CN113924460A

CN113924460A - System and method for determining recommendation information for service requests

Info

Publication number: CN113924460A
Application number: CN201980097233.9A
Authority: CN
Inventors: 仲小伟
Original assignee: Beijing Didi Infinity Technology and Development Co Ltd
Current assignee: Beijing Didi Infinity Technology and Development Co Ltd
Priority date: 2019-06-06
Filing date: 2019-06-06
Publication date: 2022-01-11
Anticipated expiration: 2039-06-06
Also published as: WO2020243963A1; CN113924460B

Abstract

The application relates to a system and method for determining recommendation information for a service request. The system may receive a service request from a terminal device. The service request includes a target location. The system may determine a plurality of candidate road segments based on the target location. The system may also determine a plurality of correlations between the plurality of candidate road segments and the target location by using the trained correlation determination model. The system may also identify a target road segment from the plurality of candidate road segments based on the plurality of correlations.

Description

System and method for determining recommendation information for service requests

Technical Field

The present application relates generally to systems and methods for online-to-offline services and, more particularly, to systems and methods for determining recommendation information associated with service requests for online-to-offline services.

Background

Online-to-offline services (e.g., online-to-offline transportation services) utilizing internet technology are becoming increasingly popular. A system providing an online-to-offline transportation service may obtain a service request from a requester that includes a service location (e.g., a starting location, a destination) and determine recommendation information (e.g., a recommended travel route that begins or ends at the service location) for the requester. However, in some cases, the service location may be a location where the vehicle cannot be parked. To determine the recommended travel route, the system should determine a suitable location or a suitable section of road on which the vehicle corresponding to the service location may stop. Accordingly, it is desirable to provide systems and methods for determining a suitable location or a suitable road segment corresponding to a service location of a service request, thereby efficiently and accurately determining recommendation information associated with the service request.

Disclosure of Invention

One aspect of the present application relates to a system for determining recommendation information for a service request. The system may include a storage medium storing a set of instructions and a processor communicatively coupled to the storage medium. The system may receive a service request from a terminal device. The service request includes a target location. The system may determine a plurality of candidate road segments based on the target location. The system may determine a plurality of correlations between the plurality of candidate road segments and the target location by using a trained correlation determination model. The system may also identify a target road segment from the plurality of candidate road segments based on the plurality of correlations.

In some embodiments, the target location may include at least one of a starting location and/or a destination. The target road segment may correspond to a road segment associated with the target location.

In some embodiments, the system may determine a first vector representation of the target location using a first model. For each of the plurality of candidate road segments, the system may determine a second vector representation for the candidate road segment using a second model and determine a corresponding degree of correlation between the second representation and the first vector representation using the trained correlation determination model.

In some embodiments, the system may determine recommendation information associated with the service request based on the target road segment.

In some embodiments, the recommended information may include a recommended travel route that starts or ends at a road segment corresponding to the target road segment.

In some embodiments, a training process may be used to determine the trained correlation determination model. The training process may include obtaining a plurality of historical trip records. Each of the plurality of historical travel records includes a sample point and one or more sample road segments. The training process may include obtaining a plurality of samples based on the plurality of historical trip records. Each of the plurality of samples includes one of the sample point and the one or more sample segments. The training process may include extracting feature information for each of the plurality of samples. The training process may include obtaining an initial correlation determination model. The training process may include determining, for each of the plurality of samples, a sample correlation between the sample point and the sample section based on the feature information by using the initial correlation determination model. The training process may include determining whether a plurality of sample correlations corresponding to the plurality of samples satisfy a preset condition. The training process may further include designating the initial correlation determination model as the trained correlation determination model in response to the plurality of sample correlations satisfying the preset condition.

In some embodiments, the training process may further include, in response to the plurality of sample correlations not satisfying the preset condition, updating the initial correlation determination model, and repeating the step of determining whether the plurality of sample correlations satisfy the preset condition until the plurality of sample correlations satisfy the preset condition.

In some embodiments, the plurality of samples may include a plurality of positive samples and a plurality of negative samples. For each of the plurality of positive samples, the sample segment is associated with the sample point in the respective historical travel record. For each of the plurality of negative examples, the sample road segment is not correlated with the sample point in the corresponding historical travel record.

In some embodiments, the feature information of each of the plurality of samples may include first feature information of the sample point and second feature information of the sample section. The first characteristic information may include at least one of position information of the sample point, an identification of a passenger, an identification of a driver, and/or time information of the sample point. The second feature information includes at least one of an identification of the sample road segment, a road segment type, a road segment direction, a road segment included angle, and/or a frequency of use of the sample road segment.

In some embodiments, the trained relevance determination Model comprises a double tower Deep Structured Semantic Model (DSSM).

Another aspect of the application relates to a method implemented on a computing device. The computing device includes at least one processor, at least one storage medium, and a communication platform connected to a network. The method may include receiving a service request from a terminal device. The service request may include a target location. The method may include determining a plurality of candidate road segments based on the target location. The method may include determining a plurality of correlations between the plurality of candidate road segments and the target location by using a trained correlation determination model. The method may also include identifying a target road segment from the plurality of candidate road segments based on the plurality of correlations.

In some embodiments, the target location comprises at least one of a starting location and/or a destination. The target road segment may correspond to a road segment associated with the target location.

In some embodiments, the method may further include determining a first vector representation of the target location using a first model. The method may comprise, for each of the plurality of candidate road segments, determining a second vector representation for the candidate road segment using a second model, and determining a corresponding degree of correlation between the second representation and the first vector representation using the trained correlation determination model.

In some embodiments, the method may further include determining recommendation information associated with the service request based on the target road segment.

In some embodiments, the recommended information includes a recommended travel route that starts or ends at a road segment corresponding to the target road segment.

In some embodiments, a training process may be used to determine the trained correlation determination model. The training process may include obtaining a plurality of historical trip records. Each of the plurality of historical travel records includes a sample point and one or more sample road segments. The training process may include obtaining a plurality of samples based on the plurality of historical trip records. Each of the plurality of samples includes one of the sample point and the one or more sample segments. The training process may include extracting feature information for each of the plurality of samples. The training process may include obtaining an initial correlation determination model. The training process may include, for each of the plurality of samples, determining a sample correlation between the sample point and the sample section based on the feature information by using the initial correlation determination model. The training process may include determining whether a plurality of sample correlations corresponding to the plurality of samples satisfy a preset condition. The training process may include, in response to the plurality of sample correlations satisfying the preset condition, designating the initial correlation determination model as the trained correlation determination model.

In some embodiments, the training process may further include updating the initial correlation determination model in response to the plurality of sample correlations not satisfying the preset condition, and repeating the step of determining whether the plurality of sample correlations satisfy the preset condition until the plurality of sample correlations satisfy the preset condition.

In some embodiments, the feature information of each of the plurality of samples may include first feature information of the sample point and second feature information of the sample section. The first characteristic information may include at least one of position information of the sample point, an identification of a passenger, an identification of a driver, and/or time information of the sample point. The second feature information may include at least one of an identification of the sample road segment, a road segment type, a road segment direction, a road segment angle, and/or a frequency of use of the sample road segment.

In some embodiments, the trained relevance determination model comprises a two-tower Deep Structured Semantic Model (DSSM).

Another aspect of the application relates to a system for determining recommendation information for a service request. The system may include a receiving module, a candidate segment determination module, a relevance determination module, and an identification module. The receiving module may be configured to receive a service request from a terminal device. The service request includes a target location. The candidate segment determination module may be configured to determine a plurality of candidate segments based on the target location. The relevance determination module may be configured to determine a plurality of degrees of relevance between the plurality of candidate road segments and the target location by using a trained relevance determination model. The identification module may be configured to identify a target road segment from the plurality of candidate road segments based on the plurality of degrees of correlation.

In some embodiments, the target location comprises at least one of a starting location and/or a destination. The target road segment corresponds to a road segment associated with the target location.

In some embodiments, the relevance determination module may be configured to determine a first vector representation of the target location using a first model. The relevance determination module may be further configured to, for each of the plurality of candidate road segments, determine a second vector representation for the candidate road segment using the second model, and determine a corresponding relevance between the second representation and the first vector representation using the trained relevance determination model.

In some embodiments, the identification module may be further configured to determine recommendation information associated with the service request based on the target road segment.

In some embodiments, the system may further include a training module. The training module may be configured to obtain a plurality of historical trip records. Each of the plurality of historical travel records includes a sample point and one or more sample road segments. The training module may be configured to obtain a plurality of samples based on the plurality of historical trip records. Each of the plurality of samples includes one of the sample point and the one or more sample segments. The training module may be configured to extract feature information for each of the plurality of samples. The training module may be configured to obtain an initial relevance determination model. The training module may be configured to determine, for each of the plurality of samples, a sample correlation between the sample point and the sample segment based on the feature information by using the initial correlation determination model. The training module may be configured to determine whether a plurality of sample correlations corresponding to the plurality of samples satisfy a preset condition. The training module may be configured to designate the initial correlation determination model as the trained correlation determination model in response to the plurality of sample correlations satisfying the preset condition.

