An intelligent customer service method based on GIS geographic information platform

Abstract

The present application discloses an intelligent customer service method based on a GIS geographic information platform, which belongs to the field of computer technology. When a target object passes through security at a target airport, and the time difference between the remaining boarding time and the time required for boarding is greater than or equal to a preset time difference, the target object's store preference information and the store information of multiple candidate stores at the target airport are obtained. Based on the target object's store preference information, the target object's current location, the location of the target boarding gate, and the store information of the multiple candidate stores, multiple target stores are determined from the multiple candidate stores. Based on the target object's current location, the target object's object information, the location of the target boarding gate, the time difference, and the store information of the multiple target stores, a boarding path is planned for the target object to obtain the target boarding path of the target object.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an intelligent customer service method based on a GIS geographic information platform. Background Art

[0002] With the increasing popularity of air travel, airports have become important hubs for people to travel. Large airports are often equipped with a large number of shops, allowing passengers to choose different shops according to their needs before and after passing through security. In particular, shops after passing through security have become a popular choice for passengers to while away the time while waiting for their flight.

[0003] Since large airports are large in area and complex in structure, with numerous boarding paths, it is difficult for users to find the stores they are interested in, resulting in a waste of store resources within the airport.

[0004] Therefore, how to recommend boarding routes that pass by stores of interest to passengers while ensuring that they do not miss their flights has become a hot topic of research. Summary of the Invention

[0005] The embodiment of the present application provides an intelligent customer service method based on a GIS geographic information platform, which can recommend boarding routes that pass through stores of interest to passengers while ensuring that the passengers do not miss their flights. The technical solution is as follows.

[0006] In one aspect, an intelligent customer service method based on a GIS geographic information platform is provided, the method comprising:

[0007] When the target subject passes security at the target airport, determining the target subject's remaining boarding time and required boarding time, where the remaining boarding time is the time between the boarding time and the current time, and the required boarding time is the time required for the target subject to reach the target boarding gate from the current location, where the target boarding gate is the boarding gate for the flight the target subject will take;

[0008] If the difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference threshold, obtaining store preference information of the target object and store information of multiple candidate stores at the target airport, where the candidate stores are accessible after passing security check, and the store information includes store descriptions and store locations;

[0009] determining a plurality of target stores from the plurality of candidate stores based on the store preference information of the target object, the current location of the target object, the location of the target boarding gate, and the store information of the plurality of candidate stores, wherein the store locations and the location of the target boarding gate are determined based on a geographic information platform;

[0010] Based on the current location of the target object, the object information of the target object, the location of the target boarding gate, the remaining boarding time, and the store information of the multiple target stores, a boarding path is planned for the target object to obtain a target boarding path for the target object, where the target boarding path passes through at least one target store among the multiple target stores to reach the target boarding gate.

[0011] In one aspect, an intelligent customer service device based on a GIS geographic information platform is provided, the device comprising:

[0012] a duration determination module, configured to determine, when the target subject passes security at the target airport, the remaining boarding time and the required boarding time of the target subject, where the remaining boarding time is the time between the boarding time and the current time, and the required boarding time is the time required for the target subject to reach a target boarding gate from the current location, where the target boarding gate is the boarding gate of the flight that the target subject will take;

[0013] an information acquisition module, configured to acquire, when the difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference threshold, store preference information of the target subject and store information of a plurality of candidate stores at the target airport, wherein the candidate stores are stores that can be reached after passing security inspection, and the store information includes a store description and a store location;

[0014] a target store determining module, configured to determine a plurality of target stores from the plurality of candidate stores based on the store preference information of the target object, the current location of the target object, the location of the target boarding gate, and the store information of the plurality of candidate stores, wherein the store locations and the location of the target boarding gate are determined based on a geographic information platform;

[0015] A path planning module is used to plan a boarding path for the target object based on the current location of the target object, the object information of the target object, the location of the target boarding gate, the remaining boarding time, and the store information of the multiple target stores, so as to obtain a target boarding path for the target object, where the target boarding path passes through at least one target store among the multiple target stores to reach the target boarding gate.

[0016] In one possible embodiment, the duration determination module is used to determine the remaining boarding time of the target object when the target object passes the security check of the target airport; obtain a first moving speed, a second moving speed and a reference moving speed of the target airport, wherein the first moving speed is the average moving speed of the target object at multiple airports, the second moving speed is the average moving speed of the target object after entering the target airport, and the reference moving speed is a moving speed determined based on the crowd density in the target airport; determine the target moving speed of the target object based on the first moving speed, the second moving speed and the reference moving speed of the target airport; divide the path length of the reference boarding path between the position of the target boarding gate and the current position of the target object by the target moving speed to obtain the time required for boarding, wherein the reference boarding path is a boarding path whose selected number of times meets the preset number condition.

[0017] In one possible embodiment, the information acquisition module is configured to, when the difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference threshold, acquire historical store residence information, historical store consumption information, real-time store residence information, and real-time store consumption information of the target subject at multiple airports, where the real-time store residence information and real-time store consumption information are the store residence information and store consumption information corresponding to the target subject after entering the target airport and before passing through security check; determine the store preference information of the target subject based on the historical store residence information, historical store consumption information, real-time store residence information, and real-time store consumption information; acquire an indoor map of the target airport from the geographic information platform; determine the multiple candidate stores from the indoor map and acquire the store location of each of the candidate stores; query a store description database of the target airport based on the multiple candidate stores to obtain a store description of each of the candidate stores; and combine the store description and store location of each of the candidate stores to obtain store information of each of the candidate stores.

[0018] In a possible embodiment, the information acquisition module is used to determine multiple first historical stores of interest to the target object based on the historical store residence information of the target object; determine multiple second historical stores of interest to the target object based on the historical store consumption information of the target object; determine multiple first real-time stores of interest to the target object based on the real-time store residence information of the target object; determine multiple second real-time stores of interest to the target object based on the real-time store consumption information of the target object; and determine the store preference information of the target object based on the multiple first historical stores of interest, the multiple second historical stores of interest, the multiple first real-time stores of interest, and the multiple second real-time stores of interest.

[0019] In one possible embodiment, the information acquisition module is used to determine the target object's store types and brands of interest based on the multiple first historical stores of interest, the multiple second historical stores of interest, the multiple first real-time stores of interest, and the multiple second real-time stores of interest; and determine the target object's store preference information based on the target object's store types and brands of interest.

[0020] In one possible embodiment, the target store determination module is used to determine a plurality of reference stores from the multiple candidate stores based on the store preference information of the target object and the store descriptions in the store information of the multiple candidate stores, where the reference stores are candidate stores in which the target object has a high degree of interest; and to determine the multiple target stores from the multiple reference stores based on the current location of the target object, the location of the target boarding gate, and the store locations in the store information of the multiple reference stores.

[0021] In one possible embodiment, the target store determination module is used to determine multiple candidate boarding paths based on the current position of the target object and the position of the target boarding gate, where the candidate boarding paths are paths moving from the current position to the target boarding gate; determine a target boarding area based on the multiple candidate boarding paths, where the target boarding area is an area covering the multiple boarding paths; and determine the multiple target stores belonging to the target boarding area from the multiple reference stores based on the store locations in the store information of the multiple reference stores.

[0022] In one possible embodiment, the path planning module is used to determine the estimated residence time of the target object in each of the target stores based on the store descriptions in the store information of the multiple target stores and the object information of the target object; create multiple location nodes based on the current location, the location of the target boarding gate and the store location in the store information of each of the target stores; assign the estimated residence time of each of the target stores to the corresponding location node, and determine the movement time between each two location nodes in the multiple location nodes; based on the movement time between each two location nodes in the multiple location nodes, add a connection between each two location nodes in the multiple location nodes after assignment to obtain a target graph network; with the location node corresponding to the current location as the starting point, the location node corresponding to the location of the target boarding gate as the end point, and the remaining boarding time as a constraint, perform linear programming on the target graph network to obtain the target boarding path of the target object.

