[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

WO2020147597A1 - 一种基于社交大数据的拼车优化方法、索引结构及系统 - Google Patents

一种基于社交大数据的拼车优化方法、索引结构及系统 Download PDF

Info

Publication number
WO2020147597A1
WO2020147597A1 PCT/CN2020/070369 CN2020070369W WO2020147597A1 WO 2020147597 A1 WO2020147597 A1 WO 2020147597A1 CN 2020070369 W CN2020070369 W CN 2020070369W WO 2020147597 A1 WO2020147597 A1 WO 2020147597A1
Authority
WO
WIPO (PCT)
Prior art keywords
carpool
candidate
social
commuting
big data
Prior art date
Application number
PCT/CN2020/070369
Other languages
English (en)
French (fr)
Inventor
夏吉喆
李清泉
乐阳
朱家松
涂伟
周宝定
Original Assignee
深圳大学
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 深圳大学 filed Critical 深圳大学
Publication of WO2020147597A1 publication Critical patent/WO2020147597A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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 invention relates to the technical field of intelligent transportation, in particular to a carpool optimization method, index structure and system based on social big data.
  • Carpooling has become a popular and convenient transportation service, and it is also an effective solution to traffic congestion, which can reduce travel costs and reduce related traffic pollution and energy consumption problems.
  • the existing carpooling methods are only aimed at optimizing the cost of carpooling for two persons, which cannot meet the carpooling needs of a large number of passengers, and has low intelligence and efficiency.
  • the technical problem to be solved by the present invention is to provide a carpool optimization method, index structure and system based on social big data in view of the above-mentioned defects of the prior art, aiming to optimize the carpool model based on social big data and road network at the same time Carpool team trust level and commuting cost, in order to obtain the maximum trust and minimum commuting cost carpooling plan, reduce complexity, reduce road congestion, reduce travel costs, and facilitate users.
  • a carpool optimization method based on social big data includes the following steps:
  • the association between the social relationship and the trust rating index is established in advance, and different trust ratings corresponding to different levels are set and stored at the same time.
  • the obtaining all carpool candidates satisfying the social relationship according to the social relationship of the driver specifically includes:
  • the social relationships include direct friends, mutual friends, and strangers;
  • the carpooling optimization method based on social big data, wherein, according to the trust level index between the driver and each carpooling candidate, the driver and several carpooling candidates are combined for carpooling, and several carpooling candidates are established
  • the team includes:
  • the driver will be combined with several carpool candidates to establish several candidate carpool team groups;
  • the trust model of the candidate carpool team is constructed, and the overall trust of each candidate carpool team is calculated.
  • the method for carpooling optimization based on social big data wherein, the commuting information of each carpool candidate and driver in each candidate carpooling team group is obtained, and the commuting route and commuting cost of each candidate carpooling team group are determined.
  • the first commute information includes travel time, departure place and destination, and the second commute The information includes departure time, departure place and end of stop;
  • the first commuting information and the second commuting information determine the commuting route network of each candidate carpool team group
  • a commuter cost model of the candidate carpool team is constructed.
  • the carpool optimization method based on social big data wherein the construction of a carpool model based on social big data and road network, according to the trust level optimization and commuting cost optimization of each candidate carpool team group, matching to obtain maximum trust
  • the best carpool team group with the least commuting cost includes:
  • the objective function of the trust model is processed to maximize the degree of trust under the condition of satisfying the trust constraint, so that the value of the objective function of the trust model is maximized, and at the same time, the objective function of the commuting cost model is minimized under the condition of satisfying the commuting cost constraint Processing to minimize the value of the objective function of the commuting cost model;
  • the value of the trust weight variable is continuously changed and substituted into the objective function of the carpool model to maximize the value of the objective function of the carpool model, thereby obtaining the best carpool team group with the largest trust value and the smallest commuting cost.
  • the method for carpooling optimization based on social big data further includes: establishing a spatiotemporal carpool matching model, and optimizing the time cost under the time constraint condition, for constructing the carpooling model based on social big data and road network.
  • the factors affecting commuting cost include one or more of commuting route distance, driving time, congestion factor, oil cost, toll fee, and cost factors related to commuting.
  • the present invention also provides a car-pooling optimization system based on social big data.
  • the car-pooling optimization system based on social big data includes a number of terminals, a social network center connected to the terminal, a transportation network center, and a social network center.
  • the back-end servers respectively connected to the transportation network center, when the social big data-based carpool optimization system is executed, the aforementioned social big data-based carpool optimization method is implemented.
  • the present invention also provides a carpool optimization index structure based on social big data.
  • the carpool optimization index structure includes a memory connected to a processor and the processor, and the memory stores a carpool optimization program based on social big data.
  • the big data carpool optimization program is executed by the processor, it is used to realize the aforementioned social big data-based carpool optimization method.
  • the present invention discloses a carpooling optimization method, index structure and system based on social big data.
  • the carpooling optimization method based on social big data includes: obtaining all carpool candidates satisfying the social relationship according to the social relationship of the driver ;According to the trust level index between the driver and each carpool candidate, the driver and several carpool candidates are combined to form a number of candidate carpool team groups; each carpool candidate in each candidate carpool team group is obtained
  • the commuting information of people and drivers determines the commuting route and commuting cost of each candidate carpool team; builds a carpool model based on social big data and road network, optimizes the trust level and commute cost of each candidate carpool team, Match the best carpool team group that maximizes trust and minimizes commuting costs.
  • the present invention optimizes the carpool team trust level and commuting cost through the carpool model based on social big data and road network, so as to obtain the carpool scheme with the greatest trust degree and the least commuting cost, reduce complexity, reduce road congestion, and reduce travel cost and traffic Control costs and facilitate users.