In some embodiments, the training module may be further configured to, in response to the plurality of sample correlations not satisfying the preset condition, update the initial correlation determination model and repeat the step of determining whether the plurality of sample correlations satisfy the preset condition until the plurality of sample correlations satisfy the preset condition.

In some embodiments, the trained relevance determination model may include a two-tower Deep Structured Semantic Model (DSSM).

Yet another aspect of the application relates to a non-transitory computer-readable medium comprising instructions for execution. The execution instructions, when executed by at least one processor, instruct the at least one processor to perform a method. The method may include receiving a service request from a terminal device. The service request includes a target location. The method may include determining a plurality of candidate road segments based on the target location. The method may include determining a plurality of correlations between the plurality of candidate road segments and the target location by using a trained correlation determination model. The method may also include identifying a target road segment from the plurality of candidate road segments based on the plurality of correlations.

Additional features of a portion of the present application may be set forth in the description that follows. Additional features of some aspects of the present application will be apparent to those of ordinary skill in the art in view of the following description and accompanying drawings, or in view of the operation or manufacture of the embodiments. The features of the present application may be realized and obtained by the practice or use of various aspects of the methods, instrumentalities and combinations set forth in the detailed examples discussed below.

Drawings

The present application will be further described by way of exemplary embodiments. These exemplary embodiments will be described in detail by means of the accompanying drawings. These embodiments are non-limiting exemplary embodiments in which like reference numerals represent similar structures throughout the several views:

FIG. 1 is a schematic diagram of an exemplary online-to-offline service system, shown in accordance with some embodiments of the present application;

FIG. 2 is a schematic diagram of exemplary hardware and/or software components of an exemplary computing device shown in accordance with some embodiments of the present application;

FIG. 3 is a schematic diagram of exemplary hardware and/or software components of an exemplary mobile device shown in accordance with some embodiments of the present application;

FIG. 4 is a block diagram of an exemplary processing device shown in accordance with some embodiments of the present application;

FIG. 5 is a flow diagram of an exemplary process for determining recommendation information associated with a service request, according to some embodiments of the present application;

FIG. 6 is a schematic diagram of an exemplary process for determining a recommended travel route associated with a service request according to some embodiments of the present application;

FIG. 7 is a flow diagram of an exemplary process for determining a trained relevance determination model according to some embodiments of the present application; and

FIG. 8 is a diagram illustrating an exemplary structure of a relevance determination model according to some embodiments of the present application.

Detailed Description

The following description is presented to enable any person skilled in the art to make and use the application, and is provided in the context of a particular application and its requirements. It will be apparent to those skilled in the art that various modifications to the disclosed embodiments are possible, and that the general principles defined in this application may be applied to other embodiments and applications without departing from the spirit and scope of the application. Thus, the present application is not limited to the embodiments shown, but is to be accorded the widest scope consistent with the claims.

The terminology used in the description presented herein is for the purpose of describing particular example embodiments only and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, components, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, components, and/or groups thereof.

These and other features and characteristics of the present application, as well as the methods of operation, various components of the described systems, functions of structurally related elements, and combinations of parts and economies of manufacture, may become more apparent. It is to be understood, however, that the drawings are designed solely for the purposes of illustration and description and are not intended as a definition of the limits of the application. It should be understood that the drawings are not to scale.

The flow charts used in this application illustrate the operation of a system implemented according to some embodiments of the present application. It should be expressly understood that the operations of the flow diagrams may be performed out of order. Rather, various steps may be processed in reverse order or simultaneously. Also, one or more other operations may be added to the flowcharts. One or more operations may also be deleted from the flowchart.

Further, while the systems and methods disclosed in this application are primarily directed to an online-to-offline transportation service, it should also be understood that this is merely one exemplary embodiment. The systems and methods of the present application may be applied to any other type of on-demand service. For example, the systems and methods of the present application may be applied to transportation systems in different environments including land (e.g., on or off highway), water (e.g., river, lake, or ocean), air, aerospace, and the like, or any combination thereof. The vehicles of the transportation system may include taxis, private cars, pick-up cars, buses, trains, bullet trains, high speed railways, subways, boats, ships, airplanes, space vehicles, hot air balloons, unmanned vehicles, and the like, or any combination thereof. The transport system may also include any transport system for managing and/or distributing, for example, systems for sending and/or receiving courier. Applications of the systems and methods of the present application may include mobile device (e.g., smartphone or pad) applications, web pages, browser plug-ins, clients, customization systems, internal analytics systems, artificial intelligence robots, and the like, or any combination thereof.

The terms "passenger," "requestor," "service requestor," and "customer" are used interchangeably in this application to refer to an individual, entity, or tool that can request or subscribe to a service. Further, the terms "driver," "provider," "service provider," and "provider" are used interchangeably herein to refer to an individual, entity, or tool that can provide a service or facilitate providing a service. The term "user" is used in this application to refer to an individual, entity, or tool that can request a service, subscribe to a service, provide a service, or facilitate the provision of a service. In this application, the terms "requestor" and "requesting terminal" may be used interchangeably, and the terms "provider" and "provider terminal" may be used interchangeably.

The terms "request," "service request," and "order" are used interchangeably herein to refer to a request that may be initiated by a passenger, a requester, a service requester, a customer, a driver, a provider, a service provider, a supplier, etc., or any combination thereof. Depending on the context, the service request may be accepted by any of the passenger, requestor, service requestor, customer, driver, provider, service provider, or provider. In some embodiments, the driver, provider, service provider, or supplier accepts the service request. The service request may be for a fee or may be free of charge.

The Positioning technology used in the present application may be based on Global Positioning System (GPS), Global Navigation Satellite System (GLONASS), Compass Navigation System (Compass), galileo Positioning System, Quasi-Zenith Satellite System (QZSS), Wireless Fidelity (WiFi) Positioning technology, or any combination thereof. One or more of the above positioning systems may be used interchangeably in this application.

One aspect of the present application relates to systems and methods for determining recommended information (e.g., recommended travel routes, estimated arrival times) associated with service requests for online-to-offline services (e.g., network appointment services). When a passenger sends a service request to an online-to-offline service system, the system may receive the service request from the passenger's terminal device. The service request includes a target location (e.g., starting location, destination) for the intended service. Based on the target location, the system may determine a plurality of candidate road segments (e.g., road segments within a predetermined range of the target location). The system may also determine a plurality of correlations between the plurality of candidate road segments and the target location by using a trained correlation determination model (e.g., a two-tower deep structured semantic model). Further, the system may identify a target road segment from the plurality of candidate road segments based on the plurality of correlations. Further, the system may determine a recommended travel route that begins or ends at a road segment corresponding to the target road segment. According to the system and the method, the target road section is identified from the plurality of candidate road sections according to the correlation between the candidate road sections and the target position determined based on the trained correlation determination model, so that the accuracy and the efficiency of determination of the recommendation information associated with the service request are improved.

It should be noted that online-to-offline transportation services, such as online car booking services including combined services of internet car booking, are a new service form that only grows in the post-internet era. It provides users and service providers with a technical solution that can only be proposed in the late internet era. In the former internet era, when a passenger calls a taxi on the street, taxi requests and receptions only occur between the passenger and the taxi driver who sees the passenger. If the passenger calls a taxi by telephone, the service request and acceptance can only be made between the passenger and a service provider (e.g., a taxi company or agent). However, online taxis allow a user of a service to automatically distribute service requests to a large number of service providers (e.g., taxis) remote from the user in real time. It also allows multiple service providers to respond to service requests simultaneously and in real time. Thus, over the internet, an online-to-offline transportation system may provide a more efficient trading platform for users and service providers, which is never encountered in conventional prior internet transportation service systems.

FIG. 1 is a schematic diagram of an exemplary online-to-offline service system, shown in accordance with some embodiments of the present application. In some embodiments, the online-to-offline service system 100 may be an online transport service platform for transport services, such as net appointments, driver services, delivery vehicles, express, carpools, bus services, driver rentals, regular service, and the like. The online-to-offline service system 100 may include a server 110, a network 120, a requester terminal 130, a provider terminal 140, and a storage device 150.