[0023] In a possible implementation, the path planning module is used to perform feature extraction on the store description of each of the target stores to obtain store description features of each of the stores, wherein the store description includes store type, store brand, store product information and store area, and the store description features include store type sub-features, store brand sub-features, store product sub-features and store area sub-features; perform feature extraction on the object information of the target object to obtain object features of the target object, wherein the object information includes object type, store preference information and object status; and determine the estimated residence time of the target object in each of the target stores based on the store description features of each of the target stores and the object features of the target object.

[0024] In one possible implementation, the path planning module is used to determine multiple initial boarding paths in the target graph network with the location node corresponding to the current location as the starting point and the location node corresponding to the location of the target boarding gate as the end point; add the sum of the movement time corresponding to each of the initial boarding paths and the sum of the estimated residence time to obtain the total boarding time of each of the initial boarding paths; use the remaining boarding time to constrain the total boarding time of each of the initial boarding paths, perform multiple rounds of iterative adjustments on the location nodes in each of the initial boarding paths, and obtain the target boarding path of the target object.

[0025] On the one hand, a computer device is provided, which includes one or more processors and one or more memories, wherein at least one computer program is stored in the one or more memories, and the computer program is loaded and executed by the one or more processors to implement the intelligent customer service method based on the GIS geographic information platform.

[0026] On the one hand, a computer-readable storage medium is provided, in which at least one computer program is stored. The computer program is loaded and executed by a processor to implement the intelligent customer service method based on the GIS geographic information platform.

[0027] On the one hand, a computer program product or computer program is provided, which includes a program code, which is stored in a computer-readable storage medium. The processor of a computer device reads the program code from the computer-readable storage medium, and the processor executes the program code, so that the computer device executes the above-mentioned intelligent customer service method based on the GIS geographic information platform.

[0028] Through the technical solution provided by the embodiments of the present application, when a target subject passes security at a target airport, the target subject's remaining boarding time and the required boarding time are determined. If the difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference, the target subject's store preference information and the store information of multiple candidate stores at the target airport are obtained. The candidate stores are stores that can be reached after passing security. Based on the target subject's store preference information, the target subject's current location, the location of the target boarding gate, and the store information of the multiple candidate stores, multiple target stores are determined from the multiple candidate stores. The store locations and the location of the target boarding gate are determined based on a geographic information platform. Based on the target subject's current location, the target subject's object information, the location of the target boarding gate, the time difference, and the store information of the multiple target stores, a boarding path is planned for the target subject to obtain the target boarding path. The target boarding path is displayed to the target subject, providing both boarding guidance for the target subject and guidance to the target stores, thereby meeting the target subject's needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0030] Figure 1 This is a schematic diagram of an implementation environment of an intelligent customer service method based on a GIS geographic information platform provided in an embodiment of the present application;

[0031] Figure 2 This is a flow chart of an intelligent customer service method based on a GIS geographic information platform provided in an embodiment of the present application;

[0032] Figure 3 This is a flowchart of another intelligent customer service method based on the GIS geographic information platform provided in an embodiment of the present application;

[0033] Figure 4 This is a schematic diagram of the structure of an intelligent customer service device based on a GIS geographic information platform provided in an embodiment of the present application;

[0034] Figure 5 This is a structural diagram of a server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to make the objectives, technical solutions and advantages of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.

[0036] In this application, the terms "first", "second", etc. are used to distinguish identical or similar items with substantially the same effects and functions. It should be understood that there is no logical or temporal dependency between "first", "second", and "nth", nor is there any limitation on the quantity and execution order.

[0037] A geographic information system (GIS) is a system that uses computer technology to collect, store, manage, analyze, and display spatial data on the Earth's surface. It encompasses not only geographic spatial data but also related attribute data. GIS functions include, but are not limited to, the storage, query, analysis, and visualization of spatial data, as well as providing users with services such as positioning, navigation, and route planning.

[0038] Path planning: refers to the method of finding the optimal or satisfactory path from the starting point to the end point under given environment and conditions.

[0039] Artificial Intelligence (AI) refers to the theories, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, to perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that seeks to understand the essence of intelligence and produce new intelligent machines that can respond in a manner similar to human intelligence. AI also involves studying the design principles and implementation methods of various intelligent machines, enabling them to possess the capabilities of perception, reasoning, and decision-making.

[0040] Artificial intelligence (AI) technology is a comprehensive discipline encompassing a wide range of fields, encompassing both hardware and software technologies. Foundational AI technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0041] Machine learning (ML) is a multidisciplinary field that encompasses probability theory, statistics, approximation theory, convex analysis, and algorithmic complexity theory. It specifically studies how computers can simulate or implement human learning behaviors to acquire new knowledge or skills and reorganize existing knowledge sub-models to continuously improve their performance. Machine learning is at the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications span all areas of AI. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and learning through demonstration.

[0042] Graph networks are a collection of functions organized as a graph in a topological space, used for relational reasoning. Graph networks are composed of graph network blocks (GN blocks), have a flexible topological structure, and can be specialized into various connectionist models, such as feedforward neural networks and recurrent neural networks.

[0043] Normalization: Mapping sequences of numbers with different value ranges to the interval (0, 1) facilitates data processing. In some cases, the normalized values ​​can be directly implemented as probabilities.

[0044] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, storage, and display, etc.), and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. In particular, the current location, object information, and store preference signals of the target object involved in the embodiments of this application are all obtained and used with the permission of the target object.

[0045] Figure 1 This is a schematic diagram of the implementation environment of an intelligent customer service method based on a GIS geographic information platform provided in an embodiment of the present application, see Figure 1 , the implementation environment may include a terminal 110 and a server 140.

[0046] Terminal 110 is connected to server 140 via a wireless network or a wired network. Optionally, terminal 110 is a smart phone, a smart watch, or smart glasses, but is not limited thereto. Terminal 1101 is a mobile terminal used by a user, and has installed and run an application that supports intelligent customer service based on a GIS geographic information platform.

[0047] Server 140 is an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. Server 140 can provide backend services for applications running on terminal 110.

[0048] The following describes the intelligent customer service method based on the GIS geographic information platform provided in the embodiment of the present application. Figure 2 This is a flowchart of an intelligent customer service method based on a GIS geographic information platform provided in an embodiment of the present application. Figure 2 Taking the execution subject as a server as an example, the method includes the following steps.

[0049] 201. When the target object passes the security check at the target airport, the server determines the target object's remaining boarding time and the time required for boarding, where the remaining boarding time is the time between the boarding time and the current time, and the time required for boarding is the time required for the target object to reach the target boarding gate from the current location, where the target boarding gate is the boarding gate of the flight that the target object will take.

[0050] Among them, the target object is a user who is about to take a flight, the target airport is a large airport, and the target object passing the security check means that the target object has completed the ticket checking and security check process and entered the waiting area of ​​the target airport. The time required for boarding is the personalized boarding time required for the target object, and the time required for different objects to reach the target boarding gate from the same security checkpoint may be different. The current location of the target object is determined by the target object's mobile terminal, and the current location refers to the location within the target airport. In addition, the boarding time and the target boarding gate may change while the target object is waiting, so the remaining boarding time and the time required for boarding may also change. In the process of using the technical solution provided by the embodiment of the present application, in response to changes in the boarding time and / or the target boarding gate, the technical solution provided by the embodiment of the present application is re-executed.

[0051] 202. When the time difference between the remaining boarding time and the time required for boarding is greater than or equal to a preset time difference threshold, the server obtains the store preference information of the target object and the store information of multiple candidate stores at the target airport. The candidate stores are stores that can be reached after passing the security check, and the store information includes a store description and a store location.

[0052] Among them, if the duration difference is greater than or equal to the preset duration difference threshold, it means that the target object has ample time to board the plane, and the target object's boarding path passing through the store will not affect the target object's normal boarding. The preset duration difference threshold is set by technical personnel based on actual conditions, or determined based on the airport information of the target airport, and this embodiment of the application does not limit this. The store description of the candidate store is used to describe the store situation of the candidate store, and the store location refers to the location of the candidate store in the target airport.