  • Fig. 1 is a flowchart of a first preferred embodiment of a carpool optimization method based on social big data of the present invention.
  • Fig. 2 is a schematic diagram of carpool activities in the carpool optimization method based on social big data of the present invention.
  • Fig. 3 is a structural block diagram of the carpool optimization system based on social big data of the present invention.
  • Fig. 4 is a structural block diagram of an optimized index structure for carpooling based on social big data in the present invention.
  • FIG. 1 is a flowchart of a first preferred embodiment of a carpool optimization method based on social big data of the present invention. As shown in Figure 1, the carpool optimization method based on social big data includes the following steps:
  • Step S100 According to the social relationship of the driver, all carpooling candidates satisfying the social relationship are obtained.
  • the association between the social relationship and the trust level index is established in advance, and different levels are set to correspond to different trust level indexes, and stored at the same time.
  • the social relationship includes direct friend relationship, mutual friend relationship and stranger relationship, as shown in Table 1 below. In Table 1, it is assumed that a person has 100 direct friends.
  • the division of the trust level index is obtained according to the six-degree separation theory. In the embodiment of the present invention, based on the scale of the social network in the car-pooling activity, potential carpool candidates are screened according to the six-degree separation theory.
  • the direct friend relationship means that the two in Figure 2 are directly connected, which is defined as 1 degree of separation;
  • the common friend relationship means that there is at least one friend in common between the two in Figure 2, and is indirectly connected. It is defined as a 2-degree separation;
  • a strange friend relationship means that there is no mutual friend between the two in Figure 2, that is, a stranger, which is defined as a 3-degree separation;
  • the absence of any chain friendship relationship represents the marginal blank area in Figure 2.
  • Arbitrary carpool arrangements are made according to the degree of separation corresponding to the trust level index to form an arbitrary carpool team. Figure 2 only illustrates one of the carpool arrangements.
  • step S100 includes:
  • Step S101 Obtain social relationships of all drivers based on the social network; wherein, the social relationships include direct friend relationships, mutual friend relationships, and stranger relationships;
  • Step S102 Search and obtain carpool candidates that satisfy all the social relationships of each driver in the carpool candidate database.
  • step S200 according to the trust level index between the driver and each carpool candidate, the driver is combined with several carpool candidates to establish several candidate carpool team groups.
  • step S200 specifically includes:
  • Step S201 obtaining the association between the social relationship and the trust level index
  • Step S202 Determine the trust level index between the driver and each carpool candidate according to the association and the social relationship between the driver and each carpool candidate;
  • Step S203 According to the capacity of the vehicle, the driver and several carpool candidates are combined to form a carpool group to establish a number of candidate carpool teams;
  • Step S204 Construct a trust model of the candidate carpool team based on the trust level index between the driver and each carpool candidate in each candidate carpool team group, and calculate the overall trust of each candidate carpool team group.
  • the objective function of the trust model of the candidate carpool team is as shown in formula (1):
  • the objective function (1) aims to maximize the trust between passengers and drivers in the entire carpool, and the trust constraint condition (2) requires a given number of passengers to be allocated to the carpool team; c represents the number of carpool candidates ; CP represents the carpool capacity, N ⁇ CP and N is a natural number; Tc represents the trust level index between the driver and the carpool candidate; Xc represents 1 when the carpool candidate is selected and combined into a candidate carpool team, no If selected, it is 0.
  • carpooling candidates are 10 people (c1, c2,...,c10), carpooling candidates are c2, c4, c5, c6, and the vehicle capacity CP is 4
  • Tc1 represents the trust rating index of the carpool members of the carpool candidate c1.
  • the objective function (1) of the trust model of the candidate carpool team is selected to maximize the value of the trust level index.
  • the value of the objective function (1) is the largest, the trust among the corresponding carpool candidate teams is the highest. In this way, the entire team of carpooling candidates is more stable, with a stronger experience, and the highest natural travel efficiency.
  • Step S300 Obtain commuting information of each carpool candidate and driver in each candidate carpool team group, and determine the commuting route and commuting cost of each candidate carpool team group.
  • step S300 specifically includes:
  • Step S301 Obtain the first commute information of each carpool candidate in each candidate carpool team and the second commute information of the driver;
  • Step S302 according to the first commuting information and the second commuting information, determine the commuting route network of each candidate carpool team group;
  • Step S303 at the same time, construct a commuting cost model of the candidate carpool team according to the commuter route network.
  • the first commute information includes the travel time, departure place and destination of the carpool candidate (passenger), and the second commute information includes the departure time of the driver, the departure place and the end of the stop
  • the starting point of the vehicle refers to the location of the driver's residence
  • the end point of the parking refers to the location of the driver's workplace.
  • the commuting cost in the embodiment of the present invention is defined as the cost incurred in the process of picking up and dropping carpooling candidates and not only refers to the additional cost of driving alone.
  • the route between the residence location and the workplace location is defined as a commuter route, and the residence location and the workplace location are positioned as nodes on the commuter route. Therefore, starting from the current actual residence location, the driver has several different routes to receive each team member, so that different emergence routes are established, thus forming a commuting route network.
  • a complete commuter route means that the driver starts from the current actual residence location, receives all other team members in a predetermined order, puts all team members to their destination node, and finally arrives at the driver’s workplace node.
  • CP represents the size of the carpool team, that is, the total number of people in the carpool team
  • VP represents the capacity of the vehicle
  • s represents the driver’s residence
  • t represents the driver’s workplace
  • t represents the driver’s workplace
  • i and j are the index of the node (house or workplace)
  • d ij is the impedance i and node j of the edge between nodes; d ij represents any cost function suitable for the driver to commute.
  • the factors affecting commuting cost include commuting route distance, travel time, congestion factor, oil cost, toll And one or more of the cost factors related to commuting;
  • the above objective function (3) aims to minimize the commuting cost of the entire carpool route.
  • Commuter constraints (4) and (5) require that the carpool route must start from the driver's residence and end with the driver's workplace.
  • Commuter constraints (6) and (7) are used to indicate that the driver picks up a given number of passengers (CP at the residence) and drops a given number of passengers (CP at the workplace).
  • the commuting constraint (8) is used to indicate that the driver will not throw passengers off at will to form a logical commuting path.
  • the commuting constraint (9) is used to indicate that only one edge can be selected for any step of the carpool route.
  • the commuting constraint (10) and the commuting constraint (11) ensure that only one edge enters or exits the selected node.
  • the commuting constraint (12) ensures the topology of the commuter route network, and the commuting constraint (13) is used to indicate that the size of the carpool team is smaller than the capacity of the vehicle.
  • the aforementioned impedance associated with the edge represents a cost function, that is, it can be any combination of factors that affect commuting costs.
  • the commuting cost of the objective function (3) must be minimized in the end, that is, the commuting cost should be minimized.
  • Step S400 Construct a carpool model based on social big data and road network, optimize the trust level and commute cost of each candidate carpool team group, and match the best carpool team group that maximizes trust and minimizes commuting cost.
  • step S400 specifically includes:
  • Step S401 set a trust weight variable (W);
  • Step S402 construct a carpool model based on social big data and road network according to the obtained trust model and commuting cost model of the candidate carpool team and trust weight variables;
  • step S403 the objective function of the trust model is processed to maximize the degree of trust under the condition of satisfying the trust constraint, so that the value of the objective function of the trust model is maximized, and the objective function of the commuting cost model is performed under the condition of satisfying the commuting cost constraint.
  • step S404 the value of the trust weight variable is continuously changed at the same time, and substituted into the objective function of the carpool model respectively, so that the value of the objective function of the carpool model is maximized, thereby obtaining the best carpool with the largest trust value and the smallest commuting cost Team group.
  • W represents the trust weight variable, which is used to balance the trust of the carpool team and the commuting cost.
  • the value of W ranges from [0,1]. As the value increases, W continuously enhances social comfort; Trust normalized represents the entire carpool The normalized trust value of the team; Cost normalized represents the normalized cost value of the carpool commuting route, 0 ⁇ Cost normalized ⁇ 1, Trust team and Cost carpool correspond to formulas (1) and (3), Cost original represents commuting without passengers cost.
  • the value of the objective function (14) is made the highest.
  • the best carpool team group with the largest trust value and the smallest commuting cost can be obtained.
  • the carpool team with high trust and the lowest commuting cost is obtained, which reduces complexity, reduces road congestion, reduces travel costs, and facilitates users.
  • FIG. 3 illustrates a structural block diagram of the carpool optimization system based on social big data.
  • the carpool optimization system based on social big data
  • the optimization system includes several terminals 100, a social network center 201 connected to the terminal 100, a transportation network center 202, and a backend server 301 and a backend server 302 respectively connected to the social network center 201 and the transportation network center 202.
  • the terminal 100 may be a mobile phone, a computer, a tablet, or the like.
  • the terminal 100 includes a memory 20 connected to the processor 10 and the processor 10.
  • the memory 20 stores a carpool optimization program based on social big data.
  • the processor 10 may be a central processing unit (CPU), microprocessor or other data processing chip, which is used to run the program code or process data stored in the memory 20, For example, execute a carpool optimization program based on social big data, etc.
  • CPU central processing unit
  • microprocessor or other data processing chip, which is used to run the program code or process data stored in the memory 20, For example, execute a carpool optimization program based on social big data, etc.
  • the system for carpooling optimization based on social big data when executed, it is used to realize the steps of the method for carpooling optimization based on social big data, as described in the first embodiment.
  • the present invention also provides a carpool optimization index structure based on social big data.
  • the carpool optimization index structure includes a first memory 12 connected to a first processor 11 and a first processor 11.
  • a memory 12 stores a carpool optimization program based on social big data, and the first processor 11 executes the above-mentioned carpool optimization method based on social big data; the specifics are as described above.
  • the present invention discloses a carpool optimization method, index structure, and system based on social big data.
  • the social big data-based carpool optimization method includes: obtaining, according to the social relationship of the driver, satisfying the social relationship All carpooling candidates in the company; according to the trust level index between the driver and each carpooling candidate, the driver and several carpooling candidates are combined to form a number of candidate carpooling team groups; each candidate carpooling team group is obtained To determine the commuting route and commuting cost of each candidate carpooling team in the commute information of each carpool candidate and driver; build a carpool model based on social big data and road network, and optimize according to the trust level of each candidate carpool team group Optimized with commuting cost, matching the best carpool team group that maximizes trust and minimizes commuting cost.
  • the present invention optimizes the carpool team trust level and commuting cost through the carpool model based on social big data and road network, so as to obtain the carpool scheme with the greatest trust degree and the least commuting cost, reduce complexity, reduce road congestion, and reduce travel cost and traffic Control costs and facilitate users.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Tourism & Hospitality (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Theoretical Computer Science (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Game Theory and Decision Science (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种基于社交大数据的拼车优化方法、索引结构及系统，方法包括：根据驾驶员的社会关系，获取满足所述社会关系的所有拼车候选人（S100）；根据驾驶员与每个拼车候选人之间的信任等级指数，将驾驶员与若干个拼车候选人进行拼车组合，建立若干个候选拼车团队组（S200）；获取每个候选拼车团队组中每个拼车候选人及驾驶员的通勤信息，确定候选拼车团队组的通勤路线和通勤成本（S300）；构建拼车模型，根据每个候选拼车团队组的信任等级优化与通勤成本优化，匹配得到最大化信任度且最小通勤成本的最佳拼车团队组（S400）。通过基于社交大数据与道路网络的拼车模型，同时优化拼车团队信任级别和通勤上述方法成本，以得到最大信任度和最小通勤成本的拼车方案，方便用户。