In some embodiments, the server 110 may be a single server or a group of servers. The set of servers can be centralized or distributed (e.g., the servers 110 can be a distributed system). In some embodiments, the server 110 may be local or remote. For example, server 110 may access information and/or data stored in requester terminal 130, provider terminal 140, and/or storage device 150 via network 120. As another example, server 110 may be directly connected to requester terminal 130, provider terminal 140, and/or storage device 150 to access stored information and/or data. In some embodiments, the server 110 may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-tiered cloud, and the like, or any combination thereof. In some embodiments, server 110 may be implemented on a computing device 200 that includes one or more of the components shown in fig. 2.

In some embodiments, the server 110 may include a processing device 112. Processing device 112 may process information and/or data related to the service request to perform one or more of the functions described herein. For example, the processing device 112 may identify a target road segment associated with the service request by using the trained relevance determination model and determine recommendation information (e.g., recommended driving route, estimated arrival time) related to the service request according to the target road segment. In some embodiments, processing device 112 may include one or more processing devices (e.g., a single core processing device or a multi-core processor device). The Processing Device 112 may include a Central Processing Unit (CPU), an Application-Specific Integrated Circuit (ASIC), an Application-Specific Instruction Set Processor (ASIP), a Graphics Processing Unit (GPU), a Physical Processing Unit (PPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), a controller, a microcontroller Unit, a Reduced Instruction Set Computer (Reduced Instruction Set Computer, RISC), a microprocessor, or the like, or any combination thereof. In some embodiments, the processing device 112 may be integrated in the requester terminal 130 or the provider terminal 140.

Network 120 may facilitate information and/or data exchange. In some embodiments, one or more components of the online-to-offline service system 100 (e.g., the server 110, the requester terminal 130, the provider terminal 140, or the storage device 150) may send information and/or data to other components of the online-to-offline service system 100 via the network 120. For example, the server 110 may obtain a service request from the requester terminal 130 via the network 120. In some embodiments, the network 120 may be any type of wired or wireless network, or any combination thereof. Merely by way of example, Network 120 may include a cable Network, a wired Network, a fiber optic Network, a telecommunications Network, an intranet, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Public Switched Telephone Network (PSTN), a bluetooth Network, a zigbee Network, a Near Field Communication (NFC) Network, and the like, or any combination thereof. In some embodiments, network 120 may include one or more network access points. For example, network 120 may include wired or wireless network access points, such as base stations and/or Internet switching points 120-1, 120-2, …, through which one or more components of online-to-offline service system 100 may connect to network 120 to exchange data and/or information.

In some embodiments, the service requester may be a user of the requester terminal 130. In some embodiments, the user of requester terminal 130 may be a person other than the service requester. For example, user a of the requesting terminal 130 may use the requester terminal 130 to send a service request for user B or to receive a service confirmation and/or information or instructions from the server 110. In some embodiments, the service provider may be a user of the provider terminal 140. In some embodiments, the user of provider terminal 140 may be a person other than a service provider. For example, user C of provider terminal 140 may receive a service request and/or information or instructions from server 110 for user D using provider terminal 140.

In some embodiments, the requester terminal 130 may include a mobile device 130-1, a tablet computer 130-2, a laptop computer 130-3, a built-in device in a vehicle 130-4, etc., or any combination thereof. In some embodiments, the mobile device 130-1 may include a smart home device, a wearable device, a smart mobile device, a virtual reality device, an augmented reality device, or the like, or any combination thereof. In some embodiments, the smart home devices may include smart lighting devices, smart appliance control devices, smart monitoring devices, smart televisions, smart cameras, interphones, and the like, or any combination thereof. In some embodiments, the wearable device may include a smart bracelet, a smart footwear, smart glasses, a smart helmet, a smart watch, a smart garment, a smart backpack, a smart accessory, or the like, or any combination thereof. In some embodiments, the smart mobile device may include a smart phone, a Personal Digital Assistant (PDA), a gaming device, a navigation device, a Point of Sale (POS) device, and the like, or any combination thereof. In some embodiments, the virtual reality device and/or the augmented reality device may include a virtual reality helmet, virtual reality glasses, virtual reality eyeshields, augmented reality helmets, augmented reality glasses, augmented reality eyeshields, and the like, or any combination thereof. For example, virtual reality devicesThe alternate and/or augmented reality device may comprise a Google Glass^TM、Oculus Rift^TM、Hololens^TM、Gear VR^TMAnd the like. In some embodiments, the built-in devices in the vehicle 130-4 may include an on-board computer, an on-board television, and the like. In some embodiments, the requester terminal 130 may be a device having a location technology for locating the location of the service requester and/or the requester terminal 130.

In some embodiments, provider terminal 140 may be similar to requester terminal 130, or the same device as requester terminal 130. In some embodiments, provider terminal 140 may be a device having a location technology for locating a service provider and/or the location of provider terminal 140. In some embodiments, the requester terminal 130 and/or the provider terminal 140 may communicate with other location devices to determine the location of the service requester, requester terminal 130, service provider and/or provider terminal 140. In some embodiments, the requester terminal 130 and/or the provider terminal 140 may send location information to the server 110.

The storage device 150 may store data and/or instructions related to the service request. In some embodiments, storage device 150 may store data obtained from requester terminal 130 and/or provider terminal 140. In some embodiments, storage device 150 may store data and/or instructions that server 110 uses to perform or use to perform the exemplary methods described in this application. In some embodiments, storage device 150 may include mass storage, removable storage, volatile Read-and-write Memory, Read-Only Memory (ROM), and the like, or any combination thereof. Exemplary mass storage devices may include magnetic disks, optical disks, solid state drives, and the like. Exemplary removable memories may include flash drives, floppy disks, optical disks, memory cards, compact disks, magnetic tape, and the like. Exemplary volatile read and write Memory may include Random Access Memory (RAM). Exemplary RAMs may include Dynamic Random Access Memory (DRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Static random access memory (Static RAM, SRAM), Thyristor random access memory (Thyristor RAM, T-RAM), and Zero-capacitance random access memory (Zero-Capacitor RAM, Z-RAM), among others. Exemplary ROMs may include Mask ROMs (MROMs), Programmable ROMs (PROMs), Erasable Programmable ROMs (EPROMs), Electrically Erasable Programmable ROMs (EEPROMs), Compact Disk ROMs (CD-ROMs), digital versatile disks ROMs, and the like. In some embodiments, the storage device 150 may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-tiered cloud, and the like, or any combination thereof.

In some embodiments, a storage device 150 may be connected to network 120 to communicate with one or more components of online-to-offline service system 100 (e.g., server 110, requester terminal 130, provider terminal 140). One or more components of the online-to-offline service system 100 may access data and/or instructions stored in the storage device 150 via the network 120. In some embodiments, the storage device 150 may be directly connected to or in communication with one or more components of the online-to-offline service system 100 (e.g., the server 110, the requester terminal 130, the provider terminal 140). In some embodiments, the storage device 150 may be part of the server 110.

In some embodiments, one or more components of the online-to-offline service system 100 (e.g., the server 110, the requestor terminal 130, the provider terminal 140) may have access to the storage device 150. In some embodiments, one or more components of the online-to-offline service system 100 may read and/or modify information related to the service requester, the service provider, and/or the public when one or more conditions are satisfied. For example, the server 110 may read and/or modify information of one or more service requesters after the service is completed. For another example, the provider terminal 140 may access information related to the service requester when receiving the service request from the requester terminal 130, but the provider terminal 140 may not modify the information related to the service requester.

In some embodiments, the exchange of information by one or more components of the online-to-offline service system 100 may be accomplished by requesting a service. The object of the service request may be any product. In some embodiments, the product may be a tangible product or a non-physical product. Tangible products may include food, pharmaceuticals, commodities, chemical products, appliances, clothing, automobiles, homes, luxury goods, and the like, or any combination thereof. The non-material products may include service products, financial products, knowledge products, internet products, and the like, or any combination thereof. The internet products may include personal host products, website products, mobile internet products, commercial host products, embedded products, and the like, or any combination thereof. The mobile internet product may be software, a program, a system, etc. for a mobile terminal or any combination thereof. The mobile terminal may include a tablet computer, laptop computer, mobile phone, Personal Digital Assistant (PDA), smart watch, POS device, vehicle computer, vehicle television, wearable device, and the like, or any combination thereof. The product may be, for example, any software and/or application used on a computer or mobile phone. The software and/or applications may be related to social interaction, shopping, transportation, entertainment, learning, investment, etc., or any combination thereof. In some embodiments, transportation-related system software and/or applications may include travel software and/or applications, vehicle scheduling software and/or applications, mapping software and/or applications, and/or the like. In the vehicle scheduling software and/or application, the vehicle may include a horse, a carriage, a human powered vehicle (e.g., bicycle, tricycle), a car (e.g., taxi, bus, private car), a train, a subway, a ship, an aircraft (e.g., airplane, helicopter, space shuttle, rocket, hot air balloon), and the like, or any combination thereof.

