a vehicle.

[0122] Specifically, when the target behavior is user consumption behavior, the target behavior prediction value at least includes an expected value representing the possibility of consumption of the target user. In addition to the expected value of the target user's consumption, it may also include specific products or services that the target user may purchase, or it may also include the possible consumption amount of the target user.

[0123] Among them, in the process of generating the target behavior prediction value of the target user based on the target user's behavior data and the user behavior knowledge graph, first, in the target user's behavior data, all the behavior events and corresponding behavior objects of the target user are determined; then, based on all the behavior events and corresponding behavior objects, in the user behavior knowledge graph, the edge weights corresponding to all the behavior events in the target user's behavior data are determined; finally, based on the edge weights corresponding to all the behavior events in the target user's behavior data, the target behavior prediction value of the target user is generated.

[0124] In other words, the behavior data of the target user is decomposed by utilizing the correspondence between the behavior data and the predicted values ​​contained in the user behavior knowledge graph, and all the behavior events and corresponding behavior objects in the behavior data are determined. Then, in the user behavior knowledge graph, the function or weight relationship between each behavior event and the predicted value is matched, and finally, the target behavior prediction value of the target user is generated by utilizing the function or weight relationship between all behavior events and the predicted values.

[0125] Step 130: Generate a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user.

[0126] The recommendation strategy corresponding to the target customer refers to the targeted recommendation of products or services for the target customer. If the target customer is still in the target area, the target user can be guided on site to visit the products in the target area or experience the corresponding services. If the target customer has left the target area, the relevant products or services can be pushed to the target user online, or the recommendation strategy corresponding to the target customer is recorded, and after the target user enters the store, the recommendation strategy corresponding to the target customer is used to receive the target customer.

[0127] It should be noted that the embodiments of the present invention are intended to predict the user's target behavior based on the user's behavior data, and when the store can meet the user's target behavior, promptly recommend related services to the user, or start preparatory work for the related services to ensure that the user can more conveniently obtain information that suits the user, and after starting to receive the service, since the store has prepared the preparatory work for the related services, it can ensure a faster response to user needs.

[0128] The embodiment of the present invention provides a method for recommending vehicle services based on user behavior. The method establishes a user behavior knowledge graph through the historical behavior data of multiple historical store users in the target area, and then uses the user behavior knowledge graph to predict the target behavior prediction value corresponding to the target user's behavior data, thereby generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user. The technical solution of the embodiment of the present invention can generate a recommendation strategy corresponding to the target customer to help sales staff accurately identify users with stronger intentions to implement the target behavior, and adaptively serve users according to the recommendation strategy, thereby improving the overall service quality of the store and increasing the transaction volume of products or services.

[0129] Figure 4 The flowchart of the second embodiment of a method for recommending vehicle services based on user behavior provided by the present invention is shown. The method for recommending vehicle services based on user behavior is executed by a device for recommending vehicle services based on user behavior. Figure 4 As shown, the method for recommending vehicle services based on user behavior includes the following steps:

[0130] Step 410: Obtain historical behavior data, and establish a user behavior knowledge graph based on the historical behavior data.

[0131] Step 420: Based on the user behavior knowledge graph, analyze the behavior objects in the historical behavior data of multiple historical store-entering users, the frequencies corresponding to the behavior objects, and the layout environment in the target area, and determine the behavior objects with abnormal frequencies and the corresponding abnormal types in the historical behavior data of the multiple historical store-entering users.

[0132] It should be noted that the abnormality type includes at least too low frequency and / or too high frequency. The layout environment of the target area refers to the placement of the behavior objects in the target area (behavior objects of non-staff members, such as cars, desks used in the reception area, coffee machines, etc.). For example, the coffee machine is set in a prominent position in the rest area. Among the behavior objects of historical users entering the store, the sofa in the rest area appears more frequently, but the frequency of the coffee machine is significantly lower, indicating that the setting of the coffee machine is unreasonable. In this case, the behavior object with abnormal frequency is the coffee machine, and the abnormality type is too low frequency.