Description

一种基于社交大数据的拼车优化方法、索引结构及系统 技术领域
本发明涉及智能交通技术领域，具体涉及一种基于社交大数据的拼车优化方法、索引结构及系统。
背景技术
随着生活水平的提高，车辆出行已成为人们必不可少的交通工具。拼车已成为时下流行的便捷交通出现服务，也是解决交通拥堵的有效方案，能够降低出行费用，减轻相关的交通污染和能源消耗问题。然而，现有的拼车方式仅仅针对的是双人拼车成本优化，无法满足大量乘客的拼车需求，智能化程度低，效率低。
因此，现有技术还有待于改进和发展。
发明内容
本发明要解决的技术问题在于，针对现有技术的上述缺陷，提供一种基于社交大数据的拼车优化方法、索引结构及系统，旨在通过基于社交大数据与道路网络的拼车模型，同时优化拼车团队信任级别和通勤成本，以得到最大信任度和最小通勤成本的拼车方案，降低复杂度，降低道路拥堵，减少出行成本，方便用户。
本发明解决技术问题所采用的技术方案如下：
一种基于社交大数据的拼车优化方法，所述基于社交大数据的拼车优化方法包括以下步骤：
根据驾驶员的社会关系，获取满足所述社会关系的所有拼车候选人；
根据驾驶员与每个拼车候选人之间的信任等级指数，将驾驶员与若干个拼车候选人进行拼车组合，建立若干个候选拼车团队组；
获取每个候选拼车团队组中每个拼车候选人以及驾驶员的通勤信息，确定每个候选拼车团队组的通勤路线和通勤成本；
构建基于社交大数据与道路网络的拼车模型，根据每个候选拼车团队组的信任等级优化与通勤成本优化，匹配得到最大化信任度且最小通勤成本的最佳拼车团队组。
所述的基于社交大数据的拼车优化方法，其中，所述根据驾驶员的社会关系， 获取满足所述社会关系的所有拼车候选人之前包括：
预先建立社会关系与信任等级指数之间的关联，并设定不同等级对应不同的信任等级指数，同时进行存储。
所述的基于社交大数据的拼车优化方法，其中，所述根据驾驶员的社会关系，获取满足所述社会关系的所有拼车候选人具体包括：
基于社交关系网，获取所有驾驶员的社会关系；其中，所述社会关系包括直接朋友关系、共同朋友关系以及陌生人关系；
搜索并获取拼车候选人数据库中满足每个驾驶员所有的社会关系的拼车候选人。
所述的基于社交大数据的拼车优化方法，其中，所述根据驾驶员与每个拼车候选人之间的信任等级指数，将驾驶员与若干个拼车候选人进行拼车组合，建立若干个候选拼车团队组具体包括：
获取社会关系与信任等级指数之间的关联；
根据所述关联以及驾驶员与每个拼车候选人的社会关系，确定驾驶员与每个拼车候选人之间的信任等级指数；
根据车辆容量，将驾驶员与若干个拼车候选人进行拼车组合，建立若干个候选拼车团队组；
根据每个候选拼车团队组中驾驶员与每个拼车候选人之间的信任等级指数，构建候选拼车团队的信任模型，同时计算每个候选拼车团队组整体的信任度。
所述的基于社交大数据的拼车优化方法，其中，所述获取每个候选拼车团队组中每个拼车候选人以及驾驶员的通勤信息，确定每个候选拼车团队组的通勤路线和通勤成本具体包括：
获取每个候选拼车团队组中每个拼车候选人的第一通勤信息以及驾驶员的第二通勤信息；其中，所述第一通勤信息包括出行时间、出发地以及目的地，所述第二通勤信息包括发车时间、发车起始地以及停车终点；
根据所述第一通勤信息和第二通勤信息，确定每个候选拼车团队组的通勤路线网；
同时根据所述通勤路线网，构建候选拼车团队的通勤成本模型。
所述的基于社交大数据的拼车优化方法，其中，所述构建基于社交大数据与 道路网络的拼车模型，根据每个候选拼车团队组的信任等级优化与通勤成本优化，匹配得到最大化信任度且最小通勤成本的最佳拼车团队组具体包括：
设定一信任权重变量(W)；
根据获取的候选拼车团队的信任模型和通勤成本模型以及信任权重变量，构建基于社交大数据与道路网络的拼车模型；
将信任模型的目标函数在满足信任约束条件下进行最大化信任度处理，使得信任模型的目标函数的值最大，同时将所述通勤成本模型的目标函数在满足通勤成本约束条件下进行最小化成本处理，使得通勤成本模型的目标函数的值最小；
同时不断改变信任权重变量的取值大小，分别代入到所述拼车模型的目标函数，使得所述拼车模型的目标函数的值最大，从而得到信任值最大且通勤成本最小的最佳拼车团队组。
所述的基于社交大数据的拼车优化方法，其中，还包括：建立时空拼车匹配模型，在满足时间约束条件下，优化时间成本，用于构建所述基于社交大数据与道路网络的拼车模型。
所述的基于社交大数据的拼车优化方法，其中，影响通勤成本的因素包括通勤路线距离、行驶时间、拥堵因子、石油成本、通行费以及与通勤相关的成本因素的一种或多种。
本发明还提供一种基于社交大数据的拼车优化系统，所述基于社交大数据的拼车优化系统包括若干个终端、与所述终端连接的社交网络中心、交通网络中心以及与所述社交网络中心、交通网络中心分别连接的后台服务器，所述基于社交大数据的拼车优化系统执行时实现上述所述基于社交大数据的拼车优化方法。
本发明还提供一种基于社交大数据的拼车优化索引结构，所述拼车优化索引结构包括处理器与处理器连接的存储器，所述存储器存储有基于社交大数据的拼车优化程序，所述基于社交大数据的拼车优化程序被处理器执行时用于实现上述所述基于社交大数据的拼车优化方法。
本发明公开了一种基于社交大数据的拼车优化方法、索引结构及系统，所述基于社交大数据的拼车优化方法包括：根据驾驶员的社会关系，获取满足所述社会关系的所有拼车候选人；根据驾驶员与每个拼车候选人之间的信任等级指数，将驾驶员与若干个拼车候选人进行拼车组合，建立若干个候选拼车团队组；获取 每个候选拼车团队组中每个拼车候选人以及驾驶员的通勤信息，确定每个候选拼车团队组的通勤路线和通勤成本；构建基于社交大数据与道路网络的拼车模型，根据每个候选拼车团队组的信任等级优化与通勤成本优化，匹配得到最大化信任度且最小通勤成本的最佳拼车团队组。本发明通过基于社交大数据与道路网络的拼车模型，同时优化拼车团队信任级别和通勤成本，以得到最大信任度和最小通勤成本的拼车方案，降低复杂度，降低道路拥堵，减少出行成本和交通管制成本，方便用户。
附图说明
图1是本发明基于社交大数据的拼车优化方法的第一较佳实施例的流程图。
图2是本发明基于社交大数据的拼车优化方法中拼车活动示意图。
图3是本发明基于社交大数据的拼车优化系统的结构框图。
图4是本发明基于基于社交大数据的拼车优化索引结构的结构框图。
具体实施方式
为使本发明的目的、技术方案及优点更加清楚、明确，以下参照附图并举实施例对本发明进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