[0053] 203. The server determines multiple target stores from the multiple candidate stores based on the store preference information of the target object, the current location of the target object, the location of the target boarding gate, and the store information of the multiple candidate stores. The store locations and the location of the target boarding gate are determined based on a geographic information platform.

[0054] The target subject's store preference information reflects the target subject's store preferences. A target store is a candidate store that does not affect the target subject's boarding and that the target subject is interested in. The geographic information platform stores a large amount of geographic information, including both indoor and outdoor geographic information. The aforementioned store locations and target boarding gate locations are examples of indoor geographic information.

[0055] 204. The server plans a boarding path for the target object based on the current location of the target object, the object information of the target object, the location of the target boarding gate, the time difference, and the store information of the multiple target stores, and obtains a target boarding path for the target object. The target boarding path passes through at least one target store among the multiple target stores to reach the target boarding gate.

[0056] The boarding path planning process involves finding a target boarding path recommended to the target passenger. This path is a path that does not affect the target passenger's normal boarding and passes by at least one target store, thus achieving both boarding guidance and target store recommendations.

[0057] Through the technical solution provided by the embodiments of the present application, when a target subject passes security at a target airport, the target subject's remaining boarding time and the required boarding time are determined. If the difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference, the target subject's store preference information and the store information of multiple candidate stores at the target airport are obtained. The candidate stores are stores that can be reached after passing security. Based on the target subject's store preference information, the target subject's current location, the location of the target boarding gate, and the store information of the multiple candidate stores, multiple target stores are determined from the multiple candidate stores. The store locations and the location of the target boarding gate are determined based on a geographic information platform. Based on the target subject's current location, the target subject's object information, the location of the target boarding gate, the time difference, and the store information of the multiple target stores, a boarding path is planned for the target subject to obtain the target boarding path. The target boarding path is displayed to the target subject, providing both boarding guidance for the target subject and guidance to the target stores, thereby meeting the target subject's needs.

[0058] The above steps 201-204 are a brief introduction to the intelligent customer service method based on the GIS geographic information platform provided by the embodiment of the present application. The following will combine some examples to more clearly illustrate the intelligent customer service method based on the GIS geographic information platform provided by the embodiment of the present application. Figure 3 Taking the execution subject as a server as an example, the method includes the following steps.

[0059] 301. When the target object passes the security check at the target airport, the server determines the target object's remaining boarding time and the required boarding time, where the remaining boarding time is the time between the boarding time and the current time, and the required boarding time is the time required for the target object to reach the target boarding gate from the current location, where the target boarding gate is the boarding gate for the flight that the target object will take.

[0060] The target subject is a user about to board a flight, and the target airport is a large airport. Passing security means that the target subject has completed the ticket check and security check processes and entered the waiting area of ​​the target airport. The boarding time is personalized for the target subject, and the time required for different subjects to reach the target boarding gate from the same security checkpoint may vary. The target subject's current location is determined by the target subject's mobile terminal, and this current location refers to the target subject's location within the target airport. Furthermore, the boarding time and target boarding gate may change while the target subject is waiting, so the remaining boarding time and the required boarding time may also change. During the use of the technical solutions provided in the embodiments of the present application, the technical solutions provided in the embodiments of the present application are re-executed in response to changes in the boarding time and / or target boarding gate. After the target subject passes security at the target airport, the target airport sends security check pass information to a server. Upon receiving the security check pass information, the server determines that the target subject has passed security at the target airport.

[0061] In one possible implementation, when a target object passes security at a target airport, the server determines the remaining boarding time for the target object. The server obtains a first moving speed, a second moving speed, and a reference moving speed for the target airport. The first moving speed is the average moving speed of the target object at multiple airports, the second moving speed is the average moving speed of the target object after entering the target airport, and the reference moving speed is a moving speed determined based on the passenger density within the target airport. The server determines a target moving speed for the target object based on the first moving speed, the second moving speed, and the reference moving speed of the target airport. The server divides the length of a reference boarding path between the target gate location and the current location of the target object by the target moving speed to obtain the time required for boarding. The reference boarding path is a boarding path that has been selected a number of times that meets a preset number of conditions.

[0062] The multiple airports include the target airport, and the first speed can reflect the target subject's usual speed at the airport. Of course, if this is the target subject's first flight, the first speed does not exist. In this case, the first preset speed can be directly determined as the target subject's first speed. This first preset speed is the average speed at the target airport, which ensures the implementation of the technical solution. The second speed can reflect the target subject's level of anxiety. The greater the second speed, the more anxious the target subject is to pass security, and can reflect the target subject's actual speed at the target airport. Because the passenger density at the target airport can affect the target subject's speed within the target airport, using a reference speed can help determine a more accurate target speed. There may be multiple boarding paths from the target subject's current location to the target gate. The reference boarding path is the boarding path among these multiple paths that has been selected a number of times that meets the preset number of conditions, that is, the boarding path that has been selected a number of times greater than or equal to the preset number. This path can be considered a regular boarding path, and using the path length of the reference boarding path to determine the boarding time is more representative.

[0063] In this embodiment, when a target subject passes security at a target airport, the remaining boarding time for the target subject is determined. The target subject's target speed is determined using the target subject's first and second speeds, along with a reference speed at the target airport. The target speed is then divided by the length of the reference boarding path to determine the remaining boarding time, resulting in a highly accurate estimate of the remaining boarding time.

[0064] In order to explain the above embodiment more clearly, the above embodiment will be described in several parts below.

[0065] Part 1: When the target object passes the security check at the target airport, the server determines the remaining boarding time of the target object.

[0066] In one possible implementation, when the target subject passes security at the target airport, the server performs a query based on the target subject's object identifier to obtain the flight number of the flight the target subject will board. The server also performs a query based on the flight number of the flight the target subject will board to obtain the target subject's remaining boarding time.

[0067] In the second part, the server obtains the first moving speed, the second moving speed of the target object and the reference moving speed of the target airport.

[0068] In one possible implementation, the server determines multiple reference airports from the multiple airports based on the object identifier of the target object. The multiple reference airports are airports that the target object has visited. Based on the object identifier of the target object, the server obtains the average moving speed of the target object at each reference airport. The server performs a weighted fusion of the average moving speeds of the target object at each reference airport to obtain a first moving speed of the target object. The server divides the moving distance of the target object after entering the airport by the moving duration to obtain a second moving speed of the target object. The server obtains the current crowd density at the target airport, inputs the crowd density into a reference moving speed determination model, and processes the crowd density using the reference moving speed determination model to obtain the reference moving speed.

[0069] Among them, the weight of weighted fusion is associated with the airport type of the reference airport. Since the target airport is a large airport, the weight corresponding to the airport type of a large airport is higher than the weight of the airport type of a medium-sized airport and a small airport. The specific value of the weight is set by technical personnel according to actual conditions, and the embodiment of the present application does not limit this. The moving distance and moving time of the target object after entering the target airport are collected by the mobile terminal of the target object and uploaded to the server. Of course, the moving distance and moving time are collected and uploaded to the server after the permission of the target object. Crowd density is the ratio of the number of people to the area. The reference moving speed determination model is obtained by training based on multiple sample crowd densities and the labeled reference moving speeds corresponding to each sample crowd density, and has the ability to predict the reference moving speed based on crowd density.

[0070] For example, the server uses the object identifier of the target object to query an airport database to obtain the multiple reference airports. The airport database stores the multiple airports, and the reference airports are airports associated with the object identifier of the target object. The server uses the object identifier of the target object to query the object database of each reference airport to obtain the average movement speed of the target object at each reference airport. The server divides the distance moved by the movement duration after the target object enters the airport to obtain the second movement speed of the target object. The server obtains the current crowd density at the target airport, inputs the crowd density into a reference movement speed determination model, and uses the reference movement speed determination model to perform feature extraction on the crowd density to obtain a crowd density feature for the crowd density. The crowd density includes crowd density in multiple areas of the target airport. The server uses the reference movement speed determination model to fully connect and normalize the crowd density feature to obtain the reference movement speed.