One of ordinary skill in the art will appreciate that when an element (or component) of the inline-to-offline service system 100 executes, the element may execute via electrical and/or electromagnetic signals. For example, when the requester terminal 130 sends a service request to the server 110, the processor of the requester terminal 130 may generate an electrical signal encoding the request. The processor of the requesting terminal 130 may then send the electrical signal to the output port. If the requester terminal 130 communicates with the server 110 via a wired network, the output port may be physically connected to a cable, which may further transmit electrical signals to the input port of the server 110. If the requester terminal 130 communicates with the server 110 via a wireless network, the output port of the requester terminal 130 may be one or more antennas that convert electrical signals to electromagnetic signals. Similarly, provider terminal 140 may process tasks through operation of logic circuits in its processor and receive instructions and/or service requests from server 110 via electrical or electromagnetic signals. In an electronic device, such as requester terminal 130, provider terminal 140, and/or server 110, when its processor processes an instruction, sends an instruction, and/or performs an action, the instruction and/or action is performed via an electrical signal. For example, when the processor retrieves or saves data from a storage medium (e.g., storage device 150), it may send electrical signals to the storage medium's read/write device, which may read or write structured data in the memory. The structured data may be transmitted to the processor in the form of electrical signals over a bus of the electronic device. Herein, an electrical signal may refer to one electrical signal, a series of electrical signals, and/or a plurality of discrete electrical signals.

It should be noted that the application scenario shown in fig. 1 is for illustrative purposes only, and is not intended to limit the scope of the present application. For example, the online-to-offline service 100 may be used as a navigation system. The navigation system may include a user terminal (e.g., requester terminal 130 or provider terminal 140) and a server (e.g., server 110). The user may input a target location (e.g., a start location, a destination) via the user terminal. Thus, the navigation system may determine recommended information (e.g., recommended travel route, estimated time of arrival) based on the target location according to the processes and/or methods described herein.

FIG. 2 is a schematic diagram of exemplary hardware and/or software components of an exemplary computing device shown in accordance with some embodiments of the present application. In some embodiments, server 110, requester terminal 130, and/or provider terminal 140 may be implemented on computing device 200. For example, the processing device 112 may be implemented on the computing device 200 and configured to perform the functions of the processing device 112 disclosed herein.

Computing device 200 may be used to implement any of the components of online-to-offline service system 100 as described herein. For example, the processing device 112 may be implemented on the computing device 200 by its hardware, software programs, firmware, or a combination thereof. Although only one such computing device is shown, for convenience, the online-to-offline service-related computer functions described herein may be implemented in a distributed manner across multiple similar platforms to spread out processing load.

The computing device 200 may include a serial Communication (COM) port 250 that connects to or from a network to facilitate data communication. Computing device 200 may also include a processor 220 in the form of one or more processors (e.g., logic circuits) for executing program instructions. For example, the processor 220 may include interface circuitry and processing circuitry therein. The interface circuit may be configured to receive electrical signals from bus 210. The electrical signals encode structured data and/or instructions that are processed by the processing circuitry. The processing circuitry may perform logical computations and then encode the conclusions, results and/or instructions into electrical signals. The interface circuit may then send the electrical signals from the processing circuit via bus 210.

Computing device 200 may also include different forms of program storage and data storage including, for example, a disk 270, Read Only Memory (ROM)230, or Random Access Memory (RAM)240 for storing various data files processed and/or transmitted by computing device 200. Computing device 200 may also include program instructions stored in ROM 230, RAM 240, and/or other types of non-transitory storage media that are executed by processor 220. The methods and/or processes of the present application may be embodied as program instructions. Computing device 200 may also include input/output (I/O) components 260 that support input/output between the computer and other components. Computing device 200 may also receive programming and data via network communications.

For illustration only, only one processor is depicted in fig. 2. Multiple processors are also contemplated, and thus, operations and/or steps performed by one processor as described herein may also be performed by multiple processors, either jointly or separately. For example, if in the present application the processors of computing device 200 perform steps a and b, it should be understood that steps a and b may also be performed jointly or separately by two different CPUs and/or processors (e.g., a first processor performing step a, a second processor performing step b, or a first and second processor performing steps a and b jointly).

Fig. 3 is a schematic diagram of exemplary hardware and/or software components of an exemplary mobile device shown in accordance with some embodiments of the present application. In some embodiments, the requester terminal 130 or the provider terminal 140 may be implemented on a mobile device 300. As shown in fig. 3, mobile device 300 may include a communication platform 310, a display 320, a Graphics Processing Unit (GPU)330, a Central Processing Unit (CPU)340, I/O350, a storage unit 360, a mobile Operating System (OS) 370, and a memory 390. In some embodiments, any other suitable component, including but not limited to a system bus or a controller (not shown), may also be included in mobile device 300.

In some embodiments, the operating system 370 is mobile (e.g., iOS)^TM、Android^TM、Windows Phone^TM) And one or more application programs 380 may be loaded from storage 390 into memory 360 for execution by CPU 340. The application 380 may include a browser or any other suitable mobile application for receiving and presenting information associated with an online-to-offline service or other information from the online-to-offline service system 100. User interaction with the information flow may be accomplished via I/O350 and provided to processing device 112 and/or other components of online-to-offline service system 100 via network 120.

FIG. 4 is a block diagram of an exemplary processing device shown in accordance with some embodiments of the present application. The processing device 112 may include a receiving module 410, a candidate segment determination module 420, a relevance determination module 430, a recognition module 440, and a training module 450.

The receiving module 410 may be configured to receive a service request from a terminal device (e.g., the requester terminal 130) via the network 120. The service request may be a request for a transportation service (e.g., a taxi service, a courier service, a network appointment service). The service request may include a target location, e.g., a starting location, a destination, etc.

The candidate segment determination module 420 may be configured to determine a plurality of candidate segments based on the target location. In some embodiments, the candidate segment determination module 420 may identify a plurality of available segments within a predetermined range of the target location and determine a plurality of distances (e.g., straight-line distances, road distances) between the plurality of available segments and the target location, respectively. Further, the candidate segment determination module 420 may determine available segments that are less than a distance threshold from the target location as the plurality of candidate segments. In some embodiments, the candidate road segment determination module 420 may obtain a plurality of historical usage road segments associated with the target location based on historical travel records over a predetermined period of time (e.g., the last three months) and determine a plurality of usage frequencies for the plurality of historical usage road segments, respectively. Further, the candidate segment determination module 420 may determine historical usage segments with usage frequencies above a frequency threshold as the plurality of candidate segments.

The relevance determination module 430 may be configured to determine a plurality of degrees of relevance between the plurality of candidate road segments and the target location by using the trained relevance determination model. In some embodiments, for each of the plurality of candidate road segments, to determine a correlation between the candidate road segment and the target location, the correlation determination module 430 may extract a first feature of the target location (e.g., location information) and a second feature of the candidate road segment (e.g., a road segment type, a distance between the candidate road segment and the target location). The relevance determination module 430 may also determine a first feature vector corresponding to the target location based on the first feature and a second feature vector corresponding to the candidate road segment based on the second feature. In addition, the relevance determination module 430 may determine the relevance between the candidate road segment and the target location based on the first feature vector and the second feature vector by using a trained relevance determination model. In some embodiments, the relevance determination module 430 can obtain a trained relevance determination model from the training module 450 or a storage device disclosed elsewhere in this application (e.g., storage device 150).

The identification module 440 may be configured to identify a target road segment from a plurality of candidate road segments based on a plurality of degrees of correlation. In some embodiments, the identification module 440 may select the target road segment from a plurality of candidate road segments based on a predetermined rule. For example, the identification module 440 may rank the plurality of candidate road segments from high to low based on the plurality of degrees of correlation. In addition, the identification module 440 may select the candidate segment with the highest degree of correlation as the target segment. In some embodiments, after identifying the target road segment, the identification module 440 may also determine recommendation information associated with the service request based on the target road segment. In some embodiments, the recommendation information may include a recommended travel route that starts or ends at a link corresponding to the target link, an Estimated Time Of Arrival (ETA) Of the service request, and the like.

The training module 450 may be configured to determine a trained correlation determination model based on the plurality of samples according to a training process. In some embodiments, training module 450 may obtain a plurality of historical trip records. Each of these historical travel records includes a sample point and one or more sample road segments. Further, the training module 450 may obtain a plurality of samples based on a plurality of historical trip records. Each of the samples includes one of a sample point and one or more sample segments. Further description of the training process may be found elsewhere in the application (e.g., fig. 7 and its description).

In some embodiments, the processing device 112 may also include a transmission module (not shown) that may be configured to transmit the recommendation information to the requester terminal 130 and/or the provider terminal 140 via the network 120 or to save the recommendation information in a storage device disclosed elsewhere herein (e.g., storage device 150).