实施例一
请参见图1，图1是本发明基于社交大数据的拼车优化方法的第一较佳实施例的流程图。如图1所示，所述基于社交大数据的拼车优化方法包括以下步骤：
步骤S100，根据驾驶员的社会关系，获取满足所述社会关系的所有拼车候选人。
基于步骤S100之前，预先建立社会关系与信任等级指数之间的关联，并设定不同等级对应不同的信任等级指数，同时进行存储。所述社会关系包括直接朋友关系、共同朋友关系以及陌生人关系，如下表1所示，表1中假设一人有100个直接朋友。所述信任等级指数的划分是根据六度分离理论得到，在本发明实施例中，基于拼车活动中社交网络的规模，根据所述六度分离理论筛选出潜在的拼车候选人。
分离程度 信任等级指数 社会关系 拼车候选人数量
1度分离 1 直接朋友 100
2度分离 0.69 共同朋友 10000
3度分离 0.07 陌生人 1000000+
无 0 无任何连锁友谊 0
表1
在参阅表1结合图2可知，直接朋友关系表示图2中两两之间直接相连接，定义为1度分离；共同朋友关系表示图2中两两之间有至少一共同朋友，间接连接，定义为2度分离；陌生朋友关系表示图2中两两之间没有共同朋友，即为陌生人，定义为3度分离；无任何连锁友谊关系表示图2中边缘空白区。根据信任等级指数对应的分离度进行任意拼车安排，以组成任意的拼车团队，图2中仅仅示例了其中一种拼车安排。
具体实施时，步骤S100包括：
步骤S101，基于社交关系网，获取所有驾驶员的社会关系；其中，所述社会关系包括直接朋友关系、共同朋友关系以及陌生人关系；
步骤S102，搜索并获取拼车候选人数据库中满足每个驾驶员所有的社会关系的拼车候选人。
即基于社交关系网，获取所有驾驶员的所有的社会关系，这是初步筛选，并计算此时对应的候选人数量。根据得到的社会关系，在具有拼车出行需求的拼车候选人数据库中筛选出符合每个驾驶员所有社会关系的拼车候选人，进行二次筛选，并组成拼车候选人集，此时，建立每个拼车候选人与驾驶员的信任等级指数列表，并保存。
步骤S200，根据驾驶员与每个拼车候选人之间的信任等级指数，将驾驶员与若干个拼车候选人进行拼车组合，建立若干个候选拼车团队组。
即步骤S200具体包括：
步骤S201，获取社会关系与信任等级指数之间的关联；
步骤S202，根据所述关联以及驾驶员与每个拼车候选人的社会关系，确定驾驶员与每个拼车候选人之间的信任等级指数；
步骤S203，根据车辆容量，将驾驶员与若干个拼车候选人进行拼车组合，建立若干个候选拼车团队组；
步骤S204，根据每个候选拼车团队组中驾驶员与每个拼车候选人之间的信任等级指数，构建候选拼车团队的信任模型，同时计算每个候选拼车团队组整体的信任度。
具体实施时，所述候选拼车团队的信任模型的目标函数如公式(1)所示：
Figure PCTCN2020070369-appb-000001
信任约束条件如公式(2)所示：
Figure PCTCN2020070369-appb-000002
其中，目标函数(1)旨在最大化乘客和驾驶员之间对整个拼车队的信任度，信任约束条件(2)要求将给定数量的乘客分配给拼车团队；c表示拼车候选人的人数；CP表示拼车的车辆容量，N≤CP且N为自然数；Tc表示驾驶员和拼车候选人之间的信任等级指数；Xc表示拼车候选人被选中组合成候选人拼车团队时则为1，没选中则为0。
例如：拼车候选人为10个人(c1，c2，...，c10)，拼车被选中的为c2，c4，c5，c6，车辆容量CP为4个
Figure PCTCN2020070369-appb-000003
其中Tc1代表拼车候选人c1的拼车成员的信任等级指数。
因此，根据信任等级指数以及车辆容量，在满足信任约束条件下，筛选出使得候选拼车团队的信任模型的目标函数(1)的值最大，即最大化处理信任等级指数。也就是说，当目标函数(1)的值最大时，所对应的拼车候选人团队间的信任度是最高的。这样，使得整个拼车候选人团队更加稳定，体验较强，自然出行效率最高。
步骤S300，获取每个候选拼车团队组中每个拼车候选人以及驾驶员的通勤信息，确定每个候选拼车团队组的通勤路线和通勤成本。
即步骤S300具体包括：
步骤S301，获取每个候选拼车团队组中每个拼车候选人的第一通勤信息以 及驾驶员的第二通勤信息；
步骤S302，根据所述第一通勤信息和第二通勤信息，确定每个候选拼车团队组的通勤路线网；
步骤S303，同时根据所述通勤路线网，构建候选拼车团队的通勤成本模型。
在本发明实施例中，所述第一通勤信息包括拼车候选人(乘客)出行时间、出发地以及目的地，所述第二通勤信息包括驾驶员的发车时间、发车起始地以及停车终点，发车起始地指的是驾驶员住所位置，停车终点指的是驾驶员工作场所位置。当然，本发明实施例中的通勤成本定义为用于接送和放弃拼车候选人的过程产生的费用而不仅仅指的是单独驾驶的额外成本。住所位置与工作场所位置的路线定义为通勤线路，住所位置和工作场所位置定位为通勤线路上的节点。因此，驾驶员从当前实际住所位置出发，有若干种不同的线路去接收每个团队成员，这样就建立了不同的出现路线，从而形成通勤路线网。
需要说明的是，一条完整的通勤路线，是指驾驶员从当前实际住所位置出发，按照预先确定的顺序接收所有其他团队成员，将所有团队成员放到他们目的地节点，最后到达驾驶员工作场所节点。
因此，基于所示通勤路线网，构建候选拼车团队的通勤成本模型的目标函数如公式(3)所示：
Figure PCTCN2020070369-appb-000004
通勤约束条件如公式(4)-(13)所示：
Figure PCTCN2020070369-appb-000005
Figure PCTCN2020070369-appb-000006
Figure PCTCN2020070369-appb-000007
Figure PCTCN2020070369-appb-000008
Figure PCTCN2020070369-appb-000009
Figure PCTCN2020070369-appb-000010
Figure PCTCN2020070369-appb-000011
Figure PCTCN2020070369-appb-000012
Figure PCTCN2020070369-appb-000013
CP<VP  (13)
其中，CP表示拼车团队的规模，即拼车团队的总人数；VP表示车辆的容量；
C表示拼车候选人和驾驶员的人数，C＝1，2，...，N，其中N是候选人的总数；
h表示拼车候选人的出发地(住所位置)的索引，h＝1，2，...，N；
w是拼车候选人的目的地(工作场所位置)的索引，w＝N+1，N+2，...，N*2；
s表示驾驶员的住所，t表示驾驶员的工作场所，其中t＝s+N；i和j是节点(住宅或工作场所)的索引，
i，j＝1，2，...，N*2，若节点是当的滞留，则I，J＜＝N，若节点是一个 工作场所，则N＜I，J＜＝N*2；
d ij是节点之间边缘的阻抗i和节点j；d ij表示适合驾驶员进行通勤的任何成本函数，所述影响通勤成本的因素包括通勤路线距离、行驶时间、拥堵因子、石油成本、通行费以及与通勤相关的成本因素的一种或多种；
r和f表示指数，r，f＝1，2，...，R；R是拼车路线中的边数，R＝CP*2+1。
上述目标函数(3)旨在最小化整个拼车路线的通勤成本。
通勤约束条件(4)和(5)要求拼车路线必须从驾驶员的住所开始，并以驾驶员的工作场所结束。通勤约束条件(6)和(7)用于表示驾驶员接载给定数量(CP在住宅处)的乘客并且下降给定数量(CP在工作场所)的乘客。通勤约束条件(8)用于表示驾驶员不会随意扔摔下乘客，以形成产生逻辑通勤路径。通勤约束条件(9)用于表示对于拼车路线的任何步骤只能选择一个边缘。通勤约束条件(10)和通勤约束条件(11)确保只有一个边进入或退出所选节点。通勤约束条件(12)确保通勤路线网络的拓扑，而通勤约束条件(13)用于表示拼车团队的规模小于车辆的容量。
上述与边缘相关联的阻抗表示成本函数，即可以是影响通勤成本因素的任意组合。
也就是说，不管安排哪一条通勤路线，最终要使得目标函数(3)的通勤成本最小，即最小化通勤成本处理。
步骤S400，构建基于社交大数据与道路网络的拼车模型，根据每个候选拼车团队组的信任等级优化与通勤成本优化，匹配得到最大化信任度且最小通勤成本的最佳拼车团队组。
即步骤S400具体包括：
步骤S401，设定一信任权重变量(W)；
步骤S402，根据获取的候选拼车团队的信任模型和通勤成本模型以及信任权重变量，构建基于社交大数据与道路网络的拼车模型；
步骤S403，将信任模型的目标函数在满足信任约束条件下进行最大化信任度处理，使得信任模型的目标函数的值最大，同时将所述通勤成本模型的目标函数在满足通勤成本约束条件下进行最小化成本处理，使得通勤成本模型的目标函数的值最小；
步骤S404，同时不断改变信任权重变量的取值大小，分别代入到所述拼车模型的目标函数，使得所述拼车模型的目标函数的值最大，从而得到信任值最大且通勤成本最小的最佳拼车团队组。
具体地，基于社交大数据与道路网络的拼车模型的目标函数如公式(14)所示：
Score＝W*Trust normalized+(1-W)Cost normalized    (14)
约束条件如公式(15)-(16)所示：
Figure PCTCN2020070369-appb-000014
Figure PCTCN2020070369-appb-000015
其中，W表示信任权重变量，用于平衡拼车团队信任和通勤成本，W值范围在区间[0,1]之间，W随着值的增大而不断增强社交舒适度；Trust normalized表示整个拼车团队的规范的信任值；Cost normalized表示拼车通勤路线的归一化成本值，0≤Cost normalized≤1，Trust team与Cost carpool对应于公式(1)和(3)，Cost original表示没有乘客的通勤成本。
也就是说，通过最大化目标函数(1)和最小化目标函数(3)以及不断调整W的取值，使得目标函数(14)的值最高。当目标函数(14)的值最高时，则可得到信任值最大且通勤成本最小的最佳拼车团队组。
这样，基于拼车模型优化，获取了高信任度且通勤成本最低的拼车团队，实现了降低复杂度，降低道路拥堵，减少出行成本，方便了用户。
当然，本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过基于社交大数据的拼车优化程序来指令相关硬件(如处理器，控制器等)来完成，所述的程序可存储于一计算机可读取的存储介质中，该程序在执行时可包括如基于社交大数据的拼车优化方法实施例的流程。其中所述的存储介质可为存储器、磁碟、光盘等。
实施例二
本发明实施例还提供了一种基于社交大数据的拼车优化系统，图3示例了所述基于社交大数据的拼车优化系统的结构框图，如图3所示，所述基于社交大数据的拼车优化系统包括若干个终端100、与所述终端100连接的社交网络中心201、交通网络中心202以及与所述社交网络中心201、交通网络中心202分别连接的后台服务器301和后台服务器302，所述终端100可以是手机或电脑或平板等。所述终端100包括处理器10与所述处理器10连接的存储器20。所述存储器20存储有基于社交大数据的拼车优化程序。
所述处理器10在一些实施例中，可以是一中央处理器(Central Processing Unit,CPU)，微处理器或其他数据处理芯片，用于运行所述存储器20中存储的程序代码或处理数据，例如执行基于社交大数据的拼车优化程序等
更进一步地，所述基于社交大数据的拼车优化系统执行时用于实现上述所述基于社交大数据的拼车优化方法步骤，具体如实施例一所述。
实施例三
本发明还提供一种基于社交大数据的拼车优化索引结构，如图4所示，所述拼车优化索引结构包括第一处理器11与第一处理器11连接的第一存储器12，所述第一存储器12存储有基于社交大数据的拼车优化程序，所述基于社交大数据的拼车优化程序被第一处理器11执行上述基于社交大数据的拼车优化方法；具体如上所述。
综上所述，本发明公开了一种基于社交大数据的拼车优化方法、索引结构及系统，所述基于社交大数据的拼车优化方法包括：根据驾驶员的社会关系，获取满足所述社会关系的所有拼车候选人；根据驾驶员与每个拼车候选人之间的信任 等级指数，将驾驶员与若干个拼车候选人进行拼车组合，建立若干个候选拼车团队组；获取每个候选拼车团队组中每个拼车候选人以及驾驶员的通勤信息，确定每个候选拼车团队组的通勤路线和通勤成本；构建基于社交大数据与道路网络的拼车模型，根据每个候选拼车团队组的信任等级优化与通勤成本优化，匹配得到最大化信任度且最小通勤成本的最佳拼车团队组。本发明通过基于社交大数据与道路网络的拼车模型，同时优化拼车团队信任级别和通勤成本，以得到最大信任度和最小通勤成本的拼车方案，降低复杂度，降低道路拥堵，减少出行成本和交通管制成本，方便用户。
应当理解的是，本发明的应用不限于上述的举例，对本领域普通技术人员来说，可以根据上述说明加以改进或变换，所有这些改进和变换都应属于本发明所附权利要求的保护范围。