[0133] For another example, in the above scenario, the coffee machine appears frequently, but there are fewer target behavior records of users that include the coffee machine in the behavior objects, indicating that users who have a stronger desire to implement the target behavior use the coffee machine less frequently. In this case, the abnormality type of the coffee machine is too high a frequency, and you can consider removing the coffee machine or moving it to an area with a relatively low location level, where the location level of the area is predetermined. For example, the position opposite the entrance is set as an area with a higher location level, and another example is setting the position away from the product placement area as an area with a lower importance level.

[0134] It should be noted that the user's target behavior record belongs to a behavior event, the behavior object is a salesperson or service personnel, and the behavior event is the purchase of a specific product or service.

[0135] Through the above technical solution, the user behavior knowledge graph and the historical behavior data of historical store users can be used to analyze whether the local behavior objects in the target area are reasonable, thereby providing guidance for transforming the layout in the target area.

[0136] Figure 5 FIG. 1 is a flow chart of a third embodiment of a method for recommending a vehicle service based on user behavior provided by the present invention, wherein the method for recommending a vehicle service based on user behavior is executed by a device for recommending a vehicle service based on user behavior. Figure 5 As shown, the method for recommending vehicle services based on user behavior includes the following steps:

[0137] Step 510: Obtain historical behavior data, and establish a user behavior knowledge graph based on the historical behavior data.

[0138] Step 520: Acquire historical behavior data containing the first behavior object, and determine the correlation between the first behavior object and the target behavior based on the target behavior record in the historical behavior data containing the first behavior object.

[0139] Among them, the correlation between the first behavior object and the target behavior is used to represent the impact of the first behavior on the target behavior. For example, the first behavior object is to check the motor or engine of the vehicle at the front of the vehicle. There are many user target behavior records containing the first behavior object, and it can be determined that the correlation between the first behavior object and the target behavior is high.

[0140] Alternatively, if the behavior object is a vehicle performance report (paper report or electronic report), and there are fewer user target behavior records containing the first behavior object, it can be determined that the correlation between the first behavior object and the target behavior is low.

[0141] It should be noted that the above-mentioned higher and lower are relative concepts. For example, if the influence factor of the first behavior object on the target behavior is greater than 0.4, the correlation is considered to be high, otherwise, the correlation is considered to be low.

[0142] Step 530: Obtain a location level of the first behavior object in the target area, where the location level is used to indicate the importance of the first behavior object in the target area.

[0143] Step 540: Based on the location level of the first behavior object in the target area and the correlation between the first behavior object and the target behavior, generate a first instruction to adjust the location level of the first behavior object, or generate a second instruction to maintain the location level of the first behavior object.

[0144] Among them, the first instruction is used to indicate that the placement position of the first behavior object needs to be adjusted, and gives the corresponding adjustment method of the location level, and the second instruction is used to indicate that the first behavior object can maintain the current placement position to ensure that the location level remains unchanged.

[0145] Among them, on the basis of obtaining the correlation between the first behavior object and the target behavior, combined with the location level of the first behavior object in the target area, it can be obtained that the location level of the behavior object matches the correlation between the first behavior object and the target behavior. Specifically, when the correlation between the first behavior object and the target behavior is low, the target object is adjusted to a position with a lower location level; when the correlation between the first behavior object and the target behavior is high, the target object is adjusted to a position with a higher location level. It should be noted that the above-mentioned first behavior object can be a product or equipment that has always existed in the target area, or it can be a product or equipment that is to be added, and the product or equipment that is to be added is pre-set in a certain area, and based on the location level of the first behavior object in the target area and the correlation between the first behavior object and the target behavior, it is analyzed whether the pre-set area for the new product or equipment is appropriate.

[0146] By applying the above technical solution, it is possible to specifically analyze whether the location of products or equipment in the target area is appropriate, and it is also possible to select a more appropriate location to add new products or equipment when preparing to add new products or equipment.

[0147] It should be noted that, since obtaining the predicted value of the user's target behavior based on the user behavior knowledge graph is a prediction method, in order to improve the accuracy of the prediction, it is necessary to continuously correct the user behavior knowledge graph.