[0071] Part three: The server determines the target moving speed of the target object based on the first moving speed, the second moving speed and the reference moving speed of the target airport.

[0072] In a possible implementation, the server performs weighted fusion on the first moving speed, the second moving speed, and the reference moving speed of the target airport of the target object to obtain the target moving speed of the target object.

[0073] 302. When the time difference between the remaining boarding time and the time required for boarding is greater than or equal to a preset time difference threshold, the server obtains the store preference information of the target object and the store information of multiple candidate stores at the target airport. The candidate stores are stores that can be reached after passing the security check, and the store information includes a store description and a store location.

[0074] Among them, if the duration difference is greater than or equal to the preset duration difference threshold, it means that the target object has ample time to board the plane, and the target object's boarding path passing through the store will not affect the target object's normal boarding. The preset duration difference threshold is set by technical personnel based on actual conditions, or determined based on the airport information of the target airport, and this embodiment of the application does not limit this. The store description of the candidate store is used to describe the store situation of the candidate store, and the store location refers to the location of the candidate store in the target airport.

[0075] In one possible implementation, if the difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference threshold, the server obtains the target subject's historical store visit information, historical store consumption information, and real-time store visit information and real-time store consumption information at the target airport. The real-time store visit information and real-time store consumption information are the target subject's store visit information and store consumption information corresponding to the target subject after entering the target airport and before passing through security. The server determines the target subject's store preference information based on the target subject's historical store visit information, historical store consumption information, real-time store visit information, and real-time store consumption information. The server obtains an indoor map of the target airport from the geographic information platform. The server identifies multiple candidate stores from the indoor map and obtains the location of each candidate store. The server queries a store description database at the target airport based on the multiple candidate stores to obtain a store description for each candidate store. The server combines the store description and store location of each candidate store to obtain store information for each candidate store.

[0076] Store visit information includes the store name, store type, store brand, store visit duration, and store visit count. Store consumption information includes the type of goods consumed, the amount spent, and the number of purchases. Historical store visit information directly reflects the target user's historical preferences for stores, while real-time store visit information directly reflects the target user's real-time preferences for stores. Historical store consumption information directly reflects the target user's historical preferences for goods. The association between goods and stores indirectly reflects the target user's historical preferences for stores. Real-time store consumption information directly reflects the target user's real-time preferences for goods, and thus indirectly reflects the target user's real-time preferences for stores. In summary, the target user's store preference information is a fusion of the target user's historical and real-time preferences for stores, and can more accurately reflect the target user's preferences for stores. The indoor map of the target airport is a three-dimensional electronic map that includes the location information of multiple points of interest (POIs) within the target airport. POIs include stores, service points, public facilities, and boarding gates within the target airport. Store descriptions include store type, store brand, store product information, and store area.

[0077] In order to explain the above embodiment more clearly, the method of determining the store preference information of the target object in the above embodiment is described below.

[0078] In one possible implementation, the server determines multiple first historical stores of interest to the target object based on the historical store residency information of the target object. The server determines multiple second historical stores of interest to the target object based on the historical store consumption information of the target object. The server determines multiple first real-time stores of interest to the target object based on the real-time store residency information of the target object. The server determines multiple second real-time stores of interest to the target object based on the real-time store consumption information of the target object. The server determines the store preference information of the target object based on the multiple first historical stores of interest, the multiple second historical stores of interest, the multiple first real-time stores of interest, and the multiple second real-time stores of interest.

[0079] In order to explain the above embodiment more clearly, the above embodiment will be described in several parts below.

[0080] In the first part, the server determines a plurality of first historical stores of interest to the target object based on the historical store residency information of the target object.

[0081] In one possible implementation, the server obtains the store names, store types, store brands, store dwelling durations, and store dwelling counts of multiple historically visited stores from the historical store dwelling information of the target object. The server determines multiple first historically interested stores of the target object based on the store names, store types, store brands, store dwelling durations, and store dwelling counts of the multiple historically visited stores.

[0082] Among them, historically visited stores are stores that the target object has visited in multiple airports.

[0083] For example, the server obtains the store names, store types, store brands, store residence times, and store residence times of multiple historically resident stores from the historical store residence information of the target object. Based on the store residence times and store residence times of the multiple historically resident stores, the server determines multiple reference historically resident stores from the multiple historically resident stores. The reference historically resident stores are historically resident stores whose store residence times are greater than or equal to a residence time threshold, and / or whose store residence times are greater than or equal to a residence time threshold. The server performs store recall based on the store names, store types, and store brands of the multiple reference historically resident stores to obtain multiple first initial candidate stores. The server filters the multiple first initial candidate stores based on the object type of the target object to obtain multiple first target candidate stores, which are first initial candidate stores with a high degree of match with the object type. The server determines the set of the multiple reference historically resident stores and the multiple first target candidate stores as the multiple first historically interested stores of the target object.

[0084] Among them, the object type of the target object includes gender, age, hobbies, etc., and the object type of the target object is obtained after the target object has given permission. The purpose of recalling stores based on the store names, store types, and store brands of multiple reference historical resident stores is to find first initial candidate stores that are similar to multiple reference historical resident stores, in the hope of digging out more stores that the target object may be interested in. Based on the object type of the target object, multiple first target candidate stores are found from multiple first initial candidate stores, and the first target candidate stores have a higher degree of matching with the target object. The residence time threshold and the residence number threshold are set by technical personnel according to actual conditions, and are not limited in the embodiments of the present application.

[0085] In order to illustrate the technical solution provided in the above example, the following describes a method in which the server in the above example screens the multiple first initial candidate stores based on the object type of the target object to obtain multiple first target candidate stores.

[0086] In some embodiments, the server extracts features of the object type of the target object to obtain object type features of the target object. The server extracts features of the store descriptions of each first initial candidate store to obtain store description features of each first initial candidate store. Based on the object type features of the target object and the store description features of each first initial candidate store, the server determines the degree of match between each first initial candidate store and the target object. The server determines the initial reference store among the multiple first initial candidate stores whose degree of match with the target object is greater than or equal to a matching degree threshold as the first target candidate store, thereby obtaining multiple first target candidate stores.

[0087] Part 2: The server determines a plurality of second historical stores of interest to the target object based on the historical store consumption information of the target object.

[0088] In one possible implementation, the server obtains multiple historically consumed products and the product type, product price, product name, and number of purchases of each historically consumed product from the target subject's historical store consumption information. Based on the multiple historically consumed products and the number of purchases of each historically consumed product, the server determines multiple second historically interested stores of the target subject.

[0089] For example, the server obtains multiple historical consumer goods and the product type, product price, product name, and product purchase frequency of each historical consumer product from the historical store consumption information of the target object. Based on the product purchase frequency of each historical consumer product, the server determines multiple reference consumer goods from the multiple historical consumer goods. The reference consumer goods are historical consumer goods with a purchase frequency greater than or equal to a purchase frequency threshold. The server performs store recall based on the product type, product price, and product name of each reference consumer product to obtain multiple second initial candidate stores. The server filters the multiple second initial candidate stores based on the object type of the target object to obtain multiple second target candidate stores. The second target candidate stores are second initial candidate stores with a high degree of match with the object type. The server determines the multiple second target candidate stores as the multiple second historical stores of interest for the target object.

[0090] The purpose of recalling stores based on the product type, product price, and product name of each reference consumer product is to find a second initial candidate store related to the historical consumer product, in the hope of discovering more stores that the target object may be interested in.

[0091] Part three: The server determines a plurality of first real-time interested stores of the target object based on the real-time store residency information of the target object.