The modules in the processing device 112 may be connected or communicate with each other through wired or wireless connections. The wired connection may include a metal cable, an optical cable, a hybrid cable, and the like, or any combination thereof. The wireless connection may include a Local Area Network (LAN), a Wide Area Network (WAN), bluetooth, zigbee network, Near Field Communication (NFC), etc., or any combination thereof. Two or more modules may be combined into a single module. Any of the modules may be divided into two or more units. For example, the receiving module 410 and the candidate segment determination module 420 may be combined into a single module that may receive a service request and determine a plurality of candidate segments. As another example, the processing device 112 may include a storage module (not shown) for storing information and/or data associated with the service request (e.g., a plurality of candidate road segments, a plurality of correlations, a target road segment, recommendation information). As another example, the training module 450 may not be necessary, and the trained relevance determination model may be obtained from a storage device disclosed elsewhere in this application (e.g., storage device 150), or may be determined by a separate training device in the online-to-offline service 100.

FIG. 5 is a flow diagram of an exemplary process for determining recommendation information associated with a service request, according to some embodiments of the present application. In some embodiments, process 500 may be implemented by a set of instructions (e.g., an application program) stored in ROM 230 or RAM 240. Processor 220 and/or the modules in fig. 4 may execute a set of instructions, and when executing the instructions, processor 220 and/or the modules may be configured to perform process 500. The operation of the process shown below is for illustration purposes only. In some embodiments, process 500 may be accomplished with one or more additional operations not described, and/or without one or more operations discussed herein. Additionally, the order of the operations of the process as shown in FIG. 5 and described below is not intended to be limiting.

In 510, the processing device 112 (e.g., the receiving module 410) (e.g., the interface circuitry of the processor 220) may receive a service request from a terminal device (e.g., the requester terminal 130) via the network 120.

In some embodiments, the service request may be a request for any location-based service. In some embodiments, the service request may be a request for a transportation service (e.g., a taxi service, a courier service, a network appointment service). The service request may be a real-time request, a reservation request, etc., or any combination thereof. As used herein, a real-time request may include a service that the requestor desires to receive at the current time or at a defined time near the current time. For example, the service request may be a real-time request if the defined time is within a time period less than a time threshold from the current time, e.g., 5 minutes from the current time, 10 minutes from the current time, 20 minutes from the current time, etc. The reservation request may include a service that the requester desires to receive at a defined time that is remote from the current time. For example, the service request may be a reservation request if the defined time is within a time period greater than a time threshold from the current time, e.g., 25 minutes from the current time, 2 hours from the current time, 1 day from the current time, etc. The time threshold may be a default setting for the online-to-offline service system 100 or may be adjusted according to different circumstances. For example, at peak traffic times, the time threshold may be relatively small (e.g., 10 minutes), while at off-peak times (e.g., 10:00-12:00a.m.), the time threshold may be relatively large (e.g., 1 hour).

In some embodiments, the service request may include a target location, e.g., a starting location, a destination, etc. As used herein, a starting location generally refers to a location where a requestor wishes to begin service (e.g., a location that the requestor wishes to be received by a service provider). The destination generally refers to a location where the requestor wishes to end the service (e.g., a location where the requestor wishes to be dropped by the service provider). In some embodiments, the starting location may be the current location of the requester terminal 130 or any location defined by the requester. In some embodiments, the starting location and/or destination may be obtained in various ways, including but not limited to manual input by the requester terminal 130, selection from historical input records, selection from system recommendations, use of location techniques, and the like, or any combination thereof. In some embodiments, the starting location and/or destination may be represented as a description of the location, an address of the location, latitude and longitude coordinates of the location, a point corresponding to the location in a map, or the like, or any combination thereof.

In 520, the processing device 112 (e.g., the candidate road segment determination module 420) (e.g., the processing circuitry of the processor 220) may determine a plurality of candidate road segments (e.g., 5, 10, 15) based on the target location. The term "road segment" as used herein may refer to an element (e.g., a section of a road) of a road or street in a map. As described above, the target location may include a starting location or a destination; accordingly, the candidate segment may be a candidate start segment corresponding to the start position or a candidate end segment corresponding to the destination. The term "start section" refers to a section of a service request where a driving route starts, which includes a boarding location corresponding to a starting location; the term "end section" refers to a section where a travel route ends in a service request, which includes a get-off position corresponding to a destination. As used herein, a drop-off location generally refers to a location where a vehicle may stop and then get on a service object (e.g., a requester or cargo). The boarding location may be the same as or different from the starting location. For example, when the start position is a position where the vehicle cannot stop, the online-to-offline service system 100 may determine an appropriate position near the start position as the boarding position. An alighting location generally refers to a location at which the vehicle may stop to allow a transport service object (e.g., a requester and/or cargo) to alight from the vehicle. The drop-off location may be the same or different from the destination. For example, when the destination is a location where the vehicle cannot stop, the online-to-offline service system 100 may determine a suitable location near the destination as the drop-off location.

In some embodiments, the processing device 112 may identify a plurality of available road segments within a predetermined range (e.g., 50 meters, 100 meters, 200 meters) of the target location and determine a plurality of distances (e.g., straight-line distances, road distances) between the plurality of available road segments and the target location, respectively. Further, the processing device 112 may determine available road segments having a distance to the target location less than a distance threshold as the plurality of candidate road segments. The distance threshold may be a default setting for the online-to-offline service system 100, or may be adjusted in different circumstances.

In some embodiments, the processing device 112 may obtain a plurality of historical usage road segments associated with the target location based on historical travel records within a predetermined period (e.g., the last three months) and determine a frequency of usage of the plurality of historical usage road segments, respectively. As used herein, a historical use road segment refers to a road segment where a historical travel route of a historical service order associated with a target location (e.g., a historical service with a historical start location or historical destination that is the same as the target location) begins or ends. The frequency of use of the specific historical use segment refers to the number of historical travel routes (corresponding to the historical service orders associated with the target location) that start or end at the specific historical use segment, or the ratio of the number of historical travel routes that start or end at the specific historical use segment to the total number of historical travel routes for the historical service orders associated with the target location. Further, the processing device 112 may determine historical usage road segments with a usage frequency above a frequency threshold as a plurality of candidate road segments. The frequency threshold may be a default setting for the online-to-offline service system 100, or may be adjusted in different circumstances.

In some embodiments, the processing device 112 may also consider reference information (e.g., traffic information associated with the target location, preference information of the requester) when determining the plurality of candidate road segments. For example, the processing device 112 may filter out available road segments having poor traffic conditions or historically used road segments. As another example, the processing device 112 may prioritize road segments that meet the requester's preferences (e.g., do not cross roads).

At 530, the processing device 112 (e.g., the relevance determination module 430) (e.g., processing circuitry of the processor 220) may determine a plurality of degrees of relevance between the plurality of candidate road segments and the target location by using the trained relevance determination model.

As used herein, taking a particular candidate segment (we may assume that the particular candidate segment is a candidate start segment) as an example, the relevance may indicate a probability that the particular candidate segment (or a particular candidate pick-up point included in the particular candidate segment) is likely to be selected by the requester as a pick-up point for the service request. The greater the degree of correlation, the higher the probability.

In some embodiments, for each of the plurality of candidate road segments, to determine a degree of correlation between the candidate road segment and the target location, the processing device 112 may extract a first feature of the target location and a second feature of the candidate road segment. The processing device 112 may also determine a first feature vector corresponding to the target location based on the first feature and a second feature vector corresponding to the candidate road segment based on the second feature. Further, by using the trained relevance determination model, the processing device 112 may determine a relevance between the candidate road segment and the target location based on the first feature vector and the second feature vector.

In some embodiments, the first characteristic of the target location may include location information of the target location (e.g., market, school, office building, hospital), an identification of the requester (e.g., passenger), an identification of the service provider that has accepted the service request (e.g., driver), GPS information associated with the requester (e.g., GPS information uploaded by requester terminal 130), GPS information associated with the service provider (e.g., GPS information uploaded by provider terminal 140), time information (e.g., time point when the service request originated, weekday, or weekend), or the like, or a combination thereof. The second characteristic of the candidate road segment may include an identification of the candidate road segment, a type of road segment (e.g., highway, overpass, tunnel), a speed of the road segment (also may be referred to as "road segment traffic condition"), a road segment angle, GPS information associated with the candidate road segment (e.g., GPS information uploaded by provider terminals 140 located on the candidate road segment), a distance between the candidate road segment and the target location, a frequency of use of the candidate road segment, or the like, or a combination thereof.