[0148] For example, after the service personnel use the target behavior prediction value obtained by the user behavior knowledge graph, they implement the corresponding recommendation strategy based on the prediction value, that is, push the service information corresponding to the recommendation strategy to the target user.

[0149] In actual application, there may be a variety of situations. For example, if the user accepts the service personnel's recommended strategy and implements the target behavior, the user behavior knowledge graph can be appropriately adjusted to make the user behavior knowledge graph give a higher target behavior prediction value under the same conditions. If the user rejects the service personnel's recommended strategy, the user behavior knowledge graph can be relatively adjusted to reduce the target behavior prediction value given by the user behavior knowledge graph under the same conditions. The embodiment of the present invention obtains feedback information of the target user corresponding to the service information and modifies the user behavior knowledge graph based on the feedback information. In actual application, the user behavior knowledge graph is continuously modified to improve the prediction accuracy of the user behavior knowledge graph.

[0150] The embodiment of the present invention provides a method for recommending vehicle services based on user behavior. The user behavior knowledge graph is established through the historical behavior data of historical store users in the target area. The target behavior prediction value corresponding to the behavior data of the target user is then predicted using the user behavior knowledge graph, thereby generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user. The technical solution of the embodiment of the present invention can generate a recommendation strategy corresponding to the target customer to help sales staff accurately identify users with stronger intentions to implement the target behavior, and adaptively serve users based on the recommendation strategy, thereby improving the overall service quality of the store and increasing the transaction volume of products or services.

[0151] Figure 6 FIG. 1 is a schematic diagram showing a structure of an embodiment of a vehicle service recommendation device based on user behavior provided by an embodiment of the present invention. Figure 6 As shown, the vehicle service recommendation device 600 based on user behavior includes: a knowledge graph establishment module 610, a target behavior prediction module 620, a recommendation strategy generation module 630, a behavior object anomaly analysis module 640, a behavior relevance generation module 650, a location level acquisition module 660, an instruction generation module 670 and a recommendation strategy execution module 680.

[0152] The knowledge graph establishment module 610 is used to obtain historical behavior data and establish a user behavior knowledge graph based on the historical behavior data; the historical behavior data includes behavior events of multiple historical store users in a target area, and behavior objects corresponding to the behavior events, wherein the behavior events include at least one behavior object, and the user behavior knowledge graph includes a master node and at least one slave node, the master node corresponds to the historical store users, the slave node corresponds to the behavior object, and the edge attributes between the master node and the slave nodes, and between the slave nodes are behavior events and edge weights corresponding to the behavior events.

[0153] The target behavior prediction module 620 is used to obtain the behavior data of the target user in the target area, and generate the target behavior prediction value of the target user based on the behavior data of the target user and the user behavior knowledge graph.

[0154] The recommendation strategy generation module 630 is used to generate a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user.

[0155] In an optional manner, the knowledge graph building module 610 is also used to obtain the total duration of the historical store-entering users' visits.

[0156] And, it is used to determine the edge weight corresponding to the behavior event based on the total time of entering the store and the duration of the behavior corresponding to the behavior event.

[0157] In an optional manner, the knowledge graph establishment module 610 is also used to allocate the edge weights corresponding to the behavior event to the edge attributes corresponding to the multiple behavior objects according to the number of the multiple behavior objects if the behavior event includes multiple behavior objects.

[0158] In an optional manner, the behavior object anomaly analysis module 640 is also used to analyze the behavior objects in the historical behavior data of multiple historical users entering the store, the frequencies corresponding to the behavior objects, and the layout environment in the target area based on the user behavior knowledge graph, and determine the behavior objects with abnormal frequencies in the historical behavior data of the multiple historical users entering the store and the corresponding anomaly types.

[0159] In an optional manner, the behavior relevance generation module 650 is used to obtain historical behavior data containing a first behavior object, and determine the relevance between the first behavior object and the target behavior based on the target behavior record in the historical behavior data containing the first behavior object.