[0092] In one possible implementation, the server obtains the store names, store types, store brands, store dwelling durations, and store dwell counts of multiple real-time resident stores from the real-time store dwelling information of the target object. The server determines multiple first real-time stores of interest to the target object based on the store names, store types, store brands, store dwelling durations, and store dwell counts of the multiple real-time resident stores.

[0093] Among them, the real-time resident stores are stores where the target object has stayed in multiple airports.

[0094] For example, the server obtains the store names, store types, store brands, store dwell time, and store dwell counts of multiple real-time resident stores from the real-time store dwell information of the target object. Based on the store dwell time and store dwell count of the multiple real-time resident stores, the server determines multiple reference real-time resident stores from the multiple real-time resident stores. The reference real-time resident stores are real-time resident stores whose store dwell time is greater than or equal to a dwell time threshold, and / or whose store dwell count is greater than or equal to a dwell count threshold. The server performs store recall based on the store names, store types, and store brands of the multiple reference real-time resident stores to obtain multiple third initial candidate stores. The server filters the multiple third initial candidate stores based on the object type of the target object to obtain multiple third target candidate stores, which are third initial candidate stores with a high degree of match with the object type. The server determines the set of the multiple reference real-time resident stores and the multiple third target candidate stores as the multiple first real-time stores of interest for the target object.

[0095] The purpose of performing store recall based on the store names, store types, and store brands of multiple reference real-time resident stores is to find third initial candidate stores that are similar to the multiple reference real-time resident stores, in the hope of discovering more stores that the target object may be interested in. By finding multiple third target candidate stores from the multiple third initial candidate stores based on the target object's object type, the third target candidate stores have a higher degree of match with the target object.

[0096] Part 4: The server determines a plurality of second real-time interested stores of the target object based on the real-time store consumption information of the target object.

[0097] In one possible implementation, the server obtains multiple real-time consumed products and the product type, product price, product name, and number of purchases of each real-time consumed product from the real-time store consumption information of the target object. Based on the multiple real-time consumed products and the number of purchases of each real-time consumed product, the server determines multiple second real-time stores of interest to the target object.

[0098] For example, the server obtains multiple real-time consumer goods and the product type, product price, product name and number of purchases of each real-time consumer product from the real-time store consumption information of the target object. Based on the number of purchases of each real-time consumer product, the server determines multiple reference consumer goods from the multiple real-time consumer goods. The reference consumer goods are real-time consumer goods whose purchase times are greater than or equal to the purchase times threshold. The server performs store recall based on the product type, product price and product name of each reference consumer product to obtain multiple fourth initial candidate stores. The server filters the multiple fourth initial candidate stores based on the object type of the target object to obtain multiple fourth target candidate stores. The fourth target candidate stores are fourth initial candidate stores with a high degree of match with the object type. The server determines the multiple fourth target candidate stores as the multiple second real-time interested stores of the target object.

[0099] The purpose of recalling stores based on the product type, product price, and product name of each reference consumer product is to find a fourth initial candidate store related to the real-time consumer product, in the hope of discovering more stores that the target object may be interested in.

[0100] Part 5: The server determines the store preference information of the target object based on the multiple first historical interest stores, the multiple second historical interest stores, the multiple first real-time interest stores, and the multiple second real-time interest stores.

[0101] In one possible implementation, the server determines the target object's store types and brands of interest based on the multiple first historical stores of interest, the multiple second historical stores of interest, the multiple first real-time stores of interest, and the multiple second real-time stores of interest. The server determines the target object's store preference information based on the target object's store types and brands of interest.

[0102] For example, the server performs store type statistics and brand statistics on the multiple first historical stores of interest, the multiple second historical stores of interest, the multiple first real-time stores of interest, and the multiple second real-time stores of interest, and obtains multiple candidate store types, the number of occurrences of each candidate store type, multiple candidate brands, and the number of occurrences of each candidate brand. The server determines multiple target store types from the multiple candidate store types based on the number of occurrences of each candidate store type. The server determines multiple target brands from the multiple candidate brands based on the number of occurrences of each candidate brand. The server determines the multiple target store types as the store types of interest to the target object, and determines the multiple target brands as the brands of interest to the target object. The server determines the set of the store types of interest and brands of interest to the target object as the store preference information of the target object.

[0103] 303. The server determines multiple target stores from the multiple candidate stores based on the store preference information of the target object, the current location of the target object, the location of the target boarding gate, and the store information of the multiple candidate stores. The store locations and the location of the target boarding gate are determined based on a geographic information platform.

[0104] The target subject's store preference information reflects the target subject's store preferences. A target store is a candidate store that does not affect the target subject's boarding and that the target subject is interested in. The geographic information platform stores a large amount of geographic information, including both indoor and outdoor geographic information. The aforementioned store locations and target boarding gate locations are examples of indoor geographic information.

[0105] In one possible implementation, the server determines a plurality of reference stores from the plurality of candidate stores based on the target subject's store preference information and the store descriptions in the store information of the plurality of candidate stores. The reference stores are candidate stores in which the target subject has a high level of interest. The server determines the plurality of target stores from the plurality of reference stores based on the target subject's current location, the location of the target boarding gate, and the store locations in the store information of the plurality of reference stores.

[0106] In order to explain the above embodiment more clearly, the above embodiment will be described in several parts below.

[0107] In the first part, the server determines a plurality of reference stores from the plurality of candidate stores based on the store preference information of the target object and the store descriptions in the store information of the plurality of candidate stores.

[0108] In one possible implementation, the server obtains the target object's store type and brand of interest from the target object's store preference information. The server uses the target object's store type and brand of interest to query multiple reference stores, obtaining multiple reference stores whose store types fall within the target object's store type of interest and / or whose brands fall within the target object's brand of interest.

[0109] In the second part, the server determines the multiple target stores from the multiple reference stores based on the current location of the target object, the location of the target boarding gate, and the store locations in the store information of the multiple reference stores.

[0110] In one possible implementation, the server determines multiple candidate boarding paths based on the current location of the target object and the location of the target boarding gate. The candidate boarding paths are paths from the current location to the target boarding gate. The server determines a target boarding area based on the multiple candidate boarding paths. The target boarding area is an area covering the multiple boarding paths. The server determines the multiple target stores belonging to the target boarding area from the multiple reference stores based on the store locations in the store information of the multiple reference stores.

[0111] Under this implementation method, the target boarding area is used to filter out multiple target stores from multiple reference stores, ensuring that the obtained target stores are all located in the area that the target object passes through when boarding the plane, avoiding planning stores that are out of the target boarding area for the target object, saving the target object's time, and reducing the probability of the target object missing the flight.

[0112] For example, the server performs path planning on an indoor map of the target airport based on the current location of the target object and the location of the target boarding gate, obtaining multiple candidate boarding paths. The server identifies the area covered by the multiple candidate boarding paths on the indoor map as the target boarding area. The server then identifies the multiple target stores within the target boarding area from the multiple reference stores based on the store locations in the store information of the multiple reference stores.

[0113] 304. The server plans a boarding path for the target object based on the current location of the target object, the object information of the target object, the location of the target boarding gate, the remaining boarding time, and the store information of the multiple target stores, and obtains a target boarding path for the target object. The target boarding path passes through at least one target store among the multiple target stores to reach the target boarding gate.

[0114] The boarding path planning process involves finding a target boarding path recommended to the target passenger. This path is a path that does not affect the target passenger's normal boarding and passes by at least one target store, thus achieving both boarding guidance and target store recommendations.

[0115] In one possible implementation, the server determines the estimated residence time of the target object in each target store based on the store description in the store information of the multiple target stores and the object information of the target object. The server creates multiple location nodes based on the current location, the location of the target boarding gate, and the store location in the store information of each target store. The server assigns the estimated residence time of each target store to the corresponding location node, and determines the movement time between each two location nodes in the multiple location nodes. Based on the movement time between each two location nodes in the multiple location nodes, the server adds a connection between each two location nodes in the multiple location nodes after assignment to obtain a target graph network. The server uses the location node corresponding to the current location as the starting point, the location node corresponding to the location of the target boarding gate as the end point, and the remaining boarding time as a constraint to perform linear programming on the target graph network to obtain the target boarding path of the target object.