In some embodiments, the processing device 112 may obtain the trained relevance determination model from the training module 450 or a storage device disclosed elsewhere in this application (e.g., storage device 150). In some embodiments, a trained relevance determination model may be determined based on a plurality of samples associated with a plurality of historical trip records. In some embodiments, the trained relevance determination Model may be a double tower Deep Structured Semantic Model (DSSM). More descriptions of the trained relevance determination model may be found elsewhere in the application (e.g., fig. 7 and its description).

In some embodiments, the processing device 112 may determine a first vector representation (also referred to as a "point vector") of the target location based on the first feature (or first feature vector) using the first model, and determine a second vector representation (also referred to as a "road segment vector") of the candidate road segment based on the second feature (or second feature vector) using the second model. Further, the processing device 112 may determine a corresponding degree of correlation between the second vector representation and the first vector representation by using a trained degree of correlation determination model. In some embodiments, the first model and the second model may be separate parts (e.g., the point network and the road segment network shown in fig. 8) included in a trained relevance determination model (e.g., the dual-tower DSSM shown in fig. 8). In some embodiments, the first model, the second model, and the trained relevance determining model are independent models that may be trained separately. For example, the first model may be a first DSSM, the second model may be a second DSSM, and the trained correlation determination model may be a classifier (e.g., a binary classifier).

In 540, the processing device 112 (e.g., the identification module 440) (e.g., the processing circuitry of the processor 220) may identify a target road segment from a plurality of candidate road segments based on a plurality of degrees of correlation.

The candidate segment may be a candidate start segment corresponding to a start position or a candidate end segment corresponding to a destination, as described in operation 520. Accordingly, the target section may be a target start section corresponding to a start position including a target getting-on position or a target end section corresponding to a destination including a target getting-off position.

In some embodiments, the processing device 112 may select a target road segment from a plurality of candidate road segments based on a predetermined rule. For example, the processing device 112 may rank the plurality of candidate road segments from high to low based on the plurality of degrees of correlation. Further, the processing device 112 may select the candidate segment with the highest degree of correlation as the target segment.

In some embodiments, after identifying the target road segment, the processing device 112 (e.g., the processing circuitry of the processor 220) may also determine recommendation information associated with the service request based on the target road segment. In some embodiments, the recommendation information may include a recommended driving route starting or ending from a road segment opposite the target road segment, an Estimated Time of Arrival (ETA) of the service request, and the like. For example, the processing device 112 may determine a recommended travel route based on the starting location, the target starting segment, the target ending segment, and the destination. For another example, the processing device 112 may determine the ETA based on the recommended travel route and/or traffic information (e.g., traffic speed, traffic flow, traffic density) associated with the service request.

In some embodiments, the processing device 112 (e.g., interface circuitry of the processor 220) may transmit the recommendation information to the requester terminal 130 and/or the provider terminal 140 via the network 120. In some embodiments, the processing device 112 may save the recommendation information to a storage device (e.g., storage device 150) disclosed elsewhere in this application.

It should be noted that the above description is provided for illustrative purposes only, and is not intended to limit the scope of the present application. Various changes and modifications will occur to those skilled in the art based on the description herein. However, such changes and modifications do not depart from the scope of the present application. For example, one or more other optional operations (e.g., a store operation) may be added elsewhere in process 500. In a storage operation, the processing device 112 may store information and/or data associated with the service request (e.g., a plurality of candidate road segments, a plurality of degrees of association, the target road segment, the recommendation information) in a storage device (e.g., storage device 150) disclosed elsewhere in this application. As another example,

operations

510 and 520 may be combined into a single operation in which the processing device 112 may receive a service request and determine a plurality of candidate road segments.

FIG. 6 is a schematic diagram of an exemplary process for determining a recommended travel route associated with a service request according to some embodiments of the present application. As shown, the service request includes a start location S and a destination D. As illustrated in fig. 5, after receiving the service request, the processing device 112 may determine a plurality of candidate start segments based on the start position S and a plurality of candidate end segments based on the destination D. Further, the processing device 112 may identify the target start link L corresponding to the start position S based on a plurality of degrees of correlation between a plurality of candidate start links and the start position S_S(ii) a And based on a plurality of facies between a plurality of candidate end segments and the destination DRelevance determination of target end link L corresponding to destination D_D. The processing device 112 may also start the road segment L based on the start position S, the target start position S_STarget end road section L_DAnd destination D determines a recommended travel route.

FIG. 7 is a flow diagram illustrating an exemplary process of determining a trained relevance determination model according to some embodiments of the present application. In some embodiments, process 700 may be implemented by a set of instructions (e.g., an application program) stored in ROM 230 or RAM 240. Processor 220 and/or training module 450 may execute the set of instructions, and upon execution of the instructions, processor 220 and/or training module 450 may be configured to perform process 700. The operation of the process shown below is for illustration purposes only. In some embodiments, process 700 may be accomplished with one or more additional operations not described, and/or without one or more operations discussed herein. Additionally, the order of the operations of the processes as shown in FIG. 7 and described below is not intended to be limiting.

At 710, the processing device 112 (e.g., the training module 450) (e.g., the interface circuitry or processing circuitry of the processor 220) may obtain a plurality of historical trip records. The processing device 112 may obtain a plurality of historical travel records from a storage device disclosed elsewhere in this application (e.g., the storage device 150, a storage module (not shown) in the processing device 112).

In some embodiments, each of the plurality of historical travel records may include service requests that have been completed (referred to as "historical service orders") and information associated therewith. For example, taking the application scenario shown in fig. 1 as an example, a requestor may send a service request to the online-to-offline service system 100 that includes a point of transportation service (e.g., a starting location, a destination). The online-to-offline service system 100 may receive the service request and determine a plurality of candidate segments (e.g., a plurality of candidate start segments, a plurality of candidate end segments) associated with the point. The online-to-offline service system 100 may further identify a start segment (which includes an entry location) from a plurality of candidate start segments and an end segment (which includes an exit location) from a plurality of candidate end segments. The online-to-offline service system 100 may also determine a travel route based on the start road segment and the end road segment. The service provider may receive the service request and provide transportation services along a travel route from the boarding location to the disembarking location. After the service provider sends the requester to the drop-off location, the online-to-offline service system 100 may store information associated with the service request (e.g., a start location, a destination, a plurality of candidate segments, a start segment, an end segment, a travel route, a pick-up location, a drop-off location) in a storage device (e.g., storage device 150) disclosed elsewhere in this application.

In some embodiments, multiple historical travel records may be selected based on a time criterion. For example, the plurality of historical travel records may be selected within a predetermined time period, e.g., the last day, the last three days, the last week, the last two weeks, the last month, the last six months, 8:00 am to 9 am of each day in six months, etc. In some embodiments, multiple historical trip records may be selected based on spatial criteria. For example, multiple historical travel records may be selected within a predetermined geographic area (e.g., city, region). In some embodiments, the plurality of historical travel records may be selected based on one or more parameters, such as "requester identity," "provider identity," "start location," "destination," "section identification," "section type," "section speed," "section angle," "section usage frequency," and the like.

In some embodiments, each of the plurality of historical travel records may include a sample point and one or more sample road segments. As used herein, a sample point refers to a historical starting location or a historical destination of a historical service order in a corresponding historical travel record. The one or more sample road segments refer to historical candidate start road segments or historical destination candidate end road segments corresponding to historical start positions in the corresponding historical travel record.

At 720, the processing device 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may obtain a plurality of samples based on a plurality of historical trip records. Each of the plurality of samples includes one of a sample point and one or more sample segments.

In some embodiments, the plurality of samples may include a plurality of positive samples (which may be labeled as "1") and a plurality of negative samples (which may be labeled as "0"). For a positive sample, the sample segment is correlated with a sample point in the corresponding historical travel record. For negative examples, the sample segment is not correlated to the sample point in the corresponding historical travel record. As used herein, "sample road segment is related to a sample point" refers to a sample road segment where the historical travel route of the historical service order associated with the sample point (i.e., the historical service order having the same historical start location or historical destination as the sample point) actually begins or ends in the historical travel record.

In some embodiments, the processing device 112 may divide the plurality of samples into a training set and a test set. The training set may be used to train the model and the test set may be used to determine whether the training process has been completed.

In 730, the processing device 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may extract a sample feature (also referred to as "feature information") for each of the plurality of samples.

In some embodiments, the sample characteristics for each of the plurality of samples may include a first sample characteristic for the sample point and a second sample characteristic for the sample segment. As set forth in operation 530, the first sample feature of the sample point may include sample location information for the sample point (e.g., market, school, office building, hospital), an identification of a history requester (e.g., passenger) that initiated a history service request associated with the sample point, an identification of a history service provider (e.g., driver) that provided history service for the history requester, time information (e.g., a history time point, weekday, or weekend at the time of initiation of the history service request), and the like, or combinations thereof. The second characteristics of the sample road segment may include an identification of the sample road segment, a type of road segment (e.g., highway, overpass, tunnel), historical road segment speed (also may be referred to as "road segment traffic conditions"), a road segment angle, a distance between the sample road segment and the sample point, a frequency of use of the sample road segment, and the like, or a combination thereof.

