WO2018201979A1 - 交通工具推荐方法、系统及其设备 - Google Patents

交通工具推荐方法、系统及其设备 Download PDF

Info

Publication number
WO2018201979A1
WO2018201979A1 PCT/CN2018/084763 CN2018084763W WO2018201979A1 WO 2018201979 A1 WO2018201979 A1 WO 2018201979A1 CN 2018084763 W CN2018084763 W CN 2018084763W WO 2018201979 A1 WO2018201979 A1 WO 2018201979A1
Authority
WO
WIPO (PCT)
Prior art keywords
passenger
driver
feature
server
information
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2018/084763
Other languages
English (en)
French (fr)
Inventor
石志彬
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Publication of WO2018201979A1 publication Critical patent/WO2018201979A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0631—Recommending goods or services
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0645—Rental transactions; Leasing transactions
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry

Definitions

  • the present application relates to the field of Internet technologies, and in particular, to a vehicle recommendation method, system, and device thereof.
  • the server pushes the vehicle to the passenger according to one or more of the situation of the nearby vehicle, the destination to which the passenger is going, and the rating level of the driver. For example, based on the position information-based taxi recommendation method, the location of the passenger's called vehicle is calculated and the location of the surrounding empty vehicle is selected, and the nearest vehicle is selected for recommendation.
  • the recommended method of taxiing in the related art only recommends the nearest vehicle to the passenger based on the distance factor, and the processing method has a single limitation, which cannot satisfy the multi-directional passenger demand of the user.
  • the present application provides a method, system and device for recommending a vehicle to solve the problem in the related art that only the nearest vehicle is recommended to the passenger based on the distance factor, and the processing mode has a single limitation, and the user's multi-directional traffic demand cannot be satisfied.
  • the first aspect of the present application provides a vehicle recommendation method, where the method is applied to a server, including: receiving a usage request sent by a first terminal, according to a location of a passenger corresponding to the usage request, and a preset schedule. Range screening a plurality of candidate vehicles that meet distance requirements; acquiring a first feature of the passenger and acquiring a second feature of each candidate vehicle driver; respectively, the first feature and each candidate vehicle driver The second feature is matched to obtain a feature matching degree, the target driver is determined according to the feature matching degree; the first response information is fed back to the first terminal, and the second terminal is sent to the second terminal corresponding to the target driver. Response information.
  • the second aspect embodiment of the present application proposes a vehicle recommendation method, the method being applied to the first terminal, comprising: sending a use request to the server, so that the server according to the location and location of the passenger corresponding to the use request Presetting a scheduling range to screen a plurality of candidate vehicles satisfying the distance requirement; receiving first response information fed back by the server, wherein the server separately sets the first feature of the passenger with each candidate vehicle driver The second feature is matched to obtain a feature matching degree, and the target driver is determined according to the feature matching degree.
  • the embodiment of the third aspect of the present application provides a vehicle recommendation method, where the method is applied to the second terminal, including: receiving second response information sent by the server, where the target driver corresponding to the second terminal is The server matches the first feature of the passenger with the second feature of each candidate vehicle driver to obtain a feature matching degree, and is determined according to the feature matching degree; when the target driver reaches the position of the passenger Sending arrival response information to the first terminal corresponding to the passenger.
  • the fourth aspect of the present application provides a server, including: a screening module, configured to receive a usage request sent by a first terminal, and filter a distance requirement according to a location of the passenger corresponding to the usage request and a preset scheduling range. a plurality of candidate vehicles; a first acquisition module for acquiring a first feature of the passenger; a second acquisition module, configured to acquire a second feature of each candidate vehicle driver; and a matching module for The first feature is matched with the second feature of each candidate vehicle driver to obtain a feature matching degree; the determining module is configured to determine the target driver according to the feature matching degree; and the sending module is configured to The terminal feeds back the first response information, and sends the second response information to the second terminal corresponding to the target driver.
  • the fifth aspect of the present application provides a first terminal, including: a first sending module, configured to send a use request to a server, so that the server according to a passenger's riding position and pre-corresponding to the use request
  • the scheduling range is set to filter a plurality of candidate vehicles satisfying the distance requirement
  • the first receiving module is configured to receive the first response information fed back by the server, wherein the server separates the first feature of the passenger with each
  • the second feature of the candidate vehicle driver is matched to obtain a feature matching degree, and the target driver is determined according to the feature matching degree.
  • the sixth aspect of the present application provides a second terminal, including: a receiving module, configured to receive second response information sent by a server, where a target driver corresponding to the second terminal is the server The first feature is respectively matched with the second feature of each candidate vehicle driver to obtain a feature matching degree, and is determined according to the feature matching degree; the first sending module is configured to: when the target driver arrives at the passenger At the time of the location, the arrival response information is transmitted to the first terminal corresponding to the passenger.
  • the seventh embodiment of the present application provides a vehicle recommendation system, including the server according to the fourth aspect, the terminal described in the fifth aspect, and the terminal in the sixth aspect.
  • An eighth aspect of the present application provides an apparatus, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program
  • the vehicle recommendation method described in the first aspect of the present application is implemented.
  • the ninth aspect of the present application provides an apparatus, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program
  • the vehicle recommendation method described in the second aspect of the present application is implemented.
  • a tenth aspect of the present application provides an apparatus, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program
  • the vehicle recommendation method described in the embodiment of the third aspect of the present application is implemented.
  • An eleventh embodiment of the present application provides a non-volatile computer storage medium storing one or more programs, when the one or more programs are executed by a device, causing the The apparatus performs the vehicle recommendation method of the first aspect of the present application.
  • a twelfth aspect of the present application provides a non-volatile computer storage medium storing one or more programs, when the one or more programs are executed by a device, causing the The apparatus performs the vehicle recommendation method of the second aspect of the present application.
  • a thirteenth aspect of the present application provides a non-volatile computer storage medium storing one or more programs, when the one or more programs are executed by a device, causing the The apparatus performs the vehicle recommendation method of the embodiment of the third aspect of the present application.
  • the fourteenth aspect of the present application provides a computer program product that, when executed by an instruction processor in the computer program product, performs the vehicle recommendation method of the first aspect.
  • the fifteenth aspect of the present application provides a computer program product that, when executed by an instruction processor in the computer program product, performs the vehicle recommendation method of the second aspect.
  • the sixteenth aspect of the present application provides a computer program product that, when executed by an instruction processor in the computer program product, performs the vehicle recommendation method of the third aspect.
  • FIG. 1 is a schematic diagram of location-based taxi recommendation according to an embodiment of the present application.
  • FIG. 2 is a schematic flowchart of a first method for recommending a vehicle according to an embodiment of the present application
  • FIG. 3 is a schematic diagram of a vehicle recommendation model provided by an embodiment of the present application matching a first feature with a second feature of each candidate vehicle driver;
  • FIG. 4 is a schematic flowchart of establishing a vehicle recommendation model according to an embodiment of the present application.
  • FIG. 5 is a schematic diagram of a result of classifying and processing data information collected by a driver and a passenger based on an SVM classifier according to an embodiment of the present application;
  • FIG. 6 is a schematic flowchart of a second method for recommending a vehicle according to an embodiment of the present application.
  • FIG. 7 is a schematic flowchart diagram of a third method for recommending a vehicle according to an embodiment of the present application.
  • FIG. 8 is a schematic flowchart diagram of a fourth method for recommending a vehicle according to an embodiment of the present application.
  • FIG. 9 is a schematic flowchart diagram of a fifth method for recommending a vehicle according to an embodiment of the present application.
  • FIG. 10 is a schematic flowchart diagram of a sixth method for recommending a vehicle according to an embodiment of the present disclosure.
  • FIG. 11 is a flow diagram of a process of a vehicle recommendation method according to an embodiment of the present application.
  • FIG. 12 is a schematic structural diagram of a server according to an embodiment of the present application.
  • FIG. 13 is a schematic structural diagram of another server according to an embodiment of the present disclosure.
  • FIG. 14 is a schematic structural diagram of still another server according to an embodiment of the present application.
  • FIG. 15 is a schematic structural diagram of a fourth acquiring module according to an embodiment of the present disclosure.
  • FIG. 16 is a schematic structural diagram of a first terminal according to an embodiment of the present application.
  • FIG. 17 is a schematic structural diagram of another first terminal according to an embodiment of the present disclosure.
  • FIG. 18 is a schematic structural diagram of a second terminal according to an embodiment of the present application.
  • FIG. 19 is a schematic structural diagram of another second terminal according to an embodiment of the present disclosure.
  • FIG. 20 is a schematic structural diagram of a vehicle recommendation system according to an embodiment of the present application.
  • FIG. 21 is a schematic structural diagram of a vehicle recommendation device according to an embodiment of the present application.
  • FIG. 22 is a schematic structural diagram of another vehicle recommendation device according to an embodiment of the present disclosure.
  • FIG. 23 is a schematic structural diagram of still another vehicle recommendation device according to an embodiment of the present application.
  • the server B acquires the location information of the passenger A, and is in the preset distance range of the passenger A (such as shown in FIG. 1).
  • the search for an empty vehicle within the circle is processed to screen out the closest target vehicle C, and after the target vehicle C driver receives the order, push the target vehicle to the passenger.
  • the present application proposes a vehicle recommendation method, which is compared with the location-based vehicle recommendation method in the related art.
  • the flexible matching of passenger characteristics and driver characteristics increases the diversity of traffic recommendation methods and meets the multi-directional traffic needs of users.
  • FIG. 2 is a schematic flowchart diagram of a first method for recommending a vehicle according to an embodiment of the present application. As shown in FIG. 2, the vehicle recommendation method includes the following steps:
  • Step 101 Receive a use request sent by the first terminal, and filter multiple candidate vehicles that meet the distance requirement according to the location of the passenger and the preset scheduling range.
  • the passenger may input a use request through a vehicle application on the first terminal or by making a call, so that the server can receive a corresponding use request and according to the location of the passenger and
  • the preset scheduling range filters out a plurality of candidate vehicles that meet the distance requirement.
  • the first terminal can be selected according to actual application requirements, such as a mobile phone, a tablet computer, and a smart wear device.
  • the server can obtain the passenger position in different ways according to specific application scenarios, as illustrated below:
  • the location information of the passenger is obtained by a positioning device on the first terminal, such as a GPS device of the mobile phone.
  • the location information of the passenger is obtained by the processing of the use request, that is, the location of the passenger is included in the usage request input by the passenger.
  • the preset scheduling range may be selected according to actual application requirements, such as 3KM, 5KM, and the like. Generally speaking, it refers to the dispatch range that most passengers and drivers can accept, that is, the scheduling range of the passengers for the waiting time and the economic requirements of the driver for the empty car.
  • Step 102 Acquire a first feature of the passenger and acquire a second feature of each candidate vehicle driver.
  • the passenger attribute registration information in the database is queried according to the identification information of the passenger, and the attribute characteristics of the passenger are obtained.
  • the passenger preference registration information in the database is queried according to the identification information of the passenger, and the preference characteristics of the passenger are obtained.
  • the first terminal is provided with a usage scene setting interface, and the usage scene characteristics of the passenger are acquired according to the setting information.
  • the first characteristics of the passenger may include, but are not limited to, attributes of the passenger (such as age, gender, etc.), preferences of the passenger (such as hobbies, customs, etc.) and characteristics of the use of the passenger (such as night taxiing). One or more of them.
  • the driver attribute registration information in the database is queried based on the identification information of each candidate vehicle driver to obtain the attribute characteristics of each candidate vehicle driver.
  • the driver preference registration information in the database is queried according to the identification information of each candidate vehicle driver, and the preference characteristics of each candidate vehicle driver are obtained.
  • the second feature of each candidate vehicle driver may include, but is not limited to, an attribute characteristic (such as age, gender, etc.) of each candidate vehicle driver and a preference feature of each candidate vehicle driver (such as a special One or more of talents, areas of expertise, etc.).
  • an attribute characteristic such as age, gender, etc.
  • a preference feature of each candidate vehicle driver such as a special One or more of talents, areas of expertise, etc.
  • Step 103 Match the first feature with the second feature of each candidate vehicle driver, and determine the target driver according to the feature matching degree between the passenger and each candidate vehicle driver.
  • Step 104 Feed back first response information to the first terminal, and send second response information to the second terminal corresponding to the target driver.
  • the first feature can be matched to the second feature of each candidate vehicle driver in a number of ways.
  • a pre-trained vehicle recommendation model by which a first feature can be matched with a second feature of each candidate vehicle driver, respectively, to determine a target driver based on the matching result .
  • FIG. 3 is a schematic diagram of the vehicle recommendation model provided by the embodiment of the present application matching the first feature with the second feature of each candidate vehicle driver.
  • the first features (a1, a2, a3, and a4) of passenger A and the second features (x1, x2, x3, and x4) of candidate vehicle driver X, candidate vehicle driver Y
  • the second feature (y1, y2, y3, and y4) is input to the vehicle recommendation model, and the vehicle recommendation model matches (a1, a2, a3, and a4) with (x1, x2, x3, and x4), and the obtained matching result is A1 matches x1, a2 and x2 match.
  • the vehicle recommendation model then matches (a1, a2, a3, and a4) with (y1, y2, y3, and y4), and the resulting matching results are a1 and y1 matches, a3 and y3, and a4 and y4 match.
  • the candidate vehicle driver Y with the highest degree of matching can be determined as the target driver.
  • the candidate driver with the highest matching degree is one, that is, Y in FIG. 3, whereby the candidate driver Y can be determined as the target driver.
  • the matching degree of the candidate driver Z is the same as that of the candidate driver Y, it is necessary to further select the candidate driver closest to the passenger's position as the target.
  • Driver For example, the distance between the candidate driver Y and the passenger position is 1.8 KM; the distance between the candidate driver Z and the current position of the passenger is 2 KM, and the candidate driver Y is selected as the target driver. That is, in the pre-scheduling range, the embodiment of the present application first considers that the passenger feature and the driver feature have a high degree of matching, and considers the distance closest when the matching degree is as high. Further meet the individual needs of passengers.
  • the pre-trained vehicle recommendation model is used to match the first feature with the second feature of each candidate vehicle driver, and may also be matched by a preset algorithm or the like. , you can choose according to the actual application needs.
  • the server After determining the target driver, the server sends the second response information to the second terminal corresponding to the target driver, and receives the confirmation order information sent by the second terminal, and finally feeds back the first response information to the first terminal.
  • the first response information may be an order notification including the passenger information
  • the second response information may include a taxi response of the target driver information and the like.
  • the vehicle recommendation method of the embodiment of the present application after receiving the use request sent by the first terminal, filters a plurality of candidate vehicles satisfying the distance requirement according to the position of the passenger and the preset scheduling range, and acquires the passenger's a first feature and a second feature of each candidate vehicle driver, and then matching the first feature to a second feature of each candidate vehicle driver, respectively, such that between the passenger and each candidate vehicle driver
  • the feature matching degree determines the target driver, and finally feeds back the first response information to the first terminal, and transmits the second response information to the second terminal corresponding to the target driver.
  • the target vehicle is flexibly matched according to the passenger characteristics and the driver characteristics, thereby increasing the diversity of the traffic recommendation manner and satisfying the multi-directional traffic demand of the user.