[0160] The location level acquisition module 660 is used to acquire the location level of the first behavior object in the target area, and the location level is used to indicate the importance of the behavior object in the target area.

[0161] The instruction generation module 670 is used to generate a first instruction for adjusting the location level of the first behavior object based on the location level of the first behavior object in the target area and the correlation between the first behavior object and the target behavior, or to generate a second instruction for maintaining the location level of the first behavior object.

[0162] In an optional manner, the recommendation strategy execution module 680 is used to push service information corresponding to the recommendation strategy to the target user.

[0163] Obtain feedback information of the target user corresponding to the service information, and modify the user behavior knowledge graph based on the feedback information.

[0164] In an optional manner, the target behavior prediction module 620 is further used to determine all behavior events and corresponding behavior objects of the target user in the behavior data of the target user.

[0165] According to all the behavior events and the corresponding behavior objects, in the user behavior knowledge graph, the edge weights corresponding to all the behavior events in the behavior data of the target user are determined.

[0166] A target behavior prediction value of the target user is generated according to the edge weights corresponding to all behavior events in the behavior data of the target user.

[0167] The embodiment of the present invention provides a vehicle service recommendation device based on user behavior. The knowledge graph establishment module 610 establishes a user behavior knowledge graph. Then, the target behavior prediction module 620 uses the user behavior knowledge graph to predict the target behavior prediction value corresponding to the target user's behavior data. Finally, the recommendation strategy generation module 630 generates a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user. The technical solution of the embodiment of the present invention can generate a recommendation strategy corresponding to the target customer to help sales staff accurately identify users with stronger intentions to implement target behaviors, and adaptively serve users according to the recommendation strategy, thereby improving the overall service quality of the store and increasing the transaction volume of products or services.

[0168] Figure 7 A schematic diagram of the structure of an embodiment of a vehicle service recommendation device based on user behavior provided by an embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the vehicle service recommendation device based on user behavior.

[0169] like Figure 7 As shown, the device for recommending vehicle services based on user behavior may include: a processor 702 , a communications interface 704 , a memory 706 , and a communication bus 708 .

[0170] The processor 702, the communication interface 704, and the memory 706 communicate with each other via the communication bus 708. The communication interface 704 is used to communicate with other devices such as a client or other server network elements. The processor 702 is used to execute the program 710, which can specifically execute the relevant steps in the above-mentioned method for recommending vehicle services based on user behavior.

[0171] Specifically, the program 710 may include program code including computer executable instructions.

[0172] The processor 702 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the recommendation device for vehicle services based on user behavior may be processors of the same type, such as one or more CPUs; or may be processors of different types, such as one or more CPUs and one or more ASICs.

[0173] The memory 706 is used to store the program 710. The memory 706 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0174] Program 710 may be specifically called by processor 702 to enable the vehicle service recommendation device based on user behavior to perform the following operations:

[0175] Acquire historical behavior data, and establish a user behavior knowledge graph based on the historical behavior data; the historical behavior data includes multiple behavior events of historical store users in a target area, and behavior objects corresponding to the behavior events, wherein the behavior events include at least one behavior object, and the user behavior knowledge graph includes a master node and at least one slave node, the master node corresponds to the historical store users, the slave node corresponds to the behavior object, and the edge attributes between the master node and the slave nodes, and between the slave nodes and the slave nodes are behavior events and edge weights corresponding to the behavior events.

[0176] The behavior data of the target user in the target area is obtained, and based on the behavior data of the target user and the user behavior knowledge graph, a target behavior prediction value of the target user is generated.

[0177] Based on the target behavior prediction value of the target user, a recommendation strategy corresponding to the target customer is generated.

[0178] In an optional manner, the historical behavior data also includes the duration of the behavior corresponding to the behavior event.

[0179] In the step of establishing a user behavior knowledge graph based on the historical behavior data, it also includes:

[0180] Get the total in-store time of the historical in-store users.

[0181] Based on the total duration of entering the store and the duration of the behavior corresponding to the behavior event, the edge weight corresponding to the behavior event is determined.

[0182] In an optional manner, the step of establishing a user behavior knowledge graph based on the historical behavior data further includes:

[0183] If the behavior event includes a plurality of behavior objects, the edge weight corresponding to the behavior event is allocated to the edge attributes corresponding to the plurality of behavior objects according to the number of the plurality of behavior objects.

[0184] In an optional manner, after the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes:

[0185] Based on the user behavior knowledge graph, the behavior objects in the historical behavior data of multiple historical users entering the store, the frequencies corresponding to the behavior objects, and the layout environment in the target area are analyzed to determine the behavior objects with abnormal frequencies and the corresponding abnormal types in the historical behavior data of the multiple historical users entering the store.

[0186] In an optional manner, after the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes:

[0187] Historical behavior data including a first behavior object is obtained, and a correlation between the first behavior object and the target behavior is determined based on the target behavior record in the historical behavior data including the first behavior object.

[0188] In an optional manner, it also includes: obtaining a location level of the first behavior object in the target area, wherein the location level is used to indicate the importance of the behavior object in the target area.

[0189] Based on the location level of the first behavior object in the target area and the correlation between the first behavior object and the target behavior, a first instruction for adjusting the location level of the first behavior object is generated, or a second instruction for maintaining the location level of the first behavior object is generated.

[0190] In an optional manner, after the step of generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user, the method further includes:

[0191] Pushing service information corresponding to the recommendation strategy to the target user.

[0192] Obtain feedback information of the target user corresponding to the service information, and modify the user behavior knowledge graph based on the feedback information.

[0193] In an optional manner, the step of generating a target behavior prediction value of the target user based on the target user's behavior data and the user behavior knowledge graph includes:

[0194] In the behavior data of the target user, all behavior events and corresponding behavior objects of the target user are determined.

[0195] According to all the behavior events and the corresponding behavior objects, in the user behavior knowledge graph, the edge weights corresponding to all the behavior events in the behavior data of the target user are determined.

[0196] A target behavior prediction value of the target user is generated according to the edge weights corresponding to all behavior events in the behavior data of the target user.

[0197] The embodiment of the present invention provides a control device for a vehicle steering system, wherein the memory 706 of the control device for the vehicle steering system is used to store a program 710, which can be specifically called by the processor 702, and a user behavior knowledge graph is established through historical behavior data of historical store users in a target area, and then the user behavior knowledge graph is used to predict the target behavior prediction value corresponding to the target user's behavior data, thereby generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user. The technical solution of the embodiment of the present invention can generate a recommendation strategy corresponding to the target customer to help sales staff accurately identify users with stronger intentions to implement the target behavior, and adaptively serve users based on the recommendation strategy, thereby improving the overall service quality of the store and increasing the transaction volume of products or services.

[0198] An embodiment of the present invention provides a computer-readable storage medium, which stores at least one executable instruction. When the executable instruction is executed on a device / apparatus for recommending vehicle services based on user behavior, the device / apparatus for recommending vehicle services based on user behavior executes the method for recommending vehicle services based on user behavior in any of the above-mentioned method embodiments.

[0199] The executable instructions may be specifically used to enable the vehicle service recommendation device based on user behavior to perform the following operations:

[0200] Acquire historical behavior data, and establish a user behavior knowledge graph based on the historical behavior data; the historical behavior data includes multiple behavior events of historical store users in a target area, and behavior objects corresponding to the behavior events, wherein the behavior events include at least one behavior object, and the user behavior knowledge graph includes a master node and at least one slave node, the master node corresponds to the historical store users, the slave node corresponds to the behavior object, and the edge attributes between the master node and the slave nodes, and between the slave nodes and the slave nodes are behavior events and edge weights corresponding to the behavior events.

[0201] The behavior data of the target user in the target area is obtained, and based on the behavior data of the target user and the user behavior knowledge graph, a target behavior prediction value of the target user is generated.

[0202] Based on the target behavior prediction value of the target user, a recommendation strategy corresponding to the target customer is generated.

[0203] In an optional manner, the historical behavior data also includes the duration of the behavior corresponding to the behavior event.

[0204] In the step of establishing a user behavior knowledge graph based on the historical behavior data, it also includes:

[0205] Get the total in-store time of the historical in-store users.

[0206] Based on the total duration of entering the store and the duration of the behavior corresponding to the behavior event, the edge weight corresponding to the behavior event is determined.

[0207] In an optional manner, the step of establishing a user behavior knowledge graph based on the historical behavior data further includes:

[0208] If the behavior event includes a plurality of behavior objects, the edge weight corresponding to the behavior event is allocated to the edge attributes corresponding to the plurality of behavior objects according to the number of the plurality of behavior objects.

[0209] In an optional manner, after the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes:

[0210] Based on the user behavior knowledge graph, the behavior objects in the historical behavior data of multiple historical users entering the store, the frequencies corresponding to the behavior objects, and the layout environment in the target area are analyzed to determine the behavior objects with abnormal frequencies and the corresponding abnormal types in the historical behavior data of the multiple historical users entering the store.

[0211] In an optional manner, after the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes:

[0212] Historical behavior data including a first behavior object is obtained, and a correlation between the first behavior object and the target behavior is determined based on the target behavior record in the historical behavior data including the first behavior object.

[0213] In an optional manner, it also includes: obtaining a location level of the first behavior object in the target area, wherein the location level is used to indicate the importance of the behavior object in the target area.

[0214] Based on the location level of the first behavior object in the target area and the correlation between the first behavior object and the target behavior, a first instruction for adjusting the location level of the first behavior object is generated, or a second instruction for maintaining the location level of the first behavior object is generated.

[0215] In an optional manner, after the step of generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user, the method further includes:

[0216] Pushing service information corresponding to the recommendation strategy to the target user.

[0217] Obtain feedback information of the target user corresponding to the service information, and modify the user behavior knowledge graph based on the feedback information.

[0218] In an optional manner, the step of generating a target behavior prediction value of the target user based on the target user's behavior data and the user behavior knowledge graph includes:

[0219] In the behavior data of the target user, all behavior events and corresponding behavior objects of the target user are determined.

[0220] According to all the behavior events and the corresponding behavior objects, in the user behavior knowledge graph, the edge weights corresponding to all the behavior events in the behavior data of the target user are determined.

[0221] A target behavior prediction value of the target user is generated according to the edge weights corresponding to all behavior events in the behavior data of the target user.

[0222] The embodiment of the present invention provides a computer-readable storage medium. When the executable instructions stored in the computer-readable storage medium are executed, the user behavior knowledge graph is established through the historical behavior data of historical store users in the target area, and then the target behavior prediction value corresponding to the target user's behavior data is predicted using the user behavior knowledge graph, so as to generate a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user. The technical solution of the embodiment of the present invention can generate a recommendation strategy corresponding to the target customer to help sales staff accurately identify users with stronger intentions to implement the target behavior, and adaptively serve users according to the recommendation strategy, thereby improving the overall service quality of the store and increasing the transaction volume of products or services.

[0223] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system or other device. In addition, the embodiments of the present invention are not directed to any particular programming language.

[0224] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. Similarly, in order to simplify the present invention and help understand one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, the various features of the embodiments of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. Wherein, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, wherein each claim itself is a separate embodiment of the present invention.

[0225] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and further may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive.

[0226] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be understood as limitations on the order of execution.

Claims

1. A method for recommending vehicle services based on user behavior, characterized in that: The method comprises: Acquire historical behavior data, and establish a user behavior knowledge graph based on the historical behavior data; the historical behavior data includes multiple behavior events of historical store users in a target area, and behavior objects corresponding to the behavior events, wherein the behavior event includes at least one behavior object, the user behavior knowledge graph includes a master node and at least one slave node, the master node corresponds to the historical store user, the slave node corresponds to the behavior object, and the edge attributes between the master node and the slave node, and between the slave nodes and the slave nodes are the behavior events and the edge weights corresponding to the behavior events; Obtaining behavior data of a target user in a target area, and generating a target behavior prediction value of the target user based on the behavior data of the target user and the user behavior knowledge graph; Based on the target behavior prediction value of the target user, a recommendation strategy corresponding to the target customer is generated.