[0116] The location node is an abstract representation of the target store, and the line between two location nodes represents the time required to travel between them, also known as the travel time. The remaining boarding time constraint means that the total time on the target boarding path cannot exceed the remaining boarding time, ensuring that the target person can board the flight on time.

[0117] In order to explain the above embodiment more clearly, the above embodiment will be described in several parts below.

[0118] In the first part, the server determines an estimated residence time of the target object in each target store based on the store descriptions in the store information of the multiple target stores and the object information of the target object.

[0119] In one possible implementation, the server performs feature extraction on the store description of each target store to obtain store description features of each store. The store description includes the store type, store brand, store product information, and store area. The store description features include store type sub-features, store brand sub-features, store product sub-features, and store area sub-features. The server performs feature extraction on the object information of the target object to obtain object features of the target object. The object information includes object type, store preference information, and object status. Based on the store description features of each target store and the object features of the target object, the server determines the estimated residence time of the target object in each target store.

[0120] The following describes how the server determines the estimated residence time of the target object in each target store based on the store description features of each store and the object features of the target object in the above embodiment.

[0121] In some embodiments, for any target store among the multiple target stores, the server concatenates the store description features of the target store and the object features of the target object to obtain a concatenated feature. The server inputs the concatenated feature into a dwell time prediction model and performs multiple full connections on the concatenated feature using the dwell time prediction model to obtain a dwell time estimation feature. The server normalizes the dwell time estimation feature using the dwell time prediction model to obtain an estimated dwell time of the target object in the target store.

[0122] Among them, the residence time prediction model is trained based on the object features of multiple sample objects, the store description features of multiple sample stores and the corresponding annotated residence times, and has the ability to use object features and store description features to determine the predicted residence time.

[0123] In the second part, the server creates a plurality of location nodes based on the current location, the location of the target boarding gate, and the store locations in the store information of each target store.

[0124] In one possible implementation, the server creates multiple location nodes on a blank map based on the current location, the location of the target boarding gate, and the store locations in the store information of each target store. The relative position relationship of the multiple location nodes on the blank map is the same as that of the current location, the target boarding gate, and the multiple target stores.

[0125] In the third part, the server assigns the estimated residence time of each target store to the corresponding location node, and determines the movement time between every two location nodes in the multiple location nodes.

[0126] In one possible implementation, the server assigns an estimated dwell time for each target store to a corresponding location node. The server then divides the distance between each two location nodes by the target object's target movement speed to obtain a travel time between each two location nodes.

[0127] In the fourth part, the server adds a connection between every two position nodes in the multiple position nodes after the assignment based on the movement time between every two position nodes in the multiple position nodes to obtain a target graph network.

[0128] The value of the connecting line is the corresponding movement time.

[0129] In the fifth part, the server uses the location node corresponding to the current location as the starting point, the location node corresponding to the location of the target boarding gate as the end point, and the remaining boarding time as a constraint, and performs linear programming on the target graph network to obtain the target boarding path of the target object.

[0130] In one possible implementation, the server determines multiple initial boarding paths within the target graph network, starting with the location node corresponding to the current location and ending with the location node corresponding to the target boarding gate. The server adds the sum of the travel time corresponding to each initial boarding path to the sum of the estimated dwell time to obtain the total boarding time for each initial boarding path. The server uses the remaining boarding time as a constraint on the total boarding time for each initial boarding path and performs multiple rounds of iterative adjustments to the location nodes in each initial boarding path to determine the target boarding path for the target object.

[0131] For example, the server takes the location node corresponding to the current location as the starting point and the location node corresponding to the location of the target boarding gate as the end point, performs path planning in the target graph network, and obtains multiple initial boarding paths. The initial boarding path is a boarding path that can reach the target boarding gate from the current location. The server adds the sum of the movement time corresponding to each initial boarding path and the sum of the estimated residence time to obtain the total boarding time of each initial boarding path. During any round of iteration of any initial boarding path, if the total boarding time of the initial boarding path is greater than the remaining boarding time, the server randomly deletes the location node corresponding to the target store in the initial boarding path, or adjusts the connection between the location nodes corresponding to the target store in the initial boarding path, and completes the round of iteration until the total boarding time of the initial boarding path after the iteration is less than or equal to the remaining boarding time. The iteration of the initial boarding path is completed, and the initial boarding path after the iteration is determined as the target boarding path.

[0132] 305. The server sends the target boarding path of the target object to the mobile terminal of the target object, so that the target boarding path is displayed through the mobile terminal.

[0133] Among them, the number of target boarding paths can be multiple, and the target object can select the desired target boarding path on the mobile terminal. In the embodiment of the present application, recommending the target boarding path to the target object is the intelligent customer service based on the GIS geographic information platform provided by the target airport. This intelligent customer service can make full use of the store resources of the target airport and recommend boarding paths that pass through the stores of interest to passengers while ensuring that passengers do not miss their flights. This intelligent service is turned on or off by the target object. When the target object chooses to turn on the intelligent service, the target object needs to authorize the server to obtain the target object's object information, store preference information, store consumption information, and store residency information, etc. If the target object turns on the intelligent service but does not authorize the server to obtain the target object's object information, store preference information, store consumption information, and store residency information, the server can directly recommend the preset boarding path to the target object.

[0134] Optionally, after step 305, the server can also perform the following steps.

[0135] In one possible implementation, the server determines the target object's object position in real time. The server determines the shortest boarding path from the target object's object position to the target boarding gate. The server divides the path length of the shortest boarding path by the target object's target moving speed to obtain the fastest boarding time. The server compares the fastest boarding time with the remaining boarding time. If the time difference between the remaining boarding time and the fastest boarding time is less than a warning time difference threshold, the server sends an alert to the target object's mobile terminal, prompting the target object to immediately proceed to the target boarding gate for boarding.

[0136] Among them, the warning duration difference threshold is set by technical personnel according to actual conditions, and the embodiments of this application do not limit this.

[0137] In this implementation, if the target person is at a target store at a target airport, the server can determine the target person's location and the fastest boarding time in real time. If boarding time is tight, the server can promptly remind the target person to board, thereby reducing the probability of the target person missing their flight.

[0138] In some embodiments, the server sends the shortest boarding path to the mobile terminal at the same time as sending the warning information to the mobile terminal, so that the target object can reach the target boarding gate according to the shortest boarding path.

[0139] All of the above optional technical solutions can be combined in any way to form optional embodiments of the present application, and will not be described in detail here.

[0140] Through the technical solution provided by the embodiments of the present application, when a target subject passes security at a target airport, the target subject's remaining boarding time and the required boarding time are determined. If the difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference, the target subject's store preference information and the store information of multiple candidate stores at the target airport are obtained. The candidate stores are stores that can be reached after passing security. Based on the target subject's store preference information, the target subject's current location, the location of the target boarding gate, and the store information of the multiple candidate stores, multiple target stores are determined from the multiple candidate stores. The store locations and the location of the target boarding gate are determined based on a geographic information platform. Based on the target subject's current location, the target subject's object information, the location of the target boarding gate, the time difference, and the store information of the multiple target stores, a boarding path is planned for the target subject to obtain the target boarding path. The target boarding path is displayed to the target subject, providing both boarding guidance for the target subject and guidance to the target stores, thereby meeting the target subject's needs.

[0141] Figure 4 This is a schematic diagram of the structure of an intelligent customer service device based on a GIS geographic information platform provided in an embodiment of the present application. Figure 4 The device includes: a duration determination module 401, an information acquisition module 402, a target store determination module 403 and a path planning module 404.

[0142] The duration determination module 401 is used to determine the target object's remaining boarding time and the required boarding time when the target object passes the security check at the target airport. The remaining boarding time is the time between the boarding time and the current time. The required boarding time is the time required for the target object to reach the target boarding gate from the current location. The target boarding gate is the boarding gate of the flight that the target object will take.

[0143] The information acquisition module 402 is used to obtain the store preference information of the target object and the store information of multiple candidate stores at the target airport when the time difference between the remaining boarding time and the time required for boarding is greater than or equal to a preset time difference threshold. The candidate stores are stores that can be reached after passing the security check, and the store information includes a store description and a store location.

[0144] The target store determination module 403 is used to determine multiple target stores from the multiple candidate stores based on the store preference information of the target object, the current location of the target object, the location of the target boarding gate and the store information of the multiple candidate stores. The store locations and the location of the target boarding gate are determined based on a geographic information platform.

[0145] The path planning module 404 is used to plan the boarding path of the target object based on the current location of the target object, the object information of the target object, the location of the target boarding gate, the remaining boarding time, and the store information of the multiple target stores, and obtain the target boarding path of the target object, where the target boarding path passes through at least one target store among the multiple target stores to reach the target boarding gate.

[0146] In one possible embodiment, the duration determination module 401 is used to determine the remaining boarding time of the target object when the target object passes the security check of the target airport. The first moving speed, the second moving speed and the reference moving speed of the target airport are obtained, wherein the first moving speed is the average moving speed of the target object at multiple airports, the second moving speed is the average moving speed of the target object after entering the target airport, and the reference moving speed is the moving speed determined based on the passenger density in the target airport. Based on the first moving speed, the second moving speed and the reference moving speed of the target airport, the target moving speed of the target object is determined. The path length of the reference boarding path between the position of the target boarding gate and the current position of the target object is divided by the target moving speed to obtain the time required for boarding. The reference boarding path is a boarding path whose selected times meet the preset number conditions.

[0147] In one possible embodiment, the information acquisition module 402 is configured to, when the difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference threshold, acquire the target subject's historical store visit information, historical store consumption information, and real-time store visit information and real-time store consumption information at multiple airports. The real-time store visit information and real-time store consumption information are the target subject's store visit information and store consumption information corresponding to the target subject after entering the target airport and before passing through security. The target subject's store preference information is determined based on the target subject's historical store visit information, historical store consumption information, real-time store visit information, and real-time store consumption information. An indoor map of the target airport is acquired from the geographic information platform. Multiple candidate stores are identified from the indoor map and the location of each candidate store is acquired. A store description database of the target airport is queried based on the multiple candidate stores to acquire a store description for each candidate store. The store description and store location of each candidate store are combined to acquire store information for each candidate store.

[0148] In one possible implementation, the information acquisition module 402 is configured to determine a plurality of first historical stores of interest to the target object based on the historical store residency information of the target object. Determine a plurality of second historical stores of interest to the target object based on the historical store consumption information of the target object. Determine a plurality of first real-time stores of interest to the target object based on the real-time store residency information of the target object. Determine a plurality of second real-time stores of interest to the target object based on the real-time store consumption information of the target object. Determine the store preference information of the target object based on the plurality of first historical stores of interest, the plurality of second historical stores of interest, the plurality of first real-time stores of interest, and the plurality of second real-time stores of interest.

[0149] In one possible implementation, the information acquisition module 402 is configured to determine the target subject's store types and brands of interest based on the plurality of first historical stores of interest, the plurality of second historical stores of interest, the plurality of first real-time stores of interest, and the plurality of second real-time stores of interest. Based on the target subject's store types and brands of interest, the target subject's store preference information is determined.

[0150] In one possible implementation, the target store determination module 403 is configured to determine a plurality of reference stores from the plurality of candidate stores based on the target subject's store preference information and the store descriptions in the store information of the plurality of candidate stores. The reference stores are candidate stores in which the target subject has a high level of interest. The plurality of target stores are determined from the plurality of reference stores based on the target subject's current location, the location of the target boarding gate, and the store locations in the store information of the plurality of reference stores.

[0151] In one possible implementation, the target store determination module 403 is configured to determine multiple candidate boarding paths based on the current location of the target object and the location of the target boarding gate, where the candidate boarding paths are paths from the current location to the target boarding gate. Based on the multiple candidate boarding paths, a target boarding area is determined, where the target boarding area is the area covering the multiple boarding paths. Based on the store locations in the store information of the multiple reference stores, the multiple target stores belonging to the target boarding area are determined from the multiple reference stores.

[0152] In one possible implementation, the path planning module 404 is used to determine the estimated residence time of the target object in each target store based on the store description in the store information of the multiple target stores and the object information of the target object. Based on the current location, the location of the target boarding gate and the store location in the store information of each target store, multiple location nodes are created. The estimated residence time of each target store is assigned to the corresponding location node, and the movement time between each two location nodes in the multiple location nodes is determined. Based on the movement time between each two location nodes in the multiple location nodes, a connection is added between each two location nodes in the multiple location nodes after assignment to obtain a target graph network. With the location node corresponding to the current location as the starting point, the location node corresponding to the location of the target boarding gate as the end point, and the remaining boarding time as the constraint, linear programming is performed on the target graph network to obtain the target boarding path of the target object.

[0153] In one possible implementation, the path planning module 404 is used to extract features from the store descriptions of each target store to obtain store description features of each store, where the store description includes the store type, store brand, store product information, and store area, and the store description features include store type sub-features, store brand sub-features, store product sub-features, and store area sub-features. Feature extraction is performed on the object information of the target object to obtain object features of the target object, where the object information includes object type, store preference information, and object status. Based on the store description features of each target store and the object features of the target object, the estimated residence time of the target object in each target store is determined.

[0154] In one possible implementation, the path planning module 404 is configured to determine multiple initial boarding paths within the target graph network, starting from a location node corresponding to the current location and ending at a location node corresponding to the target boarding gate. The total boarding time for each initial boarding path is calculated by adding the total travel time and the estimated dwell time to the total boarding time for each initial boarding path. The remaining boarding time is used to constrain the total boarding time for each initial boarding path, and multiple rounds of iterative adjustments are performed on the location nodes in each initial boarding path to determine the target boarding path for the target object.

[0155] It should be noted that the intelligent customer service device based on the GIS geographic information platform provided in the above embodiment is only illustrated by the division of the above functional modules when performing intelligent customer service based on the GIS geographic information platform. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the intelligent customer service device based on the GIS geographic information platform provided in the above embodiment and the intelligent customer service method embodiment based on the GIS geographic information platform are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0156] Through the technical solution provided by the embodiments of the present application, when a target subject passes security at a target airport, the target subject's remaining boarding time and the required boarding time are determined. If the difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference, the target subject's store preference information and the store information of multiple candidate stores at the target airport are obtained. The candidate stores are stores that can be reached after passing security. Based on the target subject's store preference information, the target subject's current location, the location of the target boarding gate, and the store information of the multiple candidate stores, multiple target stores are determined from the multiple candidate stores. The store locations and the location of the target boarding gate are determined based on a geographic information platform. Based on the target subject's current location, the target subject's object information, the location of the target boarding gate, the time difference, and the store information of the multiple target stores, a boarding path is planned for the target subject to obtain the target boarding path. The target boarding path is displayed to the target subject, providing both boarding guidance for the target subject and guidance to the target stores, thereby meeting the target subject's needs.

[0157] Figure 5 This is a schematic diagram of the structure of a server provided in an embodiment of the present application. The server 500 may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) 501 and one or more memories 502, wherein the one or more memories 502 store at least one computer program, and the at least one computer program is loaded and executed by the one or more processors 501 to implement the methods provided in the above-mentioned various method embodiments. Of course, the server 500 may also have components such as a wired or wireless network interface, a keyboard, and an input / output interface for input and output. The server 500 may also include other components for implementing device functions, which will not be described in detail here.

[0158] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including a computer program. The computer program can be executed by a processor to implement the intelligent customer service method based on the GIS geographic information platform in the above embodiment. For example, the computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, or an optical data storage device.