  • FIG. 4 is a schematic flowchart of establishing a vehicle recommendation model according to an embodiment of the present application. As shown in FIG. 4, before step 103, the method further includes:
  • Step 201 Acquire a positive sample data set and a negative sample data set of the passenger using the vehicle; wherein the positive sample data set includes: a passenger sample feature, a driver sample feature, and a correspondence relationship of positive evaluation information, where the negative sample data set includes: Correspondence between passenger sample characteristics, driver sample characteristics, and negative evaluation information.
  • Step 202 Generate a vehicle recommendation model according to the positive sample data set and the negative sample data set training model parameters, and apply the vehicle recommendation model to determine the target driver.
  • the server can collect data information of the driver and the passenger, and the data information collected to the driver and the passenger can be processed by the relevant classifier to obtain a positive sample data set and a negative sample data set of the vehicle application.
  • SVM Small Vector Machine
  • SVM Small Vector Machine
  • the basic principle is two-dimensional data. For example, the points of the training data distributed on the two-dimensional plane are clustered in different regions according to their classification, that is, the boundary points are found by training and then linearly divided.
  • the passenger scores 3 stars as a demarcation point, and the passenger scores more than 3 stars of the passenger sample feature, the driver sample feature, and the positive evaluation information as the positive sample data set is located in the right region of the boundary line.
  • the passenger scores no more than 3 stars of passenger sample characteristics, driver sample characteristics, and negative evaluation information as a negative sample data set located in the left area of the boundary line.
  • demarcation point can be selected according to the actual application needs, such as 4 stars, or a score form such as 80 points.
  • passenger sample characteristics include: one or more of age, gender, hometown, hobbies, good topics and special skills; driver sample characteristics include: age, gender, hometown, hobbies, occupations, areas of interest and special One or more of the requirements; positive evaluation information includes: one or more of good service attitude, relatively clean and tidy transportation, and delivery to the destination on time; negative evaluation information includes: poor service attitude, transportation One or more of dirty and multi-way.
  • a vehicle recommendation model is generated based on the positive sample data set and the negative sample data set training model parameters.
  • the positive sample data set and the negative sample data set may train model parameters according to a predetermined ratio, for example, a training ratio of the training model parameters to generate a vehicle recommendation model.
  • the trained vehicle recommendation model is capable of matching the first feature to the second feature of each candidate vehicle driver.
  • the target driver is thus determined based on the degree of matching.
  • FIG. 6 is a schematic flowchart diagram of a second method for recommending a vehicle according to an embodiment of the present application. As shown in FIG. 6, after step S104, the vehicle recommendation method further includes:
  • Step 301 Obtain evaluation information that the target driver performs driving service for the passenger.
  • the evaluation information of the driving service of the target driver for the passenger may be obtained in different manners according to specific application scenarios, as illustrated by the following examples:
  • the driving behavior information of the target driver is collected, the driving behavior characteristics are extracted from the driving behavior information by using a preset information processing algorithm, and the driving behavior characteristics are processed by using a machine learning logistic regression model to generate a target driver as a passenger.
  • the driving behavior information of the target driver may be collected in different manners according to specific application scenarios, as illustrated below:
  • Example One driving route information is collected by a navigation device in a target driver's vehicle.
  • the navigation route is collected by a navigation device such as a GPS positioning device in the target driver's vehicle, and the driving route is a route with the shortest distance between the departure point and the destination.
  • the driving information is extracted from the driving route information by using a preset information processing algorithm. Driving on the correct route, applying the machine learning logistic regression model to process the traffic according to the correct route, and generating the evaluation information of the target driver for the driving service of the passenger is satisfied, more specifically 5 stars, or 95 points.
  • the starting distance and the destination distance are relatively long distances or one of the most traffic jams.
  • the evaluation information of the driving driver for the driving service is not satisfied. More specifically, it can be 2 stars, or 60 points.
  • Example 2 Driving video information is collected by a camera in a target driver's vehicle.
  • driving video information such as the action, behavior, and the like of the target driver
  • a camera in the target driver's vehicle such as a traffic recorder, a terminal, and a camera
  • a preset information processing algorithm from the target driver.
  • the driving behavior is extracted for safe driving.
  • the machine learning logistic regression model is used to process the safe driving, and the evaluation information of the target driver for the driving service of the passenger is generated, which may be 5 stars or 95 points. Wait.
  • the target driver may have one-handed driving, making a phone call, etc., it can be determined that the driver is unsafe to travel.
  • the evaluation information of the target driver for driving the passenger is generated, which is not satisfactory. It is 2 stars, or 60 points.
  • Example 3 driving voice information is collected by a recorder in the target driver's vehicle.
  • the driving voice information such as the voice information of the target driver
  • the driving voice information may be collected by a sound recorder in the target driver's vehicle, and then the driving behavior characteristic is extracted from the target driver's voice information by using a preset information processing algorithm.
  • Applying the machine learning logistic regression model to deal with the civilized driving, and generating the evaluation information of the target driver for the driving service of the passenger is satisfactory, more specifically 5 stars, or 95 points.
  • the target driver may have uncivilized voice, it can be determined to be uncivilized.
  • the evaluation information of the target driver for driving the passenger is generated, which is more than 2 stars, or 60 points and so on.
  • the driving behavior information of the target driver may include one or more of driving route information, driving video information, driving voice information, and the like.
  • the target driver receiving the passenger feedback receives evaluation information for the driving service.
  • the passenger can feedback the evaluation information for the driving service of the target driver to the server through the experience experience after the end of the trip.
  • Step 302 Perform corrective training on the parameters of the vehicle recommendation model according to the first feature of the passenger, the second feature of the target driver, and the evaluation information.
  • the parameters of the vehicle recommendation model can be corrected according to the first feature of the passenger, the second feature of the target driver, and the evaluation information. Further improve the accuracy of the match, thereby improving the passenger travel experience.
  • the following describes the vehicle recommendation method of the embodiment of the present application on the first terminal side.
  • FIG. 7 is a schematic flowchart diagram of a third method for recommending a vehicle according to an embodiment of the present application. As shown in FIG. 7, the vehicle recommendation method includes:
  • Step 401 Send a use request to the server, so that the server filters the plurality of candidate vehicles satisfying the distance requirement according to the location of the passenger and the preset scheduling range.
  • Step 402 Receive first response information fed back by the server, wherein the server matches the first feature of the passenger with the second feature of each candidate vehicle driver respectively, according to the relationship between the passenger and each candidate vehicle driver The feature matching degree determines the target driver.
  • the open vehicle application sends a usage request to the server in the form of a voice input, such as "I am going to Tiananmen”.
  • the vehicle application is opened by text input, and "Tiananmen" is input to send a usage request to the server.
  • the server may filter a plurality of candidate vehicles satisfying the distance requirement according to the location of the passenger and the preset scheduling range, and respectively distinguish the first feature of the passenger with each candidate vehicle driver.
  • the second feature is matched to determine the target driver based on the degree of feature matching between the passenger and each candidate vehicle driver.
  • the server first response information can be received. It can be understood that by analyzing the first response information, target driver information, such as a phone number, a rating level, and the like, can be acquired, so that the target driver can be contacted to better satisfy the passenger travel experience.
  • target driver information such as a phone number, a rating level, and the like
  • the vehicle recommendation method of the embodiment of the present application sends a usage request to the server, so that the server selects a plurality of candidate vehicles satisfying the distance requirement according to the location of the passenger and the preset scheduling range, and then receives the first response information fed back by the server.
  • the server matches the first feature of the passenger with the second feature of each candidate vehicle driver, and determines the target driver based on the feature matching degree between the passenger and each candidate vehicle driver.
  • the target vehicle is flexibly matched according to the passenger characteristics and the driver characteristics, thereby increasing the diversity of the traffic recommendation manner and satisfying the multi-directional traffic demand of the user.
  • FIG. 8 is a schematic flowchart diagram of a fourth method for recommending a vehicle according to an embodiment of the present application. As shown in FIG. 8, the vehicle recommendation method includes:
  • Step 501 Send an attribute feature carrying the passenger identification information to the server, so that the server stores the correspondence between the passenger identification information and the attribute feature in the passenger attribute registration information in the database.
  • Step 502 Send a preference feature carrying the passenger identification information to the server, so that the server stores the correspondence between the passenger identification information and the preference feature in the passenger preference registration information in the database.
  • the passenger may store the characteristics of the passenger in advance by the first terminal, the preference characteristics of the passenger, etc., as the first feature of the passenger is stored in the server.
  • the vehicle application provides an information filling page for the passenger to input attribute characteristics (such as age, gender, etc.) and preference characteristics (such as hobbies, customs, and rides). Car ashamed, etc.).
  • each passenger has its unique account number and password, or a telephone number and the like as the passenger identification information during the above registration, to determine the uniqueness of the passenger, and to be able to match the passenger with its attribute features and preference features. .
  • the server can store the correspondence between the passenger identification information (such as the account password) and the attribute feature in the passenger attribute registration information in the database; and store the correspondence between the passenger identification information and the preference feature in the passenger preference registration information in the database.
  • Step 503 Send a usage request to the server, so that the server filters the plurality of candidate vehicles that meet the distance requirement according to the location of the passenger and the preset scheduling range.
  • step 503 corresponds to the foregoing step 401. Therefore, the description of the step 503 is referred to the description of the foregoing step 401, and details are not described herein again.
  • Step 504 Receive a usage scene setting interface fed back by the server.
  • Step 505 setting a usage scene feature of the passenger in the usage scene setting interface and transmitting the feature to the server.
  • the passenger's usage scene characteristics may be set.
  • setting the usage scene feature of the passenger in the usage scene setting interface is sent to the server by the female driver.
  • Step 506 Receive first response information fed back by the server, wherein the server matches the first feature of the passenger with the second feature of each candidate vehicle driver respectively, according to the relationship between the passenger and each candidate vehicle driver The feature matching degree determines the target driver.
  • step 506 corresponds to step 402 above. Therefore, the description of step 506 is referred to the description of step 402 above, and details are not described herein again.
  • step 507 the evaluation information of the driving service for the target driver for the passenger feedback is sent to the server.
  • the passenger can evaluate the driving service of the target driver through his own travel experience, and feed back the evaluation information to the server, so that the server can further correct the vehicle recommendation model according to the evaluation information. That is, the vehicle recommendation model is updated in real time using the collected data. Thereby, the accuracy of the vehicle recommendation is further improved, and the personalization demand of the passenger is closer.
  • the following describes the vehicle recommendation method of the embodiment of the present application on the second terminal side.
  • FIG. 9 is a schematic flowchart diagram of a fifth method for recommending a vehicle according to an embodiment of the present application. As shown in FIG. 9, the vehicle recommendation method includes:
  • Step 601 Receive second response information sent by the server, where the target driver corresponding to the second terminal is that the server matches the first feature of the passenger with the second feature of each candidate vehicle driver respectively, according to the passenger and the The degree of feature matching between each candidate vehicle driver is determined.
  • Step 602 When the target driver arrives at the location of the passenger, the arrival response information is sent to the first terminal corresponding to the passenger.
  • the first feature of the passenger is respectively matched with the second feature of each candidate vehicle driver in a manner that the server can apply a pre-trained vehicle recommendation model or the like, according to the passenger and each candidate vehicle driver.
  • the second terminal corresponding to the target driver receives the second response information sent by the server.
  • the arrival response information can be transmitted to the first terminal corresponding to the passenger.
  • the form can be in the form of a short message, a telephone, or the like.
  • the vehicle recommendation method of the embodiment of the present application receives the second response information sent by the server, wherein the target driver corresponding to the second terminal is the first feature of the server and the first driver of each candidate vehicle driver respectively.
  • the two features are matched, determined according to the feature matching degree between the passenger and each candidate vehicle driver, and when the target driver arrives at the position of the passenger, the arrival response information is transmitted to the first terminal corresponding to the passenger.
  • the target vehicle is flexibly matched according to the passenger characteristics and the driver characteristics, thereby increasing the diversity of the traffic recommendation manner and satisfying the multi-directional traffic demand of the user.
  • FIG. 10 is a schematic flowchart diagram of a sixth method for recommending a vehicle according to an embodiment of the present application. As shown in FIG. 10, before the step 601, the vehicle recommendation method further includes:
  • Step 701 Send an attribute feature carrying the driver identification information to the server, so that the server stores the correspondence between the driver identification information and the attribute feature in the driver attribute registration information in the database.
  • Step 702 Send a preference feature carrying the driver identification information to the server, so that the server stores the correspondence between the driver identification information and the preference feature in the driver preference registration information in the database.
  • the driver may pre-upload the driver's attribute characteristics, the driver's preference characteristics, and the like as a second feature of the driver in the server through the second terminal.
  • the vehicle application provides an information filling page for the driver to input attribute features (such as age, gender, etc.) and preference features (such as special talents, areas of expertise). Wait).
  • each driver has its unique account number and password, or a telephone number and the like as the driver identification information during the above registration, to determine the uniqueness of the driver, and to be able to assign the driver and its attribute characteristics and preferences.
  • the server can store the correspondence between the driver identification information (such as the account password) and the attribute feature in the driver attribute registration information in the database; the driver who stores the correspondence between the driver identification information and the preference feature in the database Preferred registration information.
  • the driver identification information such as the account password
  • the target vehicle is flexibly matched according to the passenger characteristics and the driver characteristics, thereby improving the travel experience and satisfying the personalized service of the passenger.
  • FIG. 11 is a flow diagram of a process of a vehicle recommendation method according to an embodiment of the present application.
  • the embodiment includes: a server, a first terminal, and a second terminal.
  • This embodiment describes a specific application scenario of the process of the vehicle recommendation method provided by this embodiment by using the information exchange between the devices.
  • the specific application scenario is as follows:
  • step 1.01 the first terminal sends a use request to the server.
  • the vehicle application on the first terminal is opened, and a use request is sent to the server.
  • the passenger opens the taxi application "DD” through the mobile phone, and sends a usage request to the server by manually inputting the destination "Tiananmen", the riding position is "current position", and the like.
  • step 1.02 the server receives the use request sent by the first terminal, and selects a plurality of candidate vehicles that meet the distance requirement according to the location of the passenger and the preset scheduling range.
  • step 1.03 the server acquires the first feature of the passenger and acquires the second feature of each candidate vehicle driver.
  • step 1.04 the server matches the first feature with the second feature of each candidate vehicle driver, and determines the target driver based on the feature matching degree between the passenger and each candidate vehicle driver.
  • step 1.05 the server feeds back the first response information to the first terminal, and sends the second response information to the second terminal corresponding to the target driver.
  • the server acquires the location of the passenger, that is, the specific location where the taxi is needed, and filters a plurality of candidate vehicles satisfying the distance requirement in a preset scheduling range.
  • the first feature of the passenger is acquired based on the passenger identification information, each driver identification information, and the second feature of each candidate vehicle driver is acquired in a prior database.
  • the first feature can be respectively matched with the second feature of each candidate vehicle driver by means of a pre-trained vehicle recommendation model or the like, and determined according to the feature matching degree between the passenger and each candidate vehicle driver.
  • Target driver is a pre-trained vehicle recommendation model or the like.
  • the driver with the highest degree of matching with the occupant feature has one or more, and when there is a pair, the driver who is the closest driver is selected, and the driver corresponding to the target driver is selected.
  • the second terminal sends the second response information, and simultaneously feeds back the first response information to the first terminal.
  • step 1.06 the second terminal sends the arrival response information to the first terminal corresponding to the passenger.
  • the arrival response information is transmitted to the first terminal corresponding to the passenger.
  • the first terminal sends a use request to the server, and after receiving the use request sent by the first terminal, the server selects a plurality of candidate vehicles that meet the distance requirement according to the location of the passenger and the preset scheduling range. And acquiring a first feature of the passenger and a second feature of each candidate vehicle driver, and then matching the first feature with the second feature of each candidate vehicle driver, respectively, based on the passenger and each of the candidate vehicles.
  • the feature matching degree between the drivers determines the target driver, finally feeds back the first response information to the first terminal, and transmits the second response information to the second terminal corresponding to the target driver, and the second terminal corresponds to the passenger
  • a terminal sends an arrival response message.
  • the target vehicle is flexibly matched according to the passenger characteristics and the driver characteristics, thereby improving the travel experience and satisfying the personalized service of the passenger.
  • the present application also proposes a server.
  • FIG. 12 is a schematic structural diagram of a server according to an embodiment of the present application.
  • the server includes: a screening module 11, a first obtaining module 12, a second obtaining module 13, a matching module 14, a determining module 15, and a sending module 16.
  • the screening module 11 is configured to receive a usage request sent by the first terminal, and filter multiple candidate vehicles that meet the distance requirement according to the location of the passenger and the preset scheduling range.
  • the first obtaining module 12 is configured to acquire a first feature of the passenger.
  • the second obtaining module 13 is configured to acquire a second feature of each candidate vehicle driver.
  • the matching module 14 is configured to match the first feature with the second feature of each candidate vehicle driver.
  • the determining module 15 is configured to determine the target driver according to the feature matching degree between the passenger and each candidate vehicle driver.
  • the sending module 16 is configured to feed back the first response information to the first terminal, and send the second response information to the second terminal corresponding to the target driver.
  • the first obtaining module 12 is specifically configured to query the passenger attribute registration information in the database according to the identification information of the passenger, obtain the attribute feature of the passenger, and/or; and query the passenger preference registration information in the database according to the identification information of the passenger. Obtaining a preference feature of the passenger, and/or providing a usage scene setting interface to the first terminal, and acquiring a usage scene feature of the passenger according to the setting information.
  • the second obtaining module 13 is specifically configured to query driver attribute registration information in the database according to the identification information of each candidate vehicle driver, acquire attribute characteristics of each candidate vehicle driver, and/or;
  • the identification information of each candidate vehicle driver queries the driver preference registration information in the database to obtain the preference characteristics of each candidate vehicle driver.
  • the determining module 15 is specifically configured to: if the candidate driver with the highest matching degree is one, determine the candidate driver as the target driver; if the candidate driver with the highest matching degree is multiple, select the current with the passenger The candidate driver who is closest to the location is the target driver.
  • FIG. 13 is a schematic structural diagram of another server according to an embodiment of the present application. As shown in FIG. 13, on the basis of FIG. 12, the server further includes: a third acquiring module 17 And generating module 18.
  • the third obtaining module 17 is configured to obtain a positive sample data set and a negative sample data set of the taxi application; wherein, the positive sample data set includes: a passenger sample feature, a driver sample feature, and a correspondence between positive evaluation information, and a negative relationship
  • the sample data set includes: passenger sample characteristics, driver sample characteristics, and correspondence of negative evaluation information.
  • the generating module 18 is configured to generate a vehicle recommendation model according to the positive sample data set and the negative sample data set training model parameters.
  • FIG. 14 is a schematic structural diagram of another server according to an embodiment of the present disclosure. As shown in FIG. 14, on the basis of FIG. 13, the server further includes: a fourth acquiring module 19 And training module 110.
  • the fourth obtaining module 19 is configured to obtain evaluation information that the target driver performs driving service for the passenger.
  • the training module 20 is configured to perform corrective training on parameters of the vehicle recommendation model according to the first feature of the passenger, the second feature of the target driver, and the evaluation information.
  • FIG. 15 is a schematic structural diagram of a fourth acquiring module according to an embodiment of the present disclosure.
  • the fourth obtaining module 19 includes an acquiring unit 191, an extracting unit 192, and a processing unit 193.
  • the collecting unit 191 is configured to collect driving behavior information of the target driver.
  • the extracting unit 192 is configured to extract driving behavior characteristics from the driving behavior information by using a preset information processing algorithm.
  • the processing unit 193 is configured to apply a machine learning logistic regression model to process the driving behavior feature, and generate evaluation information that the target driver performs driving service for the passenger.
  • the collecting unit 191 is specifically configured to collect driving route information by using a navigation device in the target driver vehicle; and/or; collecting driving video information through a camera in the target driver vehicle; and/or; driving through the target
  • the recorder in the vehicle's vehicle collects driving voice information.
  • the fourth obtaining module 19 further includes: a receiving unit 194.
  • the receiving unit 194 is configured to receive evaluation information of the driving driver for the driving service for the target driver.
  • the server of the embodiment of the present application after receiving the use request sent by the first terminal, filters a plurality of candidate vehicles satisfying the distance requirement according to the position of the passenger and the preset scheduling range, and acquires the first feature of the passenger and each candidate traffic. a second feature of the tool driver, then matching the first feature to the second feature of each candidate vehicle driver, respectively, to determine the target driver based on the feature match between the passenger and each candidate vehicle driver Finally, the first response information is fed back to the first terminal, and the second response information is sent to the second terminal corresponding to the target driver.
  • the target vehicle is flexibly matched according to the passenger characteristics and the driver characteristics, thereby increasing the diversity of the traffic recommendation manner and satisfying the multi-directional traffic demand of the user.
  • FIG. 16 is a schematic structural diagram of a first terminal according to an embodiment of the present disclosure.
  • the first terminal includes: a first sending module 21 and a first receiving module 22.
  • the first sending module 21 is configured to send a usage request to the server, so that the server filters the plurality of candidate vehicles that meet the distance requirement according to the location of the passenger and the preset scheduling range.
  • the first receiving module 22 is configured to receive first response information fed back by the server, wherein the server matches the first feature of the passenger with the second feature of each candidate vehicle driver respectively, according to the passenger and each candidate vehicle The feature matching degree between the drivers determines the target driver.
  • FIG. 17 is a schematic structural diagram of still another first terminal according to an embodiment of the present application.
  • the first terminal further includes: The second sending module 23, the third sending module 24, the second receiving module 25, the processing module 26 and the feedback module 27.
  • the second sending module 23 is configured to send an attribute feature carrying the passenger identification information to the server, so that the server stores the correspondence between the passenger identification information and the attribute feature in the passenger attribute registration information in the database, and/or;
  • the third sending module 24 is configured to send, to the server, a preference feature that carries the passenger identification information, so that the server stores the correspondence between the passenger identification information and the preference feature in the passenger preference registration information in the database.
  • the second receiving module 25 is configured to receive a usage scenario setting interface fed back by the server.
  • the processing module 26 is configured to set a usage scene feature of the passenger in the usage scenario setting interface and send the feature to the server.
  • the feedback module 27 is configured to send, to the server, the evaluation information of the target driver for the driving service for the driver.
  • the first terminal in the embodiment of the present application sends a use request to the server, so that the server selects a plurality of candidate vehicles satisfying the distance requirement according to the location of the passenger and the preset scheduling range, and then receives the first response information fed back by the server, Wherein the server matches the first feature of the passenger with the second feature of each candidate vehicle driver, and determines the target driver based on the feature matching degree between the passenger and each candidate vehicle driver.
  • the target vehicle is flexibly matched according to the passenger characteristics and the driver characteristics, thereby increasing the diversity of the traffic recommendation manner and satisfying the multi-directional traffic demand of the user.
  • FIG. 18 is a schematic structural diagram of a second terminal according to an embodiment of the present application.
  • the second terminal includes: a receiving module 31 and a first sending module 32.
  • the receiving module 31 is configured to receive second response information sent by the server, where the target driver corresponding to the second terminal is that the server performs the first feature of the passenger and the second feature of each candidate vehicle driver respectively.
  • the match is determined based on the degree of feature matching between the passenger and each candidate vehicle driver.
  • the first sending module 32 is configured to send the arrival response information to the first terminal corresponding to the passenger when the target driver arrives at the location of the passenger.
  • the second terminal further includes: The second sending module 33 and the third sending module 34.
  • the second sending module 33 is configured to send, to the server, an attribute feature that carries the driver identification information, so that the server stores the correspondence between the driver identification information and the attribute feature in the driver attribute registration information in the database, and / or;
  • the third sending module 34 is configured to send, to the server, a preference feature that carries the driver identification information, so that the server stores the correspondence between the driver identification information and the preference feature in the driver preference registration information in the database.
  • the second terminal of the embodiment of the present application receives the second response information sent by the server, where the target driver corresponding to the second terminal is the server, the first feature of the passenger and the second candidate of each candidate vehicle driver respectively.
  • the feature is matched, determined based on the feature matching degree between the passenger and each candidate vehicle driver, and when the target driver arrives at the position of the passenger, the arrival response information is transmitted to the first terminal corresponding to the passenger.
  • the target vehicle is flexibly matched according to the passenger characteristics and the driver characteristics, thereby increasing the diversity of the traffic recommendation manner and satisfying the multi-directional traffic demand of the user.
  • FIG. 20 is a schematic structural diagram of a taxi recommendation system according to an embodiment of the present application.
  • the taxi recommendation system includes: the server 1 described in FIG. 12 to FIG. 15 of the present application, the first terminal 2 described in FIGS. 16 to 17 of the present application, and the second terminal 3 described in FIGS. 18 to 19 of the present application.
  • the first terminal sends a use request
  • the server receives the use request sent by the first terminal, and then selects a plurality of candidate vehicles satisfying the distance requirement according to the position of the passenger and the preset scheduling range, and acquires the passenger.
  • a first feature and a second feature of each candidate vehicle driver and then matching the first feature to a second feature of each candidate vehicle driver, respectively, based on the passenger and each candidate vehicle driver
  • the feature matching degree determines the target driver, and finally feeds back the first response information to the first terminal, and sends the second response information to the second terminal corresponding to the target driver, and the second terminal sends the first response information to the first terminal corresponding to the passenger.
  • the target vehicle is flexibly matched according to the passenger characteristics and the driver characteristics, thereby increasing the diversity of the traffic recommendation manner and satisfying the multi-directional traffic demand of the user.
  • the present application also proposes a taxi recommendation device.
  • FIG. 21 is a schematic structural diagram of a taxi recommendation device according to an embodiment of the present application.
  • the taxi recommendation device may include a processor 11 and a memory 12 for storing processor-executable instructions.
  • the processor 11 is configured to: receive a usage request sent by the first terminal, filter a plurality of candidate vehicles satisfying the distance requirement according to the location of the passenger and a preset scheduling range; acquire the first feature of the passenger, and acquire each a second feature of the candidate vehicle driver; matching the first feature to the second feature of each candidate vehicle driver, and determining the target driver based on the feature matching degree between the passenger and each candidate vehicle driver Transmitting the first response information to the first terminal and transmitting the second response information to the second terminal corresponding to the target driver.
  • FIG. 22 is a schematic structural diagram of another type of taxi recommendation device according to an embodiment of the present application.
  • the taxi recommendation device may include a processor 11 and a memory 12 for storing processor-executable instructions.
  • the processor 11 is configured to: send a usage request to the server, so that the server filters the plurality of candidate vehicles that meet the distance requirement according to the location of the passenger and the preset scheduling range; and receives the first response information fed back by the server, where The server matches the first feature of the passenger with the second feature of each candidate vehicle driver, and determines the target driver based on the degree of feature matching between the passenger and each candidate vehicle driver.
  • FIG. 23 is a schematic structural diagram of still another vehicle recommendation device according to an embodiment of the present application.
  • FIG. 23 takes the second terminal side as an example.
  • the taxi recommendation device may include a processor 11 and a memory 12 for storing processor-executable instructions.
  • the processor 11 is configured to: receive second response information sent by the server, where the target driver corresponding to the second terminal is the first feature of the server and the first driver of each candidate vehicle The two features are matched according to the feature matching degree between the passenger and each candidate vehicle driver; when the target driver arrives at the passenger's riding position, the arrival response information is transmitted to the first terminal corresponding to the passenger.
  • first and second are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated.
  • features defining “first” or “second” may include at least one of the features, either explicitly or implicitly.
  • the meaning of "a plurality” is at least two, such as two, three, etc., unless specifically defined otherwise.
  • a "computer-readable medium” can be any apparatus that can contain, store, communicate, propagate, or transport a program for use in an instruction execution system, apparatus, or device, or in conjunction with the instruction execution system, apparatus, or device.
  • computer readable media include the following: electrical connections (electronic devices) having one or more wires, portable computer disk cartridges (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM).
  • the computer readable medium may even be a paper or other suitable medium on which the program can be printed, as it may be optically scanned, for example by paper or other medium, followed by editing, interpretation or, if appropriate, other suitable The method is processed to obtain the program electronically and then stored in computer memory.
  • portions of the application can be implemented in hardware, software, firmware, or a combination thereof.
  • multiple steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system.
  • a suitable instruction execution system For example, if implemented in hardware and in another embodiment, it can be implemented by any one or combination of the following techniques well known in the art: discrete with logic gates for implementing logic functions on data signals Logic circuits, application specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like.
  • each functional unit in each embodiment of the present application may be integrated into one processing module, or each unit may exist physically separately, or two or more units may be integrated into one module.
  • the above integrated modules can be implemented in the form of hardware or in the form of software functional modules.
  • the integrated modules, if implemented in the form of software functional modules and sold or used as stand-alone products, may also be stored in a computer readable storage medium.
  • the above mentioned storage medium may be a read only memory, a magnetic disk or an optical disk or the like. While the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are illustrative and are not to be construed as limiting the scope of the present application. The embodiments are subject to variations, modifications, substitutions and variations.

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Development Economics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Primary Health Care (AREA)
  • Tourism & Hospitality (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Navigation (AREA)
  • Traffic Control Systems (AREA)

Abstract

本申请公开了一种交通工具推荐方法、系统及其设备，其中，该交通工具推荐方法包括：接收第一终端发送的使用请求，根据与所述使用请求对应的乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具；获取乘客的第一特征，以及获取每个候选交通工具驾驶员的第二特征；将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度，根据特征匹配度确定目标驾驶员；向第一终端反馈第一响应信息，以及向与目标驾驶员对应的第二终端发送第二响应信息。由此，实现了根据乘客特征和驾驶员特征灵活匹配确定目标交通工具，从而增加交通推荐方式的多样性，满足用户多方位的交通需求。

Description

交通工具推荐方法、系统及其设备
本申请要求申请日为2017年5月5日、申请号为201710313989.5、发明创造名称为《交通工具推荐方法、系统及其设备》的发明专利申请的优先权。
技术领域
本申请涉及互联网技术领域，尤其涉及一种交通工具推荐方法、系统及其设备。
背景技术
随着智能终端、移动互联网应用程序的发展，打车应用越来越普及，给城市人们的出行提供了便利。
日常生活中，乘客在通过打车应用叫车时，服务器会根据附近车辆的情况、乘客要去的目的地和司机的评分等级等中的一种或者多种给乘客推送车辆。比如，基于位置信息的打车推荐方法，从乘客叫车的位置计算和周围空车车辆的位置，筛选距离最近的车辆进行推荐。
然而，相关技术中的打车推荐方法，仅仅是基于距离因素向乘客推荐最近的车辆，处理方式单一局限，不能满足用户多方位的乘车需求。
发明内容
本申请提供一种交通工具推荐方法、系统及其设备，以解决相关技术中，仅仅是基于距离因素向乘客推荐最近的交通工具，处理方式单一局限，不能满足用户多方位的交通需求的问题。
本申请第一方面实施例提出了一种交通工具推荐方法，所述方法应用于服务器，包括：接收第一终端发送的使用请求，根据与所述使用请求对应的乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具；获取所述乘客的第一特征，以及获取每个候选交通工具驾驶员的第二特征；将所述第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度，根据所述特征匹配度确定目标驾驶员；向所述第一终端反馈第一响应信息，以及向与所述目标驾驶员对应的第二终端发送第二响应信息。
本申请第二方面实施例提出了一种交通工具推荐方法，所述方法应用于第一终端，包括：向服务器发送使用请求，以使所述服务器根据与所述使用请求对应的乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具；接收所述服务器反馈的第一响应信息，其中，所述服务器将所述乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度，根据所述特征匹配度确定目标驾驶员。
本申请第三方面实施例提出了一种交通工具推荐方法，所述方法应用于第二终端，包括：接收服务器发送的第二响应信息，其中，与所述第二终端对应的目标驾驶员是所述服 务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度，并根据所述特征匹配度确定的；当所述目标驾驶员到达乘客的位置时，向与所述乘客对应的第一终端发送到达响应信息。
本申请第四方面实施例提出了一种服务器，包括：筛选模块，用于接收第一终端发送的使用请求，根据与所述使用请求对应的乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具；第一获取模块，用于获取所述乘客的第一特征；第二获取模块，用于获取每个候选交通工具驾驶员的第二特征；匹配模块，用于将所述第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度；确定模块，用于根据所述特征匹配度确定目标驾驶员；发送模块，用于向所述第一终端反馈第一响应信息，以及向与所述目标驾驶员对应的第二终端发送第二响应信息。
本申请第五方面实施例提出了一种第一终端，包括：第一发送模块，用于向服务器发送使用请求，以使所述服务器根据与所述使用请求对应的乘客的乘车位置和预设的调度范围筛选满足距离要求的多个候选交通工具；第一接收模块，用于接收所述服务器反馈的第一响应信息，其中，所述服务器将所述乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度，并根据所述特征匹配度确定目标驾驶员。
本申请第六方面实施例提出了一种第二终端，包括：接收模块，用于接收服务器发送的第二响应信息，其中，与所述第二终端对应的目标驾驶员是所述服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度，并根据所述特征匹配度确定的；第一发送模块，用于当所述目标驾驶员到达乘客的位置时，向与所述乘客对应的第一终端发送到达响应信息。
本申请第七方面实施例提供了一种交通工具推荐系统，包括第四方面实施例所述的服务器、第五方面实施例所述的终端和第六方面实施例所述的终端。
本申请第八方面实施例提供了一种设备，包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，其中，所述处理器执行所述计算机程序时，实现本申请第一方面实施例所述的交通工具推荐方法。
本申请第九方面实施例提供了一种设备，包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，其中，所述处理器执行所述计算机程序时，实现本申请第二方面实施例所述的交通工具推荐方法。
本申请第十方面实施例提供了一种设备，包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，其中，所述处理器执行所述计算机程序时，实现本申请第三方面实施例所述的交通工具推荐方法。
本申请第十一方面实施例提供了一种非易失性计算机存储介质，所述计算机存储介质存储有一个或者多个程序，当所述一个或者多个程序被一个设备执行时，使得所述设备执行本申请第一方面实施例的交通工具推荐方法。
本申请第十二方面实施例提供了一种非易失性计算机存储介质，所述计算机存储介质 存储有一个或者多个程序，当所述一个或者多个程序被一个设备执行时，使得所述设备执行本申请第二方面实施例的交通工具推荐方法。
本申请第十三方面实施例提供了一种非易失性计算机存储介质，所述计算机存储介质存储有一个或者多个程序，当所述一个或者多个程序被一个设备执行时，使得所述设备执行本申请第三方面实施例的交通工具推荐方法。
本申请第十四方面实施例提供了一种计算机程序产品，当所述计算机程序产品中的指令处理器执行时，执行第一方面实施例所述的交通工具推荐方法。
本申请第十五方面实施例提供了一种计算机程序产品，当所述计算机程序产品中的指令处理器执行时，执行第二方面实施例所述的交通工具推荐方法。
本申请第十六方面实施例提供了一种计算机程序产品，当所述计算机程序产品中的指令处理器执行时，执行第三方面实施例所述的交通工具推荐方法。
本申请实施例提供的技术方案可以包括以下有益效果：
通过接收第一终端发送的使用请求后根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具，并获取乘客的第一特征和每个候选交通工具驾驶员的第二特征，接着将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，从而根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员，最后向第一终端第一响应信息，以及向与目标驾驶员对应的第二终端发送第二响应信息，实现根据乘客特征和驾驶员特征灵活匹配确定目标交通工具，从而增加交通推荐方式的多样性，满足用户多方位的交通需求。
本申请附加的方面和优点将在下面的描述中部分给出，部分将从下面的描述中变得明显，或通过本申请的实践了解到。
附图说明
本申请上述的和/或附加的方面和优点从下面结合附图对实施例的描述中将变得明显和容易理解，其中：
图1为本申请实施例提供的基于位置的打车推荐示意图。
图2为本申请实施例提供的第一种交通工具推荐方法的流程示意图；
图3本申请实施例提供的交通工具推荐模型将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配的示意图；
图4本申请实施例提供的一种建立交通工具推荐模型的流程示意图；
图5为本申请实施例提供的基于SVM分类器对收集到驾驶员和乘客的数据信息进行分类处理的结果示意图；
图6为本申请实施例提供的第二种交通工具推荐方法的流程示意图；
图7为本申请实施例提供的第三种交通工具推荐方法的流程示意图；
图8为本申请实施例提供的第四种交通工具推荐方法的流程示意图；
图9为本申请实施例提供的第五种交通工具推荐方法的流程示意图；
图10为本申请实施例提供的第六种交通工具推荐方法的流程示意图；
图11为本申请实施例的一个交通工具推荐方法的流程交互图；
图12为本申请实施例提供的一种服务器的结构示意图；
图13为本申请实施例提供的另一种服务器的结构示意图；
图14为本申请实施例提供的又一种服务器的结构示意图；
图15为本申请实施例提供的一种第四获取模块的结构示意图；
图16为本申请实施例提供的一种第一终端的结构示意图；
图17为本申请实施例提供的另一种第一终端的结构示意图；
图18为本申请实施例提供的一种第二终端的结构示意图；
图19为本申请实施例提供的另一种第二终端的结构示意图；
图20为本申请实施例提供的一种交通工具推荐系统的结构示意图；
图21为本申请实施例提供的一种交通工具推荐装置的结构示意图；
图22为本申请实施例提供的另一种交通工具推荐装置的结构示意图；
图23为本申请实施例提供的又一种交通工具推荐装置的结构示意图。
具体实施方式
下面详细描述本申请的实施例，所述实施例的示例在附图中示出，其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附图描述的实施例是示例性的，旨在用于解释本申请，而不能理解为对本申请的限制。
下面参考附图描述本申请实施例的交通工具推荐方法、系统及其设备。
具体地，根据图1所示的基于位置的打车推荐示意图，在乘客A通过打车应用叫车时，服务器B获取乘客A的位置信息，并在乘客A的预设距离范围(比如图1所示的圆圈)内搜索空车车辆，经过处理筛选出距离最近的目标车辆C，并在目标车辆C驾驶员接单后，向乘客推送目标车辆。
因此，为了方便用户，需要向乘客推荐合适目标交通工具，本申请提出一种交通工具推荐方法，该方法相对于相关技术中的仅仅基于位置的交通工具推荐方法相比，本实施例中，涉及乘客特征和驾驶员特征的灵活匹配，增加交通推荐方式的多样性，满足用户多方位的交通需求。
为了便于描述，下面首先集中在服务器侧描述本申请实施例的交通工具推荐方法。
图2为本申请实施例提供的第一种交通工具推荐方法的流程示意图。如图2所示，该交通工具推荐方法包括以下步骤：
步骤101，接收第一终端发送的使用请求，根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具。
在实际应用当中，乘客在需要交通工具时，可以通过第一终端上的交通工具应用、或者是打电话等方式输入使用请求，从而服务器能够接收到相对应的使用请求，并根据乘客的位置和预设的调度范围筛选出满足距离要求的多个候选交通工具。
需要说明的是，第一终端可以根据实际应用需要进行选择设置，比如手机、平板电脑和智能佩戴设备等。
可以理解的是，服务器可以根据具体应用场景不同采用不同的方式获取乘客位置，举例说明如下：
第一种示例，通过第一终端上的定位装置(比如手机的GPS装置)获取乘客的位置信息。
第二种示例，通过对使用请求的处理，获取乘客的位置信息，即乘客输入的使用请求中包括乘客的位置。
需要注意的是，以上对乘客位置的获取方式仅为举例说明，可以根据实际应用需要进行选择和调整。
需要说明的是，预设的调度范围可以根据实际应用需要进行选择设置，比如3KM、5KM等。一般，指的是大部分乘客、驾驶员能够接受的调度范围，即同时满足乘客对于等待时间和驾驶员对于空车行驶的经济要求等情况下的调度范围。
步骤102，获取乘客的第一特征，以及获取每个候选交通工具驾驶员的第二特征。
具体地，确定了满足距离要求的多个候选交通工具后，为了向乘客提供与其特征最匹配的目标驾驶员，需要获取乘客的第一特征和每个候选交通工具驾驶员的第二特征。
需要说明的是，可以根据实际应用场景需要，采用不同的方式获取乘客的第一特征，举例说明如下：
第一种示例，根据乘客的标识信息查询数据库中的乘客属性注册信息，获取乘客的属性特征。
第二种示例，根据乘客的标识信息查询数据库中的乘客偏好注册信息，获取乘客的偏好特征。
第三种示例，向第一终端提供使用场景设置界面，根据设置信息获取乘客的使用场景特征。
由此，乘客的第一特征可以包括但不限于乘客的属性特征(比如年龄、性别等)、乘客的偏好特征(比如兴趣爱好、风俗习惯等)和乘客的使用场景(比如夜晚打车)特征等中的一种或者多种。
需要说明的是，以上方式仅为获取乘客的第一特征的举例说明，可以根据实际应用需要选择或者设置其他的方式。
需要说明的是，可以根据实际应用场景需要，采用不同的方式获取每个候选交通工具驾驶员的第二特征，举例说明如下：
第一种示例，根据每个候选交通工具驾驶员的标识信息查询数据库中的驾驶员属性注 册信息，获取每个候选交通工具驾驶员的属性特征。
第二种示例，根据每个候选交通工具驾驶员的标识信息查询数据库中的驾驶员偏好注册信息，获取每个候选交通工具驾驶员的偏好特征。
由此，每个候选交通工具驾驶员的第二特征可以包括但不限于每个候选交通工具驾驶员的属性特征(比如年龄、性别等)和每个候选交通工具驾驶员的偏好特征(比如特殊才艺、擅长领域等)等中的一种或者多种。
需要说明的是，以上方式仅为获取每个候选交通工具驾驶员的第二特征的举例说明，可以根据实际应用需要选择或者设置其他的方式。
步骤103，将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员。
步骤104，向第一终端反馈第一响应信息，以及向与目标驾驶员对应的第二终端发送第二响应信息。
具体地，可以通过很多种方式将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配。作为一种示例，具有预先训练好的交通工具推荐模型，通过该交通工具推荐模型可以将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，从而根据匹配结果确定目标驾驶员。具体结合图3解释说明如下：
具体地，图3本申请实施例提供的交通工具推荐模型将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配的示意图。如图3所示，将乘客A的第一特征(a1、a2、a3和a4)和候选交通工具驾驶员X的第二特征(x1、x2、x3和x4)、候选交通工具驾驶员Y的第二特征(y1、y2、y3和y4)输入交通工具推荐模型，交通工具推荐模型将(a1、a2、a3和a4)和(x1、x2、x3和x4)进行匹配，得到的匹配结果为a1和x1匹配、a2和x2匹配。接着交通工具推荐模型将(a1、a2、a3和a4)和(y1、y2、y3和y4)进行匹配，得到的匹配结果为a1和y1匹配、a3和y3和a4和y4匹配。由此，可以确定匹配度最高的候选交通工具驾驶员Y作为目标驾驶员。
需要说明的是，在上述例子中，匹配度最高的候选驾驶员为一个即图3中的Y，由此可以将候选驾驶员Y确定为目标驾驶员。当匹配度最高的候选驾驶员为多个，继续以上述例子为例，比如还有候选驾驶员Z的匹配度和候选驾驶员Y一样，需要进一步选择与乘客的位置最近的候选驾驶员为目标驾驶员。比如，候选驾驶员Y与乘客位置的距离为1.8KM；候选驾驶员Z与乘客当前位置的距离为2KM，则选择候选驾驶员Y为目标驾驶员。即本申请实施例在预先调度范围内，首先考虑乘客特征和驾驶员特征匹配度高的，在匹配度一样高的情况下再考虑距离最近。进一步满足乘客个性化需求。
需要说明的是，不限于上述描述中将预先训练好的交通工具推荐模型将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，还可以通过预先设置的算法等方式进行匹配，可以根据实际应用需要进行选择设置。
进而，在确定目标驾驶员后，服务器向目标驾驶员对应的第二终端发送第二响应信息， 并接收到第二终端发送的确认接单信息，最后向第一终端反馈第一响应信息。
其中，第一响应信息可以是包括乘客信息的接单通知等；第二响应信息可以包括目标驾驶员信息的打车响应等。
综上所述，本申请实施例的交通工具推荐方法，通过接收第一终端发送的使用请求后根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具，并获取乘客的第一特征和每个候选交通工具驾驶员的第二特征，接着将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，从而根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员，最后向第一终端反馈第一响应信息，以及向与目标驾驶员对应的第二终端发送第二响应信息。由此，实现了根据乘客特征和驾驶员特征灵活匹配确定目标交通工具，从而增加交通推荐方式的多样性，满足用户多方位的交通需求。
基于以上实施例，为了本领域人员更加清楚本实施例的交通工具推荐模型，下面结合图4具体说明如何建立交通工具推荐模型的具体过程。
图4本申请实施例提供的一种建立交通工具推荐模型的流程示意图。如图4所示，在步骤103之前，还包括：
步骤201，获取乘客使用交通工具的正样本数据集合和负样本数据集合；其中，正样本数据集合包括：乘客样本特征、驾驶员样本特征，以及正面评价信息的对应关系，负样本数据集合包括：乘客样本特征、驾驶员样本特征，以及负面评价信息的对应关系。
步骤202，根据正样本数据集合和负样本数据集合训练模型参数生成交通工具推荐模型，以应用交通工具推荐模型确定目标驾驶员。
具体地，服务器能够收集到驾驶员和乘客的数据信息，可以通过相关分类器对收集到驾驶员和乘客的数据信息进行处理获取交通工具应用的正样本数据集合和负样本数据集合。
作为一种示例，比如通过SVM(Support Vector Machine，支持向量机)分类器对收集到驾驶员和乘客的数据信息进行分类处理，SVM是一种基于分类边界的方法，基本原理是以二维数据为例：训练数据分布在二维平面上的点按照其分类聚集在不同的区域，即通过训练找到分界点后进行线性划分。
具体地，如图5所示，以乘客打分3星为分界点，乘客打分超过3星的乘客样本特征、驾驶员样本特征，以及正面评价信息的对应关系作为正样本数据集合位于分界线右边区域；乘客打分不超过3星的乘客样本特征、驾驶员样本特征，以及负面评价信息的对应关系作为负样本数据集合位于分界线左边区域。
需要说明的是，分界点可以根据实际应用需要进行选择，比如可以为4星，还可以为分数形式比如80分等。
其中，乘客样本特征包括：年龄、性别、家乡、爱好、擅长话题和特殊技能等中的一种或者多种；驾驶员样本特征包括：年龄、性别、家乡、爱好、职业、感兴趣领域和特殊要求等中的一种或者多种；正面评价信息包括：服务态度好、交通工具比较干净整洁和准 时送达目的地等中的一种或者多种；负面评价信息包括：服务态度差、交通工具比较脏和多绕路等中的一种或者多种。
进而，根据正样本数据集合和负样本数据集合训练模型参数生成交通工具推荐模型。其中，正样本数据集合和负样本数据集合可以按照预订的比例训练模型参数，比如1:1的比例进行训练模型参数生成交通工具推荐模型。
由此，训练好的交通工具推荐模型，能够将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配。从而根据匹配度确定目标驾驶员。
图6为本申请实施例提供的第二种交通工具推荐方法的流程示意图。如图6所示，在步骤S104后，该交通工具推荐方法还包括：
步骤301，获取目标驾驶员为乘客进行驾驶服务的评价信息。
具体地，可以根据具体应用场景不同采用不同的方式获取目标驾驶员为乘客进行驾驶服务的评价信息，举例说明如下：
第一种示例，采集目标驾驶员的驾驶行为信息，采用预设的信息处理算法从驾驶行为信息中提取驾驶行为特征，应用机器学习逻辑回归模型对驾驶行为特征进行处理，生成目标驾驶员为乘客进行驾驶服务的评价信息。
具体地，可以根据具体应用场景不同采用不同的方式采集目标驾驶员的驾驶行为信息，举例说明如下：
示例一，通过目标驾驶员交通工具中的导航设备采集驾驶路线信息。
具体地，通过目标驾驶员交通工具中的GPS定位装置等导航设备采集驾驶路线是出发地和目的地距离最短的一条路线，采用预设的信息处理算法从驾驶路线信息中提取驾驶行为特征为按照正确路线行驶，应用机器学习逻辑回归模型对按照正确路线行驶进行处理，生成目标驾驶员为乘客进行驾驶服务的评价信息为满意，更具体可以为5星、或者95分等。
可以理解的是，当采集驾驶路线是出发地和目的地距离是相对比较长距离或者是比较最堵车的一条路线，通过处理后，生成目标驾驶员为乘客进行驾驶服务的评价信息为不满意，更具体可以为2星、或者60分等。
示例二，通过目标驾驶员交通工具中的摄像头采集驾驶视频信息。
具体地，可以通过目标驾驶员交通工具中的摄像头比如交通记录仪、终端和摄像装置等采集驾驶视频信息，比如目标驾驶员的动作、行为等，然后采用预设的信息处理算法从目标驾驶员的动作、行为中提取驾驶行为特征为安全行驶，应用机器学习逻辑回归模型对安全行驶进行处理，生成目标驾驶员为乘客进行驾驶服务的评价信息为满意，更具体可以为5星、或者95分等。
可以理解的是，目标驾驶员可能出现单手驾驶、打电话等行为时，可以确定为不安全行驶，通过处理后，生成目标驾驶员为乘客进行驾驶服务的评价信息为不满意，更具体可以为2星、或者60分等。
示例三，通过目标驾驶员交通工具中的录音器采集驾驶语音信息。
具体地，可以通过目标驾驶员交通工具中的录音器采集驾驶语音信息，比如目标驾驶员的语音信息，然后采用预设的信息处理算法从目标驾驶员的语音信息中提取驾驶行为特征为文明行驶，应用机器学习逻辑回归模型对文明行驶进行处理，生成目标驾驶员为乘客进行驾驶服务的评价信息为满意，更具体可以为5星、或者95分等。
可以理解的是，目标驾驶员可能出现不文明语音时，可以确定为不文明行驶，通过处理后，生成目标驾驶员为乘客进行驾驶服务的评价信息为不满意，更具体可以为2星、或者60分等。
由此，目标驾驶员的驾驶行为信息可以包括但不限于驾驶路线信息、驾驶视频信息和驾驶语音信息等中的一种或者多种。
第二种示例，接收乘客反馈的目标驾驶员为其进行驾驶服务的评价信息。
具体地，乘客可以在结束行程后，通过自身体验经历将对目标驾驶员为其进行驾驶服务的评价信息反馈给服务器。
需要说明的是，以上方式仅为获取目标驾驶员为乘客进行驾驶服务的评价信息的举例说明，可以根据实际应用需要选择或者设置其他的方式。
步骤302，根据乘客的第一特征、目标驾驶员的第二特征和评价信息对交通工具推荐模型的参数进行矫正训练。
具体地，在获取评价信息后，能够根据乘客的第一特征、目标驾驶员的第二特征和评价信息对交通工具推荐模型的参数进行矫正训练。进一步提高匹配的准确性，从而提高乘客出行体验。
由此，实现满足不用乘客的出行需求，比如能够针对外地游客推荐熟悉本地路况、景点信息的目标驾驶员；针对外国人推荐懂得相应外语的目标驾驶员；针对热爱美食的乘客推荐熟悉本地美食的驾驶员；针对购物达人，推荐熟悉本地商场打折活动的驾驶员等等。从而，提高出行乐趣，实现个性化的订制服务。
为了更加全面的说明本申请实施例的交通工具推荐方法，下面集中在第一终端侧描述本申请实施例的交通工具推荐方法。
图7为本申请实施例提供的第三种交通工具推荐方法的流程示意图。如图7所示，该交通工具推荐方法包括：
步骤401，向服务器发送使用请求，以使服务器根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具。
步骤402，接收服务器反馈的第一响应信息，其中，服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员。
可以理解的是，可以根据具体应用场景不同采用不同的方式输入使用请求，举例说明如下：
第一种示例，打开交通工具应用通过语音输入形式，比如“我要去天安门”等形式向服务器发送使用请求。
第二种示例，打开交通工具应用通过文字输入形式，输入“天安门”向服务器发送使用请求。
具体地，在向服务器发送使用请求后，服务器可以根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具，并将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员。
进而，能够接收到服务器第一响应信息。可以理解的是，通过对第一响应信息的解析，能够获取目标驾驶员信息，比如电话号码、评分等级等等，从而可以与目标驾驶员取得联系，更好满足乘客出行体验。
本申请实施例的交通工具推荐方法，通过向服务器发送使用请求，以使服务器根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具，然后接收服务器反馈的第一响应信息，其中，服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员。由此，实现了根据乘客特征和驾驶员特征灵活匹配确定目标交通工具，从而增加交通推荐方式的多样性，满足用户多方位的交通需求。
图8为本申请实施例提供的第四种交通工具推荐方法的流程示意图。如图8所示，该交通工具推荐方法包括：
步骤501，向服务器发送携带乘客标识信息的属性特征，以使服务器将乘客标识信息和属性特征的对应关系存储在数据库中的乘客属性注册信息中。
步骤502，向服务器发送携带乘客标识信息的偏好特征，以使服务器将乘客标识信息和偏好特征的对应关系存储在数据库中的乘客偏好注册信息中。
具体地，乘客可以通过第一终端预先上传乘客的属性特征、乘客的偏好特征等作为乘客的第一特征存储在服务器中。作为一种实现形式，乘客在第一终端上的交通工具应用注册时，交通工具应用提供信息填写页面以便乘客输入属性特征(比如年龄、性别等)和偏好特征(比如兴趣爱好、风俗习惯和乘车忌讳等)。
可以理解的是，在上述注册时每个乘客具有其唯一账号和密码、或者是电话号码等形式作为乘客标识信息，以确定乘客的唯一性，以及能够将乘客与其属性特征、偏好特征一一对应。
进而，服务器能够将乘客标识信息(比如账号密码)和属性特征的对应关系存储在数据库中的乘客属性注册信息中；将乘客标识信息和偏好特征的对应关系存储在数据库中的乘客偏好注册信息中。
步骤503，向服务器发送使用请求，以使服务器根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具。
需要说明的是，步骤503的描述与上述步骤401相对应，因此对步骤503的描述参考上述步骤401的描述，在此不再赘述。
步骤504，接收服务器反馈的使用场景设置界面。
步骤505，在使用场景设置界面中设置乘客的使用场景特征并发送给服务器。
具体地，为了进一步满足乘客的出行需求，可以设置乘客的使用场景特征。作为一种实现形式，比如女乘客晚上回家，可以接收服务器反馈的使用场景设置界面后，在使用场景设置界面中设置乘客的使用场景特征为女驾驶员发送给服务器。
步骤506，接收服务器反馈的第一响应信息，其中，服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员。
需要说明的是，步骤506的描述与上述步骤402相对应，因此对的步骤506的描述参考上述步骤402的描述，在此不再赘述。
步骤507，向服务器发送乘客反馈的目标驾驶员为其进行驾驶服务的评价信息。
具体地，乘客在结束行程后，可以通过自己的出行体验，对目标驾驶员的驾驶服务进行评价，并将评价信息反馈给服务器，以便服务器能够进一步根据评价信息对交通工具推荐模型进行矫正训练，即利用采集的数据对交通工具推荐模型进行实时更新。由此，进一步提高交通工具推荐的准确性，更贴近乘客的个性化需求。
为了更加全面的说明本申请实施例的交通工具推荐方法，下面集中在第二终端侧描述本申请实施例的交通工具推荐方法。
图9为本申请实施例提供的第五种交通工具推荐方法的流程示意图。如图9所示，该交通工具推荐方法包括：
步骤601，接收服务器发送的第二响应信息，其中，与第二终端对应的目标驾驶员是服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定的。
步骤602，当目标驾驶员到达乘客的位置时，向与乘客对应的第一终端发送到达响应信息。
具体地，在服务器可以应用预先训练的交通工具推荐模型等方式将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定的目标驾驶员后，与目标驾驶员对应的第二终端接收服务器发送的第二响应信息。
可以理解的是，通过对第二响应信息的解析，可以获取乘客要去的目的地、电话号码等信息。由此，在目标驾驶员到达目的地后，可以向乘客对应的第一终端发送到达响应信息。其形式可以是短信、电话等形式。
本申请实施例的交通工具推荐方法，通过接收服务器发送的第二响应信息，其中，与第二终端对应的目标驾驶员是服务器将乘客的第一特征分别与每个候选交通工具驾驶员 的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定的，并当目标驾驶员到达乘客的位置时，向与乘客对应的第一终端发送到达响应信息。由此，实现了根据乘客特征和驾驶员特征灵活匹配确定目标交通工具，从而增加交通推荐方式的多样性，满足用户多方位的交通需求。
图10为本申请实施例提供的第六种交通工具推荐方法的流程示意图。如图10所示，在步骤601之前，该交通工具推荐方法还包括：
步骤701，向服务器发送携带驾驶员标识信息的属性特征，以使服务器将驾驶员标识信息和属性特征的对应关系存储在数据库中的驾驶员属性注册信息中。
步骤702，向服务器发送携带驾驶员标识信息的偏好特征，以使服务器将驾驶员标识信息和偏好特征的对应关系存储在数据库中的驾驶员偏好注册信息中。
具体地，驾驶员可以通过第二终端预先上传驾驶员的属性特征、驾驶员的偏好特征等作为驾驶员的第二特征存储在服务器中。作为一种实现形式，驾驶员在第二终端上的交通工具应用注册时，交通工具应用提供信息填写页面以便驾驶员输入属性特征(比如年龄、性别等)和偏好特征(比如特殊才艺、擅长领域等)。
可以理解的是，在上述注册时每个驾驶员具有其唯一账号和密码、或者是电话号码等形式作为驾驶员标识信息，以确定驾驶员的唯一性，以及能够将驾驶员与其属性特征、偏好特征一一对应。
进而，服务器能够将驾驶员标识信息(比如账号密码)和属性特征的对应关系存储在数据库中的驾驶员属性注册信息中；将驾驶员标识信息和偏好特征的对应关系存储在数据库中的驾驶员偏好注册信息中。
由此，实现了根据乘客特征和驾驶员特征灵活匹配确定目标交通工具，从而提高了出行体验，满足乘客个性化的订制服务。
图11为本申请实施例的一个交通工具推荐方法的流程交互图。
参见图11，本实施例包括：服务器、第一终端和第二终端。本实施例通过设备之间的信息交互举例描述本实施例提供的交通工具推荐方法的流程的具体应用场景，具体说明如下：
步骤1.01，第一终端向服务器发送使用请求。
具体地，乘客需要乘坐交通工具时，打开第一终端上的交通工具应用，向服务器发送使用请求。作为一种场景实现：乘客通过手机打开打车应用“DD”，通过手动输入目的地“天安门”，乘车位置为“当前位置”等发送使用请求给服务器。
步骤1.02，服务器接收第一终端发送的使用请求，根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具。
步骤1.03，服务器获取乘客的第一特征，以及获取每个候选交通工具驾驶员的第二特征。
步骤1.04，服务器将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配， 根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员。
步骤1.05，服务器向第一终端反馈第一响应信息，向与目标驾驶员对应的第二终端发送第二响应信息。
具体地，服务器在接收到使用请求后，通过对使用请求的解析等方式，获取乘客的位置，即需要打车的具体地点，并在预设的调度范围筛选满足距离要求的多个候选交通工具。
进而，在预先的数据库中根据乘客标识信息、每个驾驶员标识信息获取乘客的第一特征，以及获取每个候选交通工具驾驶员的第二特征。从而，可以应用预先训练的交通工具推荐模型等方式将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员。
可以理解的是，与乘客特征匹配度最高的驾驶员有一个、或者多个，在存在对个的情况下时，从中选择距离最近的驾驶员最为目标驾驶员，向与目标驾驶员对应的第二终端发送第二响应信息，同时向第一终端反馈第一响应信息。
步骤1.06，第二终端向与乘客对应的第一终端发送到达响应信息。
由此，在目标驾驶员到达乘客的乘车位置时，向与乘客对应的第一终端发送到达响应信息。
本申请实施例的交通工具推荐方法，第一终端向服务器发送使用请求，服务器接收第一终端发送的使用请求后根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具，并获取乘客的第一特征和每个候选交通工具驾驶员的第二特征，接着将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，从而根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员，最后向第一终端反馈第一响应信息，以及向与目标驾驶员对应的第二终端发送第二响应信息，第二终端向与乘客对应的第一终端发送到达响应信息。由此，实现了根据乘客特征和驾驶员特征灵活匹配确定目标交通工具，从而提高了出行体验，满足乘客个性化的订制服务。
为了实现上述实施例，本申请还提出一种服务器。
图12为本申请实施例提供的一种服务器的结构示意图。该服务器包括：筛选模块11、第一获取模块12、第二获取模块13、匹配模块14、确定模块15和发送模块16。
其中，筛选模块11，用于接收第一终端发送的使用请求，根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具。
第一获取模块12，用于获取乘客的第一特征。
第二获取模块13，用于获取每个候选交通工具驾驶员的第二特征。
匹配模块14，用于将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配。
确定模块15，用于根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员。
发送模块16，用于向第一终端反馈第一响应信息，以及向与目标驾驶员对应的第二终端发送第二响应信息。
进一步地，第一获取模块12，具体用于根据乘客的标识信息查询数据库中的乘客属性注册信息，获取乘客的属性特征，和/或；根据乘客的标识信息查询数据库中的乘客偏好注册信息，获取乘客的偏好特征，和/或；向第一终端提供使用场景设置界面，根据设置信息获取乘客的使用场景特征。
进一步地，第二获取模块13，具体用于根据每个候选交通工具驾驶员的标识信息查询数据库中的驾驶员属性注册信息，获取每个候选交通工具驾驶员的属性特征，和/或；根据每个候选交通工具驾驶员的标识信息查询数据库中的驾驶员偏好注册信息，获取每个候选交通工具驾驶员的偏好特征。
进一步地，确定模块15，具体用于若匹配度最高的候选驾驶员为一个，则将候选驾驶员确定为目标驾驶员；若匹配度最高的候选驾驶员为多个，则选择与乘客的当前位置距离最近的候选驾驶员为目标驾驶员。
进一步，作为一种可能的实现方式，图13为本申请实施例提供的另一种服务器的结构示意图，如图13所示，在图12的基础上，该服务器还包括：第三获取模块17和生成模块18。
其中，第三获取模块17，用于获取打车应用的正样本数据集合和负样本数据集合；其中，正样本数据集合包括：乘客样本特征、驾驶员样本特征，以及正面评价信息的对应关系，负样本数据集合包括：乘客样本特征、驾驶员样本特征，以及负面评价信息的对应关系。
生成模块18，用于根据正样本数据集合和负样本数据集合训练模型参数生成交通工具推荐模型。
进一步，作为一种可能的实现方式，图14为本申请实施例提供的又一种服务器的结构示意图，如图14所示，在图13的基础上，该服务器还包括：第四获取模块19和训练模块110。
其中，第四获取模块19，用于获取目标驾驶员为乘客进行驾驶服务的评价信息。
训练模块20，用于根据乘客的第一特征、目标驾驶员的第二特征和评价信息对交通工具推荐模型的参数进行矫正训练。
图15为本申请实施例提供的一种第四获取模块的结构示意图。第四获取模块19包括：采集单元191、提取单元192和处理单元193。
其中，采集单元191，用于采集目标驾驶员的驾驶行为信息。
提取单元192，用于采用预设的信息处理算法从驾驶行为信息中提取驾驶行为特征。
处理单元193，用于应用机器学习逻辑回归模型对驾驶行为特征进行处理，生成目标驾驶员为乘客进行驾驶服务的评价信息。
进一步地，采集单元191，具体用于通过目标驾驶员交通工具中的导航设备采集驾驶路线信息；和/或；通过目标驾驶员交通工具中的摄像头采集驾驶视频信息；和/或；通过目标驾驶员交通工具中的录音器采集驾驶语音信息。
进一步地，第四获取模块19还包括：接收单元194。
其中，接收单元194，用于接收乘客反馈的目标驾驶员为其进行驾驶服务的评价信息。
需要说明的是，前述对交通工具推荐方法实施例的解释说明也适用于本实施例的服务器，此处不再赘述。
本申请实施例的服务器，通过接收第一终端发送的使用请求后根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具，并获取乘客的第一特征和每个候选交通工具驾驶员的第二特征，接着将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，从而根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员，最后向第一终端反馈第一响应信息，以及向与目标驾驶员对应的第二终端发送第二响应信息。由此，实现了根据乘客特征和驾驶员特征灵活匹配确定目标交通工具，从而增加交通推荐方式的多样性，满足用户多方位的交通需求。
图16为本申请实施例提供的一种第一终端的结构示意图。该第一终端包括：第一发送模块21和第一接收模块22。
其中，第一发送模块21，用于向服务器发送使用请求，以使服务器根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具。
第一接收模块22，用于接收服务器反馈的第一响应信息，其中，服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员。
进一步，作为一种可能的实现方式，图17为本申请实施例提供的又一种第一终端的结构示意图，如图17所示，在图16的基础上，该第一终端还包括：第二发送模块23、第三发送模块24、第二接收模块25、处理模块26和反馈模块27。
其中，第二发送模块23，用于向服务器发送携带乘客标识信息的属性特征，以使服务器将乘客标识信息和属性特征的对应关系存储在数据库中的乘客属性注册信息中，和/或；
第三发送模块24，用于向服务器发送携带乘客标识信息的偏好特征，以使服务器将乘客标识信息和偏好特征的对应关系存储在数据库中的乘客偏好注册信息中。
第二接收模块25，用于接收服务器反馈的使用场景设置界面。
处理模块26，用于在使用场景设置界面中设置乘客的使用场景特征并发送给服务器。
反馈模块27，用于向服务器发送乘客反馈的所述目标驾驶员为其进行驾驶服务的评价信息。
需要说明的是，前述对交通工具推荐方法实施例的解释说明也适用于本实施例的终端，此处不再赘述。
本申请实施例的第一终端，通过向服务器发送使用请求，以使服务器根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具，然后接收服务器反馈的第一响应信息，其中，服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行 匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员。由此，实现了根据乘客特征和驾驶员特征灵活匹配确定目标交通工具，从而增加交通推荐方式的多样性，满足用户多方位的交通需求。
图18为本申请实施例提供的一种第二终端的结构示意图。该第二终端包括：接收模块31和第一发送模块32。
其中，接收模块31，用于接收服务器发送的第二响应信息，其中，与第二终端对应的目标驾驶员是服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定的。
第一发送模块32，用于当目标驾驶员到达乘客的位置时，向与乘客对应的第一终端发送到达响应信息。
进一步，作为一种可能的实现方式，图19为本申请实施例提供的又一种第二终端的结构示意图，如图19所示，在图18的基础上，该第二终端还包括：第二发送模块33、第三发送模块34。
其中，第二发送模块33，用于向服务器发送携带驾驶员标识信息的属性特征，以使服务器将所述驾驶员标识信息和属性特征的对应关系存储在数据库中的驾驶员属性注册信息中，和/或；
第三发送模块34，用于向服务器发送携带驾驶员标识信息的偏好特征，以使服务器将所述驾驶员标识信息和偏好特征的对应关系存储在数据库中的驾驶员偏好注册信息中。
需要说明的是，前述对交通工具推荐方法实施例的解释说明也适用于本实施例的终端，此处不再赘述。
本申请实施例的第二终端，通过接收服务器发送的第二响应信息，其中，与第二终端对应的目标驾驶员是服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定的，并当目标驾驶员到达乘客的位置时，向与乘客对应的第一终端发送到达响应信息。由此，实现了根据乘客特征和驾驶员特征灵活匹配确定目标交通工具，从而增加交通推荐方式的多样性，满足用户多方位的交通需求。
图20为本申请实施例提供的一种打车推荐系统的结构示意图。该打车推荐系统包括：本申请图12至图15所述的服务器1、本申请图16至图17所述的第一终端2和本申请图18至图19所述第二终端3。
需要说明的是，前述对交通工具推荐方法实施例的解释说明也适用于本实施例的打车推荐系统，此处不再赘述。
本申请实施例的打车推荐系统，第一终端发送使用请求，服务器接收第一终端发送的使用请求后根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具，并获取乘客的第一特征和每个候选交通工具驾驶员的第二特征，接着将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，从而根据乘客与每个候选交通工具驾驶员之间 的特征匹配度确定目标驾驶员，最后向第一终端反馈第一响应信息，以及向与目标驾驶员对应的第二终端发送第二响应信息，第二终端向与乘客对应的第一终端发送到达响应信息。由此，实现了根据乘客特征和驾驶员特征灵活匹配确定目标交通工具，从而增加交通推荐方式的多样性，满足用户多方位的交通需求。
为了实现上述实施例，本申请还提出一种打车推荐装置。
图21为本申请实施例提供的一种打车推荐装置的结构示意图。
其中，需要说明的是，图21以服务器侧为例。如图21所示，该打车推荐装置可以包括：处理器11；用于存储处理器可执行指令的存储器12。其中，处理器11被配置为：接收第一终端发送的使用请求，根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具；获取乘客的第一特征，以及获取每个候选交通工具驾驶员的第二特征；将第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员；向第一终端反馈第一响应信息，以及向与目标驾驶员对应的第二终端发送第二响应信息。
图22为本申请实施例提供的另一种打车推荐装置的结构示意图。
其中，需要说明的是，图22以第一终端侧为例。如图22所示，该打车推荐装置可以包括：处理器11；用于存储处理器可执行指令的存储器12。其中，处理器11被配置为：向服务器发送使用请求，以使服务器根据乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具；接收服务器反馈的第一响应信息，其中，服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员。
图23为本申请实施例提供的又一种打车推荐装置的结构示意图。
其中，需要说明的是，图23以第二终端侧为例。如图23所示，该打车推荐装置可以包括：处理器11；用于存储处理器可执行指令的存储器12。其中，处理器11被配置为：接收服务器发送的第二响应信息，其中，与第二终端对应的目标驾驶员是所述服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配，根据乘客与每个候选交通工具驾驶员之间的特征匹配度确定的；当目标驾驶员到达乘客的乘车位置时，向与乘客对应的第一终端发送到达响应信息。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本申请的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不必须针对的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任一个或多个实施例或示例中以合适的方式结合。此外，在不相互矛盾的情况下，本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性 或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。在本申请的描述中，“多个”的含义是至少两个，例如两个，三个等，除非另有明确具体的限定。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为，表示包括一个或更多个用于实现定制逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分，并且本申请的优选实施方式的范围包括另外的实现，其中可以不按所示出或讨论的顺序，包括根据所涉及的功能按基本同时的方式或按相反的顺序，来执行功能，这应被本申请的实施例所属技术领域的技术人员所理解。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用，或结合这些指令执行系统、装置或设备而使用。就本说明书而言，"计算机可读介质"可以是任何可以包含、存储、通信、传播或传输程序以供指令执行系统、装置或设备或结合这些指令执行系统、装置或设备而使用的装置。计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(电子装置)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。
应当理解，本申请的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器中且由合适的指令执行系统执行的软件或固件来实现。如，如果用硬件来实现和在另一实施方式中一样，可用本领域公知的下列技术中的任一项或他们的组合来实现：具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路，具有合适的组合逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。
本技术领域的普通技术人员可以理解实现上述实施例方法携带的全部或部分步骤是可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，该程序在执行时，包括方法实施例的步骤之一或其组合。
此外，在本申请各个实施例中的各功能单元可以集成在一个处理模块中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。
上述提到的存储介质可以是只读存储器，磁盘或光盘等。尽管上面已经示出和描述了本申请的实施例，可以理解的是，上述实施例是示例性的，不能理解为对本申请的限制，本领域的普通技术人员在本申请的范围内可以对上述实施例进行变化、修改、替换和变型。

Claims (18)

  1. 一种交通工具推荐方法，其中，所述方法应用于服务器，包括：
    接收第一终端发送的使用请求；
    根据与所述使用请求对应的乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具；
    获取所述乘客的第一特征，以及获取每个候选交通工具驾驶员的第二特征；
    将所述第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度；
    根据所述特征匹配度确定目标驾驶员；
    向所述第一终端反馈第一响应信息，以及向与所述目标驾驶员对应的第二终端发送第二响应信息。
  2. 如权利要求1所述的方法，其中，所述根据所述乘客与每个候选交通工具驾驶员之间的特征匹配度确定目标驾驶员，包括：
    若匹配度最高的候选驾驶员为一个，则将所述候选驾驶员确定为目标驾驶员；
    若匹配度最高的候选驾驶员为多个，则选择与所述乘客的当前位置距离最近的候选驾驶员为目标驾驶员。
  3. 如权利要求1所述的方法，其中，在所述将所述第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配之前，还包括：
    获取乘客使用交通工具的正样本数据集合和负样本数据集合；其中，所述正样本数据集合包括：乘客样本特征、驾驶员样本特征，以及正面评价信息的对应关系，所述负样本数据集合包括：乘客样本特征、驾驶员样本特征，以及负面评价信息的对应关系；
    根据所述正样本数据集合和所述负样本数据集合训练模型参数生成交通工具推荐模型，以应用所述交通工具推荐模型确定所述目标驾驶员。
  4. 如权利要求3所述的方法，其中，还包括：
    获取所述目标驾驶员为乘客进行驾驶服务的评价信息；
    根据所述乘客的第一特征、所述目标驾驶员的第二特征和所述评价信息对所述交通工具推荐模型的参数进行矫正训练。
  5. 如权利要求4所述的方法，其中，所述获取所述目标驾驶员为乘客进行驾驶服务的评价信息，包括：
    采集所述目标驾驶员的驾驶行为信息；
    采用预设的信息处理算法从所述驾驶行为信息中提取驾驶行为特征；
    应用机器学习逻辑回归模型对所述驾驶行为特征进行处理，生成所述目标驾驶员为所述乘客进行驾驶服务的评价信息。
  6. 如权利要求5所述的方法，其中，所述采集所述目标驾驶员的驾驶行为信息，包括：
    通过所述目标驾驶员交通工具中的导航设备采集驾驶路线信息；和/或；
    通过所述目标驾驶员交通工具中的摄像头采集驾驶视频信息；和/或；
    通过所述目标驾驶员交通工具中的录音器采集驾驶语音信息。
  7. 如权利要求4所述的方法，其中，所述获取所述目标驾驶员为乘客进行驾驶服务的评价信息，包括：
    接收所述乘客反馈的所述目标驾驶员为所述乘客进行驾驶服务的评价信息。
  8. 如权利要求1所述的方法，其中，所述获取所述乘客的第一特征，包括：
    根据所述乘客的标识信息查询数据库中的乘客属性注册信息，获取所述乘客的属性特征；和/或
    根据所述乘客的标识信息查询数据库中的乘客偏好注册信息，获取所述乘客的偏好特征；和/或
    向所述第一终端提供使用场景设置界面，根据设置信息获取所述乘客的使用场景特征。
  9. 如权利要求1所述的方法，其中，所述获取每个候选交通工具驾驶员的第二特征，包括：
    根据每个候选交通工具驾驶员的标识信息查询数据库中的驾驶员属性注册信息，获取每个候选交通工具驾驶员的属性特征；和/或
    根据每个候选交通工具驾驶员的标识信息查询数据库中的驾驶员偏好注册信息，获取每个候选交通工具驾驶员的偏好特征。
  10. 一种交通工具推荐方法，其中，所述方法应用于第一终端，包括：
    向服务器发送使用请求，以使所述服务器根据与所述使用请求对应的乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具；
    接收所述服务器反馈的第一响应信息，其中，所述服务器将所述乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度，并根据所述特征匹配度确定目标驾驶员。
  11. 如权利要求10所述的方法，其中，在所述向服务器发送使用请求之前，还包括：
    向所述服务器发送携带乘客标识信息的属性特征，以使所述服务器将所述乘客标识信息和所述属性特征的对应关系存储在数据库中的乘客属性注册信息中，和/或；
    向所述服务器发送携带乘客标识信息的偏好特征，以使所述服务器将所述乘客标识信息和所述偏好特征的对应关系存储在数据库中的乘客偏好注册信息中。
  12. 如权利要求10所述的方法，其中，在所述向服务器发送使用请求之后，还包括：
    接收所述服务器反馈的使用场景设置界面；
    在所述使用场景设置界面中设置所述乘客的使用场景特征并发送给所述服务器。
  13. 一种交通工具推荐方法，其中，所述方法应用于第二终端，包括：
    接收服务器发送的第二响应信息，其中，与所述第二终端对应的目标驾驶员是所述服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹 配度，并根据所述特征匹配度确定的；
    当所述目标驾驶员到达乘客的位置时，向与所述乘客对应的第一终端发送到达响应信息。
  14. 如权利要求13所述的方法，其中，在所述接收服务器发送的第二响应信息之前，还包括：
    向所述服务器发送携带驾驶员标识信息的属性特征，以使所述服务器将所述驾驶员标识信息和所述属性特征的对应关系存储在数据库中的驾驶员属性注册信息中，和/或；
    向所述服务器发送携带驾驶员标识信息的偏好特征，以使所述服务器将所述驾驶员标识信息和所述偏好特征的对应关系存储在数据库中的驾驶员偏好注册信息中。
  15. 一种服务器，其中，包括：
    筛选模块，用于接收第一终端发送的使用请求，根据与所述使用请求对应的乘客的位置和预设的调度范围筛选满足距离要求的多个候选交通工具；
    第一获取模块，用于获取所述乘客的第一特征；
    第二获取模块，用于获取每个候选交通工具驾驶员的第二特征；
    匹配模块，用于将所述第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度；
    确定模块，用于根据所述特征匹配度确定目标驾驶员；
    发送模块，用于向所述第一终端反馈第一响应信息，以及向与所述目标驾驶员对应的第二终端发送第二响应信息。
  16. 一种第一终端，其中，包括：
    第一发送模块，用于向服务器发送使用请求，以使所述服务器根据与所述使用请求对应的乘客的乘车位置和预设的调度范围筛选满足距离要求的多个候选交通工具；
    第一接收模块，用于接收所述服务器反馈的第一响应信息，其中，所述服务器将所述乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度，并根据所述特征匹配度确定目标驾驶员。
  17. 一种第二终端，其中，包括：
    接收模块，用于接收服务器发送的第二响应信息，其中，与所述第二终端对应的目标驾驶员是所述服务器将乘客的第一特征分别与每个候选交通工具驾驶员的第二特征进行匹配得到特征匹配度，并根据所述特征匹配度确定的；
    第一发送模块，用于当所述目标驾驶员到达乘客的位置时，向与所述乘客对应的第一终端发送到达响应信息。
  18. 一种交通工具推荐系统，其中，包括：
    如权利要求15所述的服务器；
    如权利要求16所述的第一终端；
    以及如权利要求17所述的第二终端。
PCT/CN2018/084763 2017-05-05 2018-04-27 交通工具推荐方法、系统及其设备 Ceased WO2018201979A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710313989.5 2017-05-05
CN201710313989.5A CN107194762A (zh) 2017-05-05 2017-05-05 交通工具推荐方法、系统及其设备

Publications (1)

Publication Number Publication Date
WO2018201979A1 true WO2018201979A1 (zh) 2018-11-08

Family

ID=59872401

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/084763 Ceased WO2018201979A1 (zh) 2017-05-05 2018-04-27 交通工具推荐方法、系统及其设备

Country Status (2)

Country Link
CN (1) CN107194762A (zh)
WO (1) WO2018201979A1 (zh)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109886440A (zh) * 2019-01-08 2019-06-14 平安科技（深圳）有限公司 网约车控制方法、装置、计算机设备及存储介质
CN113869864A (zh) * 2021-09-29 2021-12-31 苏州达家迎信息技术有限公司 设备调度方法、装置、设备及存储介质
US11367121B2 (en) * 2020-06-15 2022-06-21 Capital One Services, Llc Recommendation engine that utilizes travel history to recommend vehicles for customers
CN115169432A (zh) * 2022-05-25 2022-10-11 上海钧正网络科技有限公司 司机类型的确定方法和装置、数据处理方法
CN117218879A (zh) * 2023-09-05 2023-12-12 广州小鹏汽车科技有限公司 无人驾驶交通工具调度方法、电子装置、系统及存储介质
CN118797169A (zh) * 2024-09-11 2024-10-18 浙江吉利控股集团有限公司 一种主动提供车辆服务的方法、设备、介质及程序产品
CN118822818A (zh) * 2024-06-26 2024-10-22 北京白龙马云行科技有限公司 基于异质代理人模型及SaaS的网约车行为分析方法及系统
WO2025256224A1 (zh) * 2024-06-12 2025-12-18 浙江吉利控股集团有限公司 车辆控制处理方法、装置、设备及介质

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107194762A (zh) * 2017-05-05 2017-09-22 腾讯科技（深圳）有限公司 交通工具推荐方法、系统及其设备
CN108038707A (zh) * 2017-11-01 2018-05-15 平安科技（深圳）有限公司 项目推荐方法、电子设备及计算机可读存储介质
CN108153829B (zh) * 2017-12-12 2021-12-28 北京字节跳动网络技术有限公司 一种简历评估方法及装置
CN108009735B (zh) * 2017-12-12 2022-07-22 北京字节跳动网络技术有限公司 一种简历评估方法及装置
CN107993019B (zh) * 2017-12-12 2022-07-22 北京字节跳动网络技术有限公司 一种简历评估方法及装置
CN108182512B (zh) * 2017-12-12 2022-07-22 北京字节跳动网络技术有限公司 一种简历评估方法及装置
CN108376354A (zh) * 2018-01-10 2018-08-07 链家网（北京）科技有限公司 一种基于网络图结构的推荐方法及装置
CN110619551A (zh) * 2018-06-19 2019-12-27 北京嘀嘀无限科技发展有限公司 订单分配方法、订单分配系统、计算机设备及存储介质
CN110689150A (zh) * 2018-06-19 2020-01-14 北京嘀嘀无限科技发展有限公司 基于机器学习的订单预测方法、预测装置及计算机设备
CN110839346A (zh) * 2018-06-19 2020-02-25 北京嘀嘀无限科技发展有限公司 用于分配服务请求的系统和方法
CN111353104A (zh) * 2018-12-21 2020-06-30 深圳市优必选科技有限公司 车辆查询方法、系统、装置、计算机设备及存储介质
CN111442779B (zh) 2019-01-16 2022-03-22 百度在线网络技术（北京）有限公司 出行方式推荐方法及装置
CN109919708A (zh) * 2019-01-24 2019-06-21 平安科技（深圳）有限公司 一种打车软件派单方法和装置
CN111756783B (zh) * 2019-03-29 2021-11-12 比亚迪股份有限公司 终端的信息推送方法、服务器和终端的信息推送系统
CN110929157A (zh) * 2019-11-28 2020-03-27 深圳市元征科技股份有限公司 用于共享出行的数据推荐方法、装置、及可读存储介质
CN111858847A (zh) * 2020-05-19 2020-10-30 北京嘀嘀无限科技发展有限公司 一种订单处理方法、装置、电子设备及可读存储介质
CN114519438A (zh) * 2020-11-19 2022-05-20 上海擎感智能科技有限公司 预约驾驶员的方法及电子设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103996290A (zh) * 2014-06-09 2014-08-20 北京东方车云信息技术有限公司 一种提供叫车服务的方法、服务器及系统
CN105489001A (zh) * 2015-12-11 2016-04-13 惠州Tcl移动通信有限公司 一种出租车调度优化方法及系统
CN105719173A (zh) * 2016-01-19 2016-06-29 滴滴（中国）科技有限公司 订单处理方法和订单处理设备
CN105894359A (zh) * 2016-03-31 2016-08-24 百度在线网络技术（北京）有限公司 订单推送方法、装置及系统
CN105956908A (zh) * 2016-05-13 2016-09-21 深圳市永兴元科技有限公司 订单分配方法及系统
CN107194762A (zh) * 2017-05-05 2017-09-22 腾讯科技（深圳）有限公司 交通工具推荐方法、系统及其设备

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7627422B2 (en) * 2003-06-24 2009-12-01 At&T Intellectual Property I, Lp Methods, systems and computer program products for ride matching based on selection criteria and drive characteristic information
CN106127586A (zh) * 2016-06-17 2016-11-16 上海经达信息科技股份有限公司 大数据时代下车险费率辅助决策系统
CN106372817A (zh) * 2016-10-25 2017-02-01 先锋智道（北京）科技有限公司 一种司机驾驶行为生成方法、装置及系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103996290A (zh) * 2014-06-09 2014-08-20 北京东方车云信息技术有限公司 一种提供叫车服务的方法、服务器及系统
CN105489001A (zh) * 2015-12-11 2016-04-13 惠州Tcl移动通信有限公司 一种出租车调度优化方法及系统
CN105719173A (zh) * 2016-01-19 2016-06-29 滴滴（中国）科技有限公司 订单处理方法和订单处理设备
CN105894359A (zh) * 2016-03-31 2016-08-24 百度在线网络技术（北京）有限公司 订单推送方法、装置及系统
CN105956908A (zh) * 2016-05-13 2016-09-21 深圳市永兴元科技有限公司 订单分配方法及系统
CN107194762A (zh) * 2017-05-05 2017-09-22 腾讯科技（深圳）有限公司 交通工具推荐方法、系统及其设备

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109886440A (zh) * 2019-01-08 2019-06-14 平安科技（深圳）有限公司 网约车控制方法、装置、计算机设备及存储介质
US11367121B2 (en) * 2020-06-15 2022-06-21 Capital One Services, Llc Recommendation engine that utilizes travel history to recommend vehicles for customers
US11928721B2 (en) 2020-06-15 2024-03-12 Capital One Services, Llc Recommendation engine that utilizes travel history to recommend vehicles for customers
CN113869864A (zh) * 2021-09-29 2021-12-31 苏州达家迎信息技术有限公司 设备调度方法、装置、设备及存储介质
CN115169432A (zh) * 2022-05-25 2022-10-11 上海钧正网络科技有限公司 司机类型的确定方法和装置、数据处理方法
CN117218879A (zh) * 2023-09-05 2023-12-12 广州小鹏汽车科技有限公司 无人驾驶交通工具调度方法、电子装置、系统及存储介质
WO2025256224A1 (zh) * 2024-06-12 2025-12-18 浙江吉利控股集团有限公司 车辆控制处理方法、装置、设备及介质
CN118822818A (zh) * 2024-06-26 2024-10-22 北京白龙马云行科技有限公司 基于异质代理人模型及SaaS的网约车行为分析方法及系统
CN118797169A (zh) * 2024-09-11 2024-10-18 浙江吉利控股集团有限公司 一种主动提供车辆服务的方法、设备、介质及程序产品

Also Published As

Publication number Publication date
CN107194762A (zh) 2017-09-22

Similar Documents

Publication Publication Date Title
WO2018201979A1 (zh) 交通工具推荐方法、系统及其设备
AU2018284301B2 (en) Customized communications for network systems
CN110249349B (zh) 用于定制共享车辆中的舒适性的系统和方法
US10989552B2 (en) Systems and methods for adaptive content filtering
CN104618858B (zh) 利用公交卡进行信息推送的智能控制方法
US20140108307A1 (en) Methods and systems for providing personalized and context-aware suggestions
CN111859185A (zh) 一种上车点推荐方法、系统、装置及存储介质
KR101551037B1 (ko) 차량에서의 정보 제공을 위한 시스템
JP6915195B2 (ja) 指定受付システム、検索システム、検索端末、指定受付プログラム、検索プログラムおよび検索端末制御プログラム
CN105431882A (zh) 环境感知的分布式出租车汽车调度
US20190108559A1 (en) Evaluation-information generation system and vehicle-mounted device
US12056198B2 (en) Method and apparatus for enhancing a geolocation database
CN108292308A (zh) 根据移动习惯为用户推荐车辆/乘客资源
CN106327311B (zh) 订单处理方法、装置及系统
CN107274257A (zh) 一种酒店服务推荐方法及装置
WO2022142493A1 (zh) 业务数据处理方法、装置、电子设备和存储介质
CN109214878A (zh) 业务处理方法及装置
CN112927052A (zh) 基于多人协同决策及多出发地的商户推荐方法和装置
KR20230038166A (ko) 현지인 여행 플래너 매칭 서비스를 통한 여행 상품 제공 방법, 장치 및 컴퓨터프로그램
CN116049548A (zh) 车辆服务的推送方法及装置
CN108351886A (zh) 确定车辆使用者共同兴趣的系统
JP2020017130A (ja) 情報処理装置、情報処理システム、及び、情報処理方法
US11869048B2 (en) User recommendations and reviews using autonomous vehicle information
US20190353491A1 (en) Vehicle Navigation System
US20200111239A1 (en) Image generation apparatus, image generation method, and non-transitory recording medium recording program

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18795030

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18795030

Country of ref document: EP

Kind code of ref document: A1