2. The method for recommending vehicle services based on user behavior according to claim 1, characterized in that: The historical behavior data also includes the duration of the behavior corresponding to the behavior event; In the step of establishing a user behavior knowledge graph based on the historical behavior data, it also includes: Obtain the total store-entry time of the historical store-entering users; Based on the total duration of entering the store and the duration of the behavior corresponding to the behavior event, the edge weight corresponding to the behavior event is determined.

3. The method for recommending vehicle services based on user behavior according to claim 1 or 2, characterized in that: In the step of establishing a user behavior knowledge graph based on the historical behavior data, it also includes: If the behavior event includes a plurality of behavior objects, the edge weight corresponding to the behavior event is allocated to the edge attributes corresponding to the plurality of behavior objects according to the number of the plurality of behavior objects.

4. The method for recommending vehicle services based on user behavior according to claim 1, characterized in that: After the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes: Based on the user behavior knowledge graph, the behavior objects in the historical behavior data of multiple historical users entering the store, the frequencies corresponding to the behavior objects, and the layout environment in the target area are analyzed to determine the behavior objects with abnormal frequencies and the corresponding abnormal types in the historical behavior data of the multiple historical users entering the store.

5. The method for recommending vehicle services based on user behavior according to claim 1, characterized in that: After the step of establishing a user behavior knowledge graph based on the historical behavior data, the method further includes: Historical behavior data including a first behavior object is obtained, and a correlation between the first behavior object and the target behavior is determined based on the target behavior record in the historical behavior data including the first behavior object.

6. The method for recommending vehicle services based on user behavior according to claim 5, characterized in that: Also includes: Acquire a location level of the first behavior object in the target area, where the location level is used to indicate the importance of the first behavior object in the target area; Based on the location level of the first behavior object in the target area and the correlation between the first behavior object and the target behavior, a first instruction for adjusting the location level of the first behavior object is generated, or a second instruction for maintaining the location level of the first behavior object is generated.

7. The method for recommending vehicle services based on user behavior according to claim 1, characterized in that: After the step of generating a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user, the method further includes: Pushing service information corresponding to the recommendation strategy to the target user; Obtain feedback information of the target user corresponding to the service information, and modify the user behavior knowledge graph based on the feedback information.

8. The method for recommending vehicle services based on user behavior according to any one of claims 1-2 or 4-7, characterized in that: The step of generating a target behavior prediction value of the target user based on the target user's behavior data and the user behavior knowledge graph includes: In the behavior data of the target user, determining all behavior events and corresponding behavior objects of the target user; According to all the behavior events and the corresponding behavior objects, in the user behavior knowledge graph, determining the edge weights corresponding to all the behavior events in the behavior data of the target user; A target behavior prediction value of the target user is generated according to the edge weights corresponding to all behavior events in the behavior data of the target user.

9. A vehicle service recommendation device based on user behavior, characterized in that: The vehicle service recommendation device based on user behavior includes: A knowledge graph establishment module, used to obtain historical behavior data, and establish a user behavior knowledge graph based on the historical behavior data; the historical behavior data includes multiple behavior events of historical store users in a target area, and behavior objects corresponding to the behavior events, wherein the behavior event includes at least one behavior object, the user behavior knowledge graph includes a master node and at least one slave node, the master node corresponds to the historical store user, the slave node corresponds to the behavior object, and the edge attributes between the master node and the slave node, and between the slave nodes are behavior events and edge weights corresponding to the behavior events; A target behavior prediction module is used to obtain the behavior data of the target user in the target area, and generate a target behavior prediction value of the target user based on the behavior data of the target user and the user behavior knowledge graph; The recommendation strategy generation module is used to generate a recommendation strategy corresponding to the target customer based on the target behavior prediction value of the target user.

10. A vehicle service recommendation device based on user behavior, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform the operation of the method for recommending vehicle services based on user behavior as described in any one of claims 1-8.

Images