Claims (10)

  1. 一种基于社交大数据的拼车优化方法，其特征在于，所述基于社交大数据的拼车优化方法包括以下步骤：
    根据驾驶员的社会关系，获取满足所述社会关系的所有拼车候选人；
    根据驾驶员与每个拼车候选人之间的信任等级指数，将驾驶员与若干个拼车候选人进行拼车组合，建立若干个候选拼车团队组；
    获取每个候选拼车团队组中每个拼车候选人以及驾驶员的通勤信息，确定每个候选拼车团队组的通勤路线和通勤成本；
    构建基于社交大数据与道路网络的拼车模型，根据每个候选拼车团队组的信任等级优化与通勤成本优化，匹配得到最大化信任度且最小通勤成本的最佳拼车团队组。
  2. 根据权利要求1所述的基于社交大数据的拼车优化方法，其特征在于，所述根据驾驶员的社会关系，获取满足所述社会关系的所有拼车候选人之前包括：
    预先建立社会关系与信任等级指数之间的关联，并设定不同等级对应不同的信任等级指数，同时进行存储。
  3. 根据权利要求1所述的基于社交大数据的拼车优化方法，其特征在于，所述根据驾驶员的社会关系，获取满足所述社会关系的所有拼车候选人具体包括：
    基于社交关系网，获取所有驾驶员的社会关系；其中，所述社会关系包括直接朋友关系、共同朋友关系以及陌生人关系；
    搜索并获取拼车候选人数据库中满足每个驾驶员所有的社会关系的拼车候选人。
  4. 根据权利要求2所述的基于社交大数据的拼车优化方法，其特征在于，所述根据驾驶员与每个拼车候选人之间的信任等级指数，将驾驶员与若干个拼车候选人进行拼车组合，建立若干个候选拼车团队组具体包括：
    获取社会关系与信任等级指数之间的关联；
    根据所述关联以及驾驶员与每个拼车候选人的社会关系，确定驾驶员与每个拼车候选人之间的信任等级指数；
    根据车辆容量，将驾驶员与若干个拼车候选人进行拼车组合，建立若干个候选拼车团队组；
    根据每个候选拼车团队组中驾驶员与每个拼车候选人之间的信任等级指数， 构建候选拼车团队的信任模型，同时计算每个候选拼车团队组整体的信任度。
  5. 根据权利要求4所述的基于社交大数据的拼车优化方法，其特征在于，所述获取每个候选拼车团队组中每个拼车候选人以及驾驶员的通勤信息，确定每个候选拼车团队组的通勤路线和通勤成本具体包括：
    获取每个候选拼车团队组中每个拼车候选人的第一通勤信息以及驾驶员的第二通勤信息；其中，所述第一通勤信息包括出行时间、出发地以及目的地，所述第二通勤信息包括发车时间、发车起始地以及停车终点；
    根据所述第一通勤信息和第二通勤信息，确定每个候选拼车团队组的通勤路线网；
    同时根据所述通勤路线网，构建候选拼车团队的通勤成本模型。
  6. 根据权利要求5所述的基于社交大数据的拼车优化方法，其特征在于，所述构建基于社交大数据与道路网络的拼车模型，根据每个候选拼车团队组的信任等级优化与通勤成本优化，匹配得到最大化信任度且最小通勤成本的最佳拼车团队组具体包括：
    设定一信任权重变量(W)；
    根据获取的候选拼车团队的信任模型和通勤成本模型以及信任权重变量，构建基于社交大数据与道路网络的拼车模型；
    将信任模型的目标函数在满足信任约束条件下进行最大化信任度处理，使得信任模型的目标函数的值最大，同时将所述通勤成本模型的目标函数在满足通勤成本约束条件下进行最小化成本处理，使得通勤成本模型的目标函数的值最小；
    同时不断改变信任权重变量的取值大小，分别代入到所述拼车模型的目标函数，使得所述拼车模型的目标函数的值最大，从而得到信任值最大且通勤成本最小的最佳拼车团队组。
  7. 根据权利要求1所述的基于社交大数据的拼车优化方法，其特征在于，还包括：建立时空拼车匹配模型，在满足时间约束条件下，优化时间成本，用于构建所述基于社交大数据与道路网络的拼车模型。
  8. 根据权利要求1所述的基于社交大数据的拼车优化方法，其特征在于，影响通勤成本的因素包括通勤路线距离、行驶时间、拥堵因子、石油成本、通行费以及与通勤相关的成本因素的一种或多种。
  9. 一种基于社交大数据的拼车优化系统，其特征在于，所述基于社交大数据的拼车优化系统包括若干个终端、与所述终端连接的社交网络中心、交通网络中心以及与所述社交网络中心、交通网络中心分别连接的后台服务器，所述基于社交大数据的拼车优化系统执行时实现权利要求1-8任一项所述基于社交大数据的拼车优化方法。
  10. 一种基于社交大数据的拼车优化索引结构，其特征在于，所述拼车优化索引结构包括处理器与处理器连接的存储器，所述存储器存储有基于社交大数据的拼车优化程序，所述基于社交大数据的拼车优化程序被处理器执行时用于实现权利要求1-8任一项所述基于社交大数据的拼车优化方法。
PCT/CN2020/070369 2019-01-15 2020-01-06 一种基于社交大数据的拼车优化方法、索引结构及系统 WO2020147597A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910036755.X 2019-01-15
CN201910036755.XA CN109919354B (zh) 2019-01-15 2019-01-15 一种基于社交大数据的拼车优化方法、索引结构及系统

Publications (1)