In 740, the processing device 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may obtain an initial correlation determination model. The initial relevance determination model may include one or more initial parameters, which may be default settings for the online-to-offline service system 100, or may be adjusted in different circumstances. In some embodiments, the initial relevance determining model may be a double-tower Deep Structured Semantic Model (DSSM). The dual-tower DSSM may include a Fully Connected (FC) layer, a Batch Normalization (BN) layer, and a BELU layer. More descriptions of the relevance determination model may be found elsewhere in this application (e.g., FIG. 8 and its description).

In 750, for each of the plurality of samples, the processing device 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may determine a sample correlation between the sample point and the sample segment based on the first sample feature and the second sample feature using an initial correlation determination model.

In some embodiments, for each of the plurality of samples, the processing device 112 may determine a first sample feature vector corresponding to the sample point by encoding the first sample feature and determine a second sample feature vector corresponding to the sample segment by encoding the second sample feature. Further, according to the initial correlation determination model (e.g., through processing of the FC layer, the BN layer, and the RELU layer), the processing device 112 may determine a sample point vector based on the first sample feature vector and a sample segment vector based on the second sample feature vector. As used herein, the dimensions of the first sample feature vector and the dimensions of the second sample feature vector may be different, and the first sample feature vector and the second sample feature vector are not comparable; in contrast, the dimensions of the sample point vector and the sample segment vector are the same, and the sample point vector and the sample segment vector have comparability.

In some embodiments, the processing device 112 may determine the dot product (inner product) of the sample point vector and the sample segment vector according to equation (1) below:

where x represents the dot product of the sample point vector and the sample segment vector,

a vector of sample points is represented, and,

represents a sample segment vector and theta represents the angle between the sample point vector and the sample segment vector.

Further, the processing device 112 may determine the sample correlation based on a dot product of the sample point vector and the sample road segment vector according to a classification method.

For example, the processing device 112 may determine the sample correlation according to the sigmoid function shown below:

where c (x) refers to the sample correlation between a sample point and a sample segment.

For another example, the processing device 112 may determine the sample correlation according to the threshold method shown below:

where c (x) represents the sample correlation between the sample point and the sample segment, and t represents a threshold, which is a default setting (e.g., 0.5) for the online-to-offline service system 100 or can be adjusted in different situations. It should be noted that the sigmoid function or threshold method is provided for illustrative purposes, and other classification methods (e.g., softmax function) may also be used in the present application.

In 760, the processing device 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may determine whether a plurality of sample correlations corresponding to the plurality of samples satisfy a preset condition.

For example, the processing device 112 may determine a first accuracy of an initial correlation determination model corresponding to the training set and a second accuracy of an initial correlation determination model corresponding to the test set. Further, the processing device 112 may determine whether the first accuracy rate has stabilized ("stable" refers to whether the first accuracy rate in a current iteration is substantially the same as (i.e., less than a threshold value) the first accuracy rate in a previous adjacent iteration or a plurality of first fine accuracy rates in a plurality of previous iterations, and whether the second accuracy rate has reached a maximum value, as used herein, the first accuracy rate and/or the second accuracy rate may be determined based on one or more parameters associated with a plurality of sample correlations, e.g., (proximity) and a plurality of labels (i.e., "1" or "0") corresponding to the plurality of samples, in response to the first accuracy rate having stabilized and the second accuracy rate having reached a maximum value, the processing device 112 may determine that the plurality of sample correlations satisfy a preset condition, the processing device 112 may determine that the plurality of sample correlations do not satisfy the preset condition.

For another example, the processing device 112 may determine a loss function of the initial correlation determination model based on the plurality of sample correlations and the plurality of labels, and determine a value of the loss function based on the plurality of sample correlations. Further, the processing device 112 may determine whether the value of the penalty function is less than a penalty threshold. The penalty function may be a default setting for the online-to-offline service 100, or may be adjusted in different circumstances. In response to determining that the value of the loss function is less than the loss threshold, the processing device 112 may determine that the plurality of sample correlations satisfy a preset condition. In response to determining that the value of the loss function is greater than or equal to the loss threshold, the processing device 112 may determine that the plurality of sample correlations do not satisfy the preset condition.

As another example, the processing device 112 may determine whether the number of iterations is greater than a count threshold. In response to determining that the number of iterations is greater than the count threshold, the processing device 112 may determine that the plurality of sample correlations satisfy a preset condition. In response to determining that the number of iterations is less than or equal to the count threshold, the processing device 112 may determine that the plurality of sample correlations do not satisfy the preset condition.

In 770, in response to determining that the plurality of sample correlations satisfies the preset condition, the processing device 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may designate the initial correlation determination model as a trained correlation determination model, meaning that the training process has been completed.

On the other hand, in response to determining that the plurality of sample correlations do not satisfy the preset condition, the processing device 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may perform the process 700 return operation 740 to update the initial correlation determination model. For example, the processing device 112 may update one or more initial parameters to produce an updated initial relevance determination model. Further, the processing device 112 (e.g., the training module 450) (e.g., the processing circuitry of the processor 220) may repeat the step of determining whether the plurality of sample correlations satisfy the preset condition until the plurality of sample correlations satisfy the preset condition. In response to determining that the plurality of updated sample correlations under the updated correlation determination model satisfy the preset condition, the processing device 112 may designate the updated correlation determination model as a trained correlation determination model.

It should be noted that the above description is provided for illustrative purposes only, and is not intended to limit the scope of the present application. Various changes and modifications will occur to those skilled in the art based on the description herein. However, such changes and modifications do not depart from the scope of the present application. For example, the processing device 112 may update the trained correlation determination model at certain time intervals (e.g., monthly, every two months) based on a plurality of newly obtained samples. As another example, the positive and negative examples may be determined manually by an operator or according to predetermined rules by the online-to-offline service 100.

FIG. 8 is a diagram illustrating an exemplary structure of a relevance determination model according to some embodiments of the present application. As shown, the relevance determination model may include a point network and a road segment network.

The point network may be configured to obtain point information and determine a point vector based on the point information. As described in conjunction with fig. 5 and 7, the processing device 112 may extract point features (i.e., first features, first sample features) from the point information and determine a point feature vector (i.e., first feature vector, first sample feature vector) by encoding the point features. Further, the processing device 112 may determine the point vector based on the point feature vector by using a point network.

The road segment network may be configured to obtain road segment information and determine a road segment vector based on the road segment information. As described in conjunction with fig. 5 and 7, the processing device 112 may extract the link features (i.e., second features, second sample features) from the link information and determine the link feature vector (i.e., second feature vector, second sample feature vector) by encoding the link features. Further, the processing device 112 may determine the road segment vector based on the road segment feature vector by using the road segment network.

In some embodiments, to determine the point feature vector and/or the road segment feature vector, the processing device 112 may classify features (e.g., point features, road segment features) as embedded features and dense features and determine the point feature vector and/or the road segment feature vector by encoding the features. In some embodiments, the embedded features may include location information for points or road segments, identification of the requester, identification of the service provider, time information, and the like. Dense features may include frequency of use of road segments, GPS information associated with points or road segments, distances between points and road segments, and the like. Further description of features may be found elsewhere in this application (e.g., fig. 5, fig. 7, and descriptions thereof).

In some embodiments, both the point network and the segment network may include a Full Connectivity (FC) layer, a Bulk Normalization (BN) layer, and a BELU layer. The FC layer may be configured to perform a linear transformation on the point feature vector or the link feature vector. The BN layer may be configured to perform normalization on the intermediate results of the FC layer. The BELU layer may be configured to perform a non-linear transformation on the intermediate results of the BN layer.

As described above, the dimensions of the point feature vector and the dimensions of the link feature vector are different, and the point feature vector and the link feature vector are not comparable. Through three-layer processing, the dimension of the point vector is the same as that of the road segment vector, and the point vector and the road segment vector have comparability. Further, the processing device 112 may determine a degree of correlation between the point and the road segment based on the point vector and the road segment vector. For example, as described in connection with operation 750, the processing device 112 may determine a degree of correlation between the point and the road segment based on a dot product between the point vector and the road segment vector and a classification method.

Having thus described the basic concepts, it will be apparent to those of ordinary skill in the art having read this application that the foregoing disclosure is to be construed as illustrative only and is not limiting of the application. Various modifications, improvements and adaptations of the present application may occur to those skilled in the art, although they are not explicitly described herein. Such modifications, improvements and adaptations are proposed in the present application and thus fall within the spirit and scope of the exemplary embodiments of the present application.