[0159] In an exemplary embodiment, a computer program product or computer program is also provided, which includes a program code, which is stored in a computer-readable storage medium. The processor of a computer device reads the program code from the computer-readable storage medium, and the processor executes the program code, so that the computer device executes the above-mentioned intelligent customer service method based on the GIS geographic information platform.

[0160] In some embodiments, the computer program involved in the embodiments of the present application may be deployed and executed on a computer device, or on multiple computer devices located at one location, or on multiple computer devices distributed at multiple locations and interconnected through a communication network. Multiple computer devices distributed at multiple locations and interconnected through a communication network may constitute a blockchain system.

[0161] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or may be accomplished by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0162] The above are only optional embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. An intelligent customer service method based on GIS geographic information platform, characterized in that: The method comprises: When the target subject passes security at the target airport, determining the target subject's remaining boarding time and required boarding time, where the remaining boarding time is the time between the boarding time and the current time, and the required boarding time is the time required for the target subject to reach the target boarding gate from the current location, where the target boarding gate is the boarding gate for the flight the target subject will take; If the difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference threshold, obtaining store preference information of the target object and store information of multiple candidate stores at the target airport, where the candidate stores are accessible after passing security check, and the store information includes store descriptions and store locations; Determining a plurality of target stores from the plurality of candidate stores based on the store preference information of the target object, the current location of the target object, the location of the target boarding gate, and the store information of the plurality of candidate stores includes: Determining, based on the current location of the target object and the location of the target boarding gate, a plurality of candidate boarding paths, wherein the candidate boarding paths are paths moving from the current location to the target boarding gate; determining, based on the plurality of candidate boarding paths, a target boarding area, wherein the target boarding area is an area covering the plurality of candidate boarding paths; determining, based on store locations in store information of a plurality of reference stores, a plurality of target stores belonging to the target boarding area from the plurality of reference stores, wherein the reference stores are candidate stores in which the target object has a high degree of interest, and the store locations and the location of the target boarding gate are determined based on a geographic information platform; Based on the current location of the target object, the object information of the target object, the location of the target boarding gate, the remaining boarding time, and the store information of the multiple target stores, a boarding path is planned for the target object to obtain a target boarding path for the target object, and the target boarding path passes through at least one target store among the multiple target stores to reach the target boarding gate.

2. The method according to claim 1, characterized in that The step of determining the remaining boarding time and the time required for boarding the target object when the target object passes security inspection at the target airport includes: When the target object passes security at the target airport, determining the remaining boarding time of the target object; Obtaining a first moving speed and a second moving speed of the target object, and a reference moving speed of the target airport, where the first moving speed is an average moving speed of the target object at multiple airports, the second moving speed is an average moving speed of the target object after entering the target airport, and the reference moving speed is a moving speed determined based on a passenger flow density within the target airport; determining a target moving speed of the target object based on the first moving speed and the second moving speed of the target object and a reference moving speed of the target airport; The path length of the reference boarding path between the position of the target boarding gate and the current position of the target object is divided by the target moving speed to obtain the boarding time required, and the reference boarding path is a boarding path that is selected a number of times that meets the preset number of conditions.

3. The method according to claim 1, characterized in that When the time difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference threshold, acquiring the store preference information of the target object and the store information of the plurality of candidate stores at the target airport includes: When the difference between the remaining boarding time and the required boarding time is greater than or equal to a preset time difference threshold, obtaining historical store visit information and historical store consumption information of the target subject at multiple airports, and real-time store visit information and real-time store consumption information of the target subject at the target airport, wherein the real-time store visit information and the real-time store consumption information are corresponding store visit information and store consumption information of the target subject after entering the target airport and before passing through security check; Determining the store preference information of the target object based on the historical store residency information, historical store consumption information, the real-time store residency information, and the real-time store consumption information of the target object; Obtaining an indoor map of the target airport from the geographic information platform; Determining the plurality of candidate stores from the indoor map and obtaining a store location of each of the candidate stores; querying a store description database of the target airport based on the multiple candidate stores to obtain a store description of each candidate store; The store description and store location of each candidate store are combined to obtain the store information of each candidate store.

4. The method according to claim 3, characterized in that The determining of the target object's store preference information based on the target object's historical store residency information, historical store consumption information, the real-time store residency information, and the real-time store consumption information includes: Determining a plurality of first historical stores of interest to the target object based on the historical store residency information of the target object; Determining a plurality of second historical stores of interest to the target object based on the historical store consumption information of the target object; Determining a plurality of first real-time interested stores of the target object based on the real-time store residency information of the target object; Determining a plurality of second real-time interested stores of the target object based on the real-time store consumption information of the target object; The store preference information of the target object is determined based on the plurality of first historically interested stores, the plurality of second historically interested stores, the plurality of first real-time interested stores, and the plurality of second real-time interested stores.

5. The method according to claim 4, characterized in that The determining the store preference information of the target object based on the plurality of first historically interested stores, the plurality of second historically interested stores, the plurality of first real-time interested stores, and the plurality of second real-time interested stores includes: Determining the target object's store types and brands of interest based on the plurality of first historical stores of interest, the plurality of second historical stores of interest, the plurality of first real-time stores of interest, and the plurality of second real-time stores of interest; Based on the store types and brands of interest to the target object, store preference information of the target object is determined.

6. The method according to claim 1, wherein The determining of a plurality of target stores from the plurality of candidate stores based on the store preference information of the target object, the current location of the target object, the location of the target boarding gate, and the store information of the plurality of candidate stores includes: A plurality of reference stores is determined from the plurality of candidate stores based on the store preference information of the target object and the store descriptions in the store information of the plurality of candidate stores.

7. The method according to claim 1, characterized in that The step of planning a boarding path for the target object based on the current location of the target object, the object information of the target object, the location of the target boarding gate, the remaining boarding time, and the store information of the plurality of target stores to obtain a target boarding path for the target object includes: Determining an estimated dwelling time of the target object in each of the target stores based on the store descriptions in the store information of the multiple target stores and the object information of the target object; creating a plurality of location nodes based on the current location, the location of the target boarding gate, and the store locations in the store information of each of the target stores; Assigning the estimated dwell time of each target store to a corresponding location node, and determining a travel time between every two location nodes among the plurality of location nodes; Based on the movement time between each two position nodes in the plurality of position nodes, adding a connection line between each two position nodes in the plurality of position nodes after the assignment, to obtain a target graph network; With the location node corresponding to the current location as the starting point, the location node corresponding to the location of the target boarding gate as the end point, and the remaining boarding time as a constraint, linear programming is performed on the target graph network to obtain the target boarding path of the target object.

8. The method according to claim 7, characterized in that The determining, based on the store descriptions in the store information of the multiple target stores and the object information of the target objects, an estimated residence time of the target object in each of the target stores includes: Extracting features from the store descriptions of each target store to obtain store description features of each target store, wherein the store descriptions include store type, store brand, store product information, and store area, and the store description features include store type sub-features, store brand sub-features, store product sub-features, and store area sub-features; Extracting features from the target object's object information to obtain object features of the target object, wherein the object information includes object type, store preference information, and object status; Based on the store description features of each of the target stores and the object features of the target object, an estimated residence time of the target object in each of the target stores is determined.

9. The method according to claim 7, characterized in that The step of performing linear programming on the target graph network with the location node corresponding to the current location as a starting point, the location node corresponding to the location of the target boarding gate as an end point, and the remaining boarding time as a constraint to obtain a target boarding path for the target object includes: Determine a plurality of initial boarding paths in the target graph network, with the location node corresponding to the current location as a starting point and the location node corresponding to the location of the target boarding gate as an end point; Adding the sum of the travel time corresponding to each of the initial boarding paths to the sum of the estimated dwell time to obtain the total boarding time for each of the initial boarding paths; The remaining boarding time is used to constrain the total boarding time of each of the initial boarding paths, and multiple rounds of iterative adjustments are performed on the position nodes in each of the initial boarding paths to obtain the target boarding path of the target object.

Images