Publication Number Publication Date
WO2020147597A1 true WO2020147597A1 (zh) 2020-07-23

Family

ID=66960427

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/070369 WO2020147597A1 (zh) 2019-01-15 2020-01-06 一种基于社交大数据的拼车优化方法、索引结构及系统

Country Status (2)

Country Link
CN (1) CN109919354B (zh)
WO (1) WO2020147597A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112561639A (zh) * 2020-12-15 2021-03-26 北京嘀嘀无限科技发展有限公司 拼车服务处理方法、装置、设备及存储介质
CN114067597A (zh) * 2021-11-17 2022-02-18 哈尔滨工业大学 一种基于强化学习的不同合乘意愿下出租车调度方法
CN114792187A (zh) * 2022-01-12 2022-07-26 山东师范大学 基于意愿和信任双重约束的群智感知团队招募方法及系统

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109919354B (zh) * 2019-01-15 2022-04-22 深圳大学 一种基于社交大数据的拼车优化方法、索引结构及系统
CN110992122B (zh) * 2019-10-22 2024-05-14 北京交通大学 人到车的网约顺风车匹配方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106251024A (zh) * 2016-08-11 2016-12-21 淮阴工学院 一种基于数据挖掘的城市上下班拼车线路优化拟合方法
CN106651027A (zh) * 2016-12-21 2017-05-10 北京航空航天大学 一种基于社交网络的互联网班车线路优化方法
US20170243172A1 (en) * 2016-02-23 2017-08-24 International Business Machines Corporation Cognitive optimal and compatible grouping of users for carpooling
CN108304946A (zh) * 2018-01-08 2018-07-20 南京理工大学 一种基于社交的智能拼车方法
CN109919354A (zh) * 2019-01-15 2019-06-21 深圳大学 一种基于社交大数据的拼车优化方法、索引结构及系统

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9305284B2 (en) * 2010-05-23 2016-04-05 Technion Research & Development Foundation Limited Methods and systems for managing a multi participant event
GR20130100414A (el) * 2013-07-12 2015-02-20 Ανδρεας-Λεωνιδας Κυπριανου Προδρομιδης Μεθοδος και συστημα για τη μεταφορα αντικειμενων μεσω δικτυων εμπιστοσυνης
WO2016034209A1 (en) * 2014-09-02 2016-03-10 Telecom Italia S.P.A. Method and system for providing a dynamic ride sharing service
US10197410B2 (en) * 2014-11-18 2019-02-05 International Business Machines Corporation Dynamic real-time carpool matching
CN105489002B (zh) * 2016-01-05 2017-12-26 深圳大学 一种基于智能匹配和路径优化的拼车方法及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170243172A1 (en) * 2016-02-23 2017-08-24 International Business Machines Corporation Cognitive optimal and compatible grouping of users for carpooling
CN106251024A (zh) * 2016-08-11 2016-12-21 淮阴工学院 一种基于数据挖掘的城市上下班拼车线路优化拟合方法
CN106651027A (zh) * 2016-12-21 2017-05-10 北京航空航天大学 一种基于社交网络的互联网班车线路优化方法
CN108304946A (zh) * 2018-01-08 2018-07-20 南京理工大学 一种基于社交的智能拼车方法
CN109919354A (zh) * 2019-01-15 2019-06-21 深圳大学 一种基于社交大数据的拼车优化方法、索引结构及系统

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112561639A (zh) * 2020-12-15 2021-03-26 北京嘀嘀无限科技发展有限公司 拼车服务处理方法、装置、设备及存储介质
CN114067597A (zh) * 2021-11-17 2022-02-18 哈尔滨工业大学 一种基于强化学习的不同合乘意愿下出租车调度方法
CN114792187A (zh) * 2022-01-12 2022-07-26 山东师范大学 基于意愿和信任双重约束的群智感知团队招募方法及系统

Also Published As

Publication number Publication date
CN109919354B (zh) 2022-04-22
CN109919354A (zh) 2019-06-21

Similar Documents

Publication Publication Date Title
WO2020147597A1 (zh) 一种基于社交大数据的拼车优化方法、索引结构及系统
Huang et al. A genetic-algorithm-based approach to solve carpool service problems in cloud computing
Ma et al. Real-time city-scale taxi ridesharing
WO2020199524A1 (zh) 一种基于网络表示学习的网约共享出行人员匹配方法
Huang et al. Optimization of the carpool service problem via a fuzzy-controlled genetic algorithm
CN110390415A (zh) 一种基于用户出行大数据进行出行方式推荐的方法及系统
CN107145971A (zh) 一种动态调整的快递配送优化方法
US9074904B1 (en) Method for solving carpool matching problem and carpool server using the same
US20230215272A1 (en) Information processing method and apparatus, computer device and storage medium
CN111581506B (zh) 基于协同过滤的航班推荐方法及系统
CN114040272B (zh) 一种路径确定方法、装置和存储介质
CN114548723A (zh) 基于云边协同的实时车货匹配与路径推荐方法及其系统
CN106651043A (zh) 一种求解多目标多车场带时间窗车辆路径问题的智能算法
Jiau et al. Services-oriented computing using the compact genetic algorithm for solving the carpool services problem
Zhu et al. PASS: parking-lot-assisted carpool over vehicular ad hoc networks
CN113709037A (zh) 一种跨链交易路由节点选择方法与装置
Zhang et al. An efficient caching and offloading resource allocation strategy in vehicular social networks
CN111708929A (zh) 信息搜索方法、装置、电子设备及存储介质
CN110992122B (zh) 人到车的网约顺风车匹配方法
CN117133144B (zh) 一种基于物联网的智慧城市停车场车位预测方法及系统
CN105208033B (zh) 一种基于智能终端情景的群体辅助推荐方法及系统
Bozdog et al. RideMatcher: peer-to-peer matching of passengers for efficient ridesharing
CN111612286B (zh) 一种订单分配方法、装置、电子设备及存储介质
Lai et al. Utility-based matching of vehicles and hybrid requests on rider demand responsive systems
CN116484113B (zh) 一种基于动态信任感知的群体观点预测方法及系统

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: 20741151

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 20/10/2021)

122 Ep: pct application non-entry in european phase

Ref document number: 20741151

Country of ref document: EP

Kind code of ref document: A1