Also, this application uses specific language to describe embodiments of the application. For example, "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with at least one embodiment of the application. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the application may be combined as appropriate.

Moreover, those of ordinary skill in the art will understand that aspects of the present application may be illustrated and described in terms of several patentable species or situations, including any new and useful combination of processes, machines, articles, or materials, or any new and useful improvement thereof. Accordingly, various aspects of the present invention may be embodied entirely in hardware, software (including firmware, resident software, micro-code, etc.) or in a combination of software and hardware implementations, which may be generally referred to herein as a "unit," module "or" system. Furthermore, various aspects of the present invention may take the form of a computer program product embodied in one or more computer-readable media having computer-readable program code embodied therein.

A computer readable signal medium may contain a propagated data signal with computer program code embodied therewith, for example, in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including electro-magnetic, optical, and the like, or any suitable combination. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code on a computer readable signal medium may be propagated over any suitable medium, including radio, cable, fiber optic cable, RF, etc., or any combination of the preceding.

Computer program code for carrying out operations for aspects of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as java, scalla, SimalTalk, Effele, Ju jade, elmarlade, C + +, Cype, VB. NET, Python, etc., conventional procedural programming languages, such as the "C" programming language, visual basic, Fortran2003, Perl, COBOL2002, PHP, ABAP, dynamic programming languages, such as Python, Ruby, and Groovy, or other programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the use of a network service provider's network) or provided in a cloud computing environment or as a service, such as a software service (SaaS).

Additionally, the order in which elements and sequences of the processes described herein are processed, the use of alphanumeric characters, or the use of other designations, is not intended to limit the order of the processes and methods described herein, unless explicitly claimed. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it is to be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although implementations of the various components described above may be embodied in a hardware device, they may also be implemented as a pure software solution, e.g., installation on an existing server or mobile device.

Similarly, it should be noted that in the preceding description of embodiments of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more embodiments of the invention. This method of application, however, is not to be interpreted as reflecting an intention that the claimed subject matter to be scanned requires more features than are expressly recited in each claim. Indeed, the embodiments may be characterized as having less than all of the features of a single embodiment disclosed above.

Claims

1. A system, comprising:

a storage medium storing a set of instructions; and

a processor communicatively coupled with the storage medium to execute the set of instructions to:

receiving a service request from a terminal device, the service request including a target location;

determining a plurality of candidate road segments based on the target location;

determining a plurality of degrees of correlation between the plurality of candidate road segments and the target position by using a trained degree of correlation determination model; and

a target road segment is identified from the plurality of candidate road segments based on the plurality of degrees of correlation.

2. The system of claim 1,

the target location comprises at least one of a starting location or a destination, an

The target road segment corresponds to a road segment associated with the target location.

3. The system of claim 1 or 2, wherein to determine the plurality of degrees of correlation between the plurality of candidate road segments and the target location by using the trained correlation determination model, the processor is configured to:

determining a first vector representation of the target location using a first model;

for each of the plurality of candidate road segments,

determining a second vector representation of the candidate road segment using a second model; and

determining a corresponding correlation between the second representation and the first vector representation using the trained correlation determination model.

4. The system of any one of claims 1-3, wherein the processor is further configured to:

and determining recommendation information associated with the service request according to the target road segment.

5. The system of claim 4, wherein the recommendation information comprises a recommended travel route beginning or ending at a road segment corresponding to the target road segment.

6. The system according to any of claims 1-5, wherein the trained relevance determination model is determined using a training process comprising:

obtaining a plurality of historical travel records, each of the plurality of historical travel records comprising a sample point and one or more sample road segments;

obtaining a plurality of samples based on the plurality of historical travel records, each of the plurality of samples including one of the sample point and the one or more sample road segments;

extracting feature information of each of the plurality of samples;

obtaining an initial correlation degree determination model;

determining, for each of the plurality of samples, a sample correlation between the sample point and the sample segment based on the feature information by using the initial correlation determination model;

determining whether a plurality of sample correlations corresponding to the plurality of samples satisfy a preset condition; and

in response to the plurality of sample correlations satisfying the preset condition, designating the initial correlation determination model as the trained correlation determination model.

7. The system of claim 6, wherein the training process further comprises:

updating the initial correlation degree determination model in response to the plurality of sample correlation degrees not meeting the preset condition; and

repeating the step of determining whether the plurality of sample correlations satisfy the preset condition until the plurality of sample correlations satisfy the preset condition.

8. The system of claim 6 or 7, wherein the plurality of samples comprises a plurality of positive samples and a plurality of negative samples, and

for each of the plurality of positive samples, the sample segment is associated with the sample point in the respective historical travel record; and

for each of the plurality of negative examples, the sample road segment is not correlated with the sample point in the corresponding historical travel record.

9. The system of any one of claims 6-8, wherein the characteristic information of each of the plurality of samples comprises first characteristic information of the sample point and second characteristic information of the sample segment, the first characteristic information comprising at least one of location information of the sample point, an identification of a passenger, an identification of a driver, or time information of the sample point, and the second characteristic information comprising at least one of an identification of the sample segment, a segment type, a segment direction, a segment angle, or a usage frequency of the sample segment.

10. The system according to any one of claims 1-9, wherein the trained relevance determination Model comprises a double tower Deep Structured Semantic Model (DSSM).

11. A method implemented on a computing device having at least one processor, at least one storage medium, and a communication platform connected to a network, the method comprising:

12. The method of claim 11,

13. The method according to claim 11 or 12, wherein the determining the plurality of degrees of correlation between the plurality of candidate road segments and the target location by using the trained correlation determination model comprises:

for each of the plurality of candidate road segments,

14. The method according to any one of claims 11-13, further comprising:

15. The method of claim 14, wherein the recommendation information comprises a recommended travel route beginning or ending at a road segment corresponding to the target road segment.

16. The method according to any of claims 11-15, wherein the trained correlation determination model is determined using a training process comprising:

extracting feature information of each of the plurality of samples;

obtaining an initial correlation degree determination model;

17. The method of claim 16, wherein the training process further comprises:

18. The method of claim 16 or 17, wherein the plurality of samples comprises a plurality of positive samples and a plurality of negative samples, and

19. The method according to any one of claims 16-18, wherein the feature information of each of the plurality of samples comprises first feature information of the sample point and second feature information of the sample segment, the first feature information comprising at least one of location information of the sample point, an identification of a passenger, an identification of a driver, or time information of the sample point, and the second feature information comprising at least one of an identification of the sample segment, a segment type, a segment direction, a segment angle, or a usage frequency of the sample segment.

20. The method according to any of claims 11-19, wherein the trained relevance determination Model comprises a double tower Deep Structured Semantic Model (DSSM).

21. A system, comprising:

a receiving module configured to receive a service request from a terminal device, the service request including a target location;

a candidate segment determination module configured to determine a plurality of candidate segments based on the target location;

a relevance determination module configured to determine a plurality of degrees of relevance between the plurality of candidate road segments and the target location by using a trained relevance determination model; and

an identification module configured to identify a target road segment from the plurality of candidate road segments based on the plurality of degrees of correlation.

22. The system of claim 21,

23. The system of claim 21 or 22, wherein the relevance determination module is further configured to:

for each of the plurality of candidate road segments,

24. The system of any one of claims 21-23, wherein the identification module is further configured to:

25. The system of claim 24, wherein the recommendation information comprises a recommended travel route beginning or ending at a road segment corresponding to the target road segment.

26. The system of any one of claims 21-25, further comprising a training module configured to:

extracting feature information of each of the plurality of samples;

obtaining an initial correlation degree determination model;

designating the initial correlation determination model as the trained correlation determination model in response to determining that the plurality of sample correlations satisfy the preset condition.

27. The system of claim 26, wherein the training module is further configured to:

28. The system of claim 26 or 27, wherein the plurality of samples comprises a plurality of positive samples and a plurality of negative samples, and

29. The system of any one of claims 26-28, wherein the characteristic information of each of the plurality of samples comprises first characteristic information of the sample point and second characteristic information of the sample segment, the first characteristic information comprising at least one of location information of the sample point, an identification of a passenger, an identification of a driver, or time information of the sample point, and the second characteristic information comprising at least one of an identification of the sample segment, a segment type, a segment direction, a segment angle, or a frequency of use of the sample segment.

30. The system according to any of claims 21-29, wherein the trained relevance determination Model comprises a double tower Depth Structured Semantic Model (DSSM).

31. A non-transitory computer-readable medium comprising executable instructions that, when executed by at least one processor, direct the at least one processor to perform a method, the method comprising: