Online car-hailing behavior analysis method and system based on heterogeneous agent model and SaaS

Abstract

The present invention discloses an online car-hailing behavior analysis method and system based on a heterogeneous agent model and SaaS, and specifically relates to the technical field of online car-hailing management, including step one, collecting information of online car-hailing drivers and online car-hailing passengers through a SaaS platform; step two, processing the information collected in step one; step three, constructing a heterogeneous agent model and calculating the matching values ​​of online car-hailing drivers and online car-hailing passengers; step four, collecting feedback information of online car-hailing passengers in a target order, and then calculating the satisfaction value of the online car-hailing passengers; step five, performing evaluation operations according to the satisfaction value and matching value of the online car-hailing passengers to obtain different types of evaluation results; the present invention can provide passengers with personalized online car-hailing matching recommendations, enhance user stickiness and satisfaction; can promptly discover and solve problems of poor service quality or insufficient matching, reduce complaints and refunds caused thereby, and reduce cost risks.

Description

Technical Field

[0001] The present invention relates to the technical field of online car-hailing management, and more specifically, to an online car-hailing behavior analysis method and system based on a heterogeneous agent model and SaaS. Background Art

[0002] Online car-hailing is the abbreviation of online taxi booking service. It refers to the business activities of building a service platform based on Internet technology, accessing qualified vehicles and drivers, and providing non-cruising taxi booking services by integrating supply and demand information. Online car-hailing is convenient: users only need to enter the destination and departure time on the mobile application, and the system will automatically match the most suitable vehicle, without waiting for a taxi or going to the station, saving time and energy; flexibility: online car-hailing is not restricted by regular buses, and can flexibly choose travel time, location and car model according to their own needs, fully meeting the personalized needs of users; and in most cases, online car-hailing will follow the user's wishes to choose a specific route for driving service, but when delivering orders on common online car-hailing platforms, they often follow the platform's preset strategy, which cannot cater to the user's preferences well, resulting in a low degree of adaptation of the final matched order, affecting the service quality of the overall order, reducing the user's experience, and easily causing the problem of reduced user stickiness, which is not conducive to the overall operation and management of the platform. Therefore, this paper proposes an online car-hailing behavior analysis method and system based on heterogeneous agent model and SaaS. Summary of the invention

[0003] To achieve the above object, the present invention provides the following technical solutions:

[0004] The online car-hailing behavior analysis system based on the heterogeneous agent model and SaaS includes an online car-hailing driver-side information collection module, an online car-hailing passenger-side information collection module, an information processing module, a heterogeneous agent model construction module, a service feedback module, a behavior analysis module, and an evaluation module;

[0005] The online car-hailing driver information collection module is used to collect the online car-hailing driver information through the SaaS platform, obtain the online car-hailing driver information set and transmit it to the information processing module;

[0006] The online car-hailing passenger terminal information collection module is used to collect the information of online car-hailing passengers through the SaaS platform, obtain the online car-hailing passenger information set and transmit it to the information processing module;

[0007] The information processing module is used to process the online car-hailing driver information set and the online car-hailing passenger information set respectively to obtain the online car-hailing driver data processing set and the online car-hailing passenger data processing set, and then transmit the online car-hailing driver data processing set and the online car-hailing passenger data processing set to the heterogeneous agent model construction module;

[0008] The heterogeneous agent model construction module is used to construct a heterogeneous agent model based on the online car-hailing driver data processing set and the online car-hailing passenger data processing set, obtain the matching values ​​of the online car-hailing driver and the online car-hailing passenger under the same order, and output the matching values ​​as output results to the evaluation module;

[0009] The service feedback module is used to collect feedback information from online ride-hailing passengers in the target order, obtain feedback information sets and transmit them to the behavior analysis module;

[0010] The behavior analysis module is used to calculate the satisfaction value of online ride-hailing passengers based on the feedback information set and transmit it to the evaluation module;

[0011] The evaluation module is used to perform evaluation operations based on the satisfaction value and matching value of online ride-hailing passengers to obtain different types of evaluation results.

[0012] In a preferred embodiment, the online car-hailing passenger data processing set consists of a mapping data set one and a non-mapping data set one, and the online car-hailing driver data processing set consists of a mapping data set two and a non-mapping data set two.

[0013] In a preferred embodiment, each item of passenger data in mapping data set one corresponds to a unique item of driver data in mapping data set two.

[0014] In a preferred embodiment, the heterogeneous agent model construction module is used to construct a heterogeneous agent model based on the online car-hailing driver data processing set and the online car-hailing passenger data processing set, and the matching value of the online car-hailing driver and the online car-hailing passenger under the same order is:

[0015] Each passenger data item in the mapping data set 1 is mapped to a vector in a preset vector space and marked as vector Ai, and each driver data item in the mapping data set 2 is mapped to a vector in a preset vector space and marked as vector Bi. When i of vector Ai and vector Bi is the same, the two are in a corresponding relationship;

[0016] The corresponding vectors Ai and Bi are screened for similarity, and the total number of vectors that reach the preset similarity threshold is marked as m, and the total number of vectors Ai is marked as n;

[0017] The total number of passenger data items in the non-mapped dataset 1 is marked as Ci, and the total number of passenger data items in the non-mapped dataset 2 is marked as Si. The matching values ​​of online car-hailing drivers and online car-hailing passengers under the same order are calculated. f1 and f2 are both preset proportional coefficients, and PPi is the matching value between online car-hailing drivers and online car-hailing passengers under the same order.

[0018] In a preferred embodiment, performing similarity screening on the corresponding vectors Ai and Bi refers to:

[0019] First, obtain a set of vector data in a corresponding relationship: vector Ai and vector Bi, and then use cosine similarity to measure the similarity between vector Ai and vector Bi. The similarity value is marked as XS i, which is obtained by the following calculation: The preset similarity threshold is marked as XX. If the similarity value XS i is greater than or equal to the preset similarity threshold XX, it is judged that the vector data of vector Ai and vector Bi have passed the screening. If the similarity value XS i is less than the preset similarity threshold XX, it is judged that the vector data of vector Ai and vector Bi have not passed the screening.

[0020] In a preferred embodiment, the behavior analysis module is used to calculate the satisfaction value of the online car-hailing passenger based on the feedback information set, which refers to:

[0021] Get all the positive feedback data items in the feedback information set and mark them as ZXi, and all the negative feedback data items and mark them as FXi, and then calculate the satisfaction value of the online car-hailing passengers as follows: wi is the preset proportional coefficient corresponding to the positive feedback data item ZXi, ri is the preset proportional coefficient corresponding to the negative feedback data item FXi, and MYi is the satisfaction value of the online car-hailing passengers.

[0022] In a preferred embodiment, the evaluation module is used to perform evaluation operations based on the satisfaction value and matching value of the online car-hailing passengers, and different types of evaluation results are obtained:

[0023] Calculate the satisfaction value MYi and matching value PPi of the online car-hailing passengers: PG1i=k1·MYi+k2·PPi,

[0024] PG1i is the first evaluation value, k1 and k2 are both preset proportional coefficients, and the sum of the two is one. If the first evaluation value PG1i is less than the preset first evaluation threshold, evaluation type one is generated; if the first evaluation value PG1i is greater than or equal to the preset first evaluation threshold, the following calculation is continued: PG2i=k3·(MYi-Q1)+k4·(PPi-Q2), PG2i is the second evaluation value, k3 and k4 are both preset proportional coefficients, and the sum of the two is one, Q1 is the preset standard value corresponding to the satisfaction value MYi of the online car-hailing passenger, Q2 is the preset standard value corresponding to the matching value PPi, if the second evaluation value PG2i is less than the preset second evaluation threshold, evaluation type two is generated; if the second evaluation value PG2i is greater than or equal to the preset second evaluation threshold, evaluation type three is generated.

[0025] In a preferred embodiment, the method for analyzing online car-hailing behavior based on the heterogeneous agent model and SaaS includes the following steps:

[0026] Step 1: Collect the information of online car-hailing drivers through the SaaS platform to obtain the online car-hailing driver information set, and collect the information of online car-hailing passengers through the SaaS platform to obtain the online car-hailing passenger information set;

[0027] Step 2: Process the online car-hailing driver information set and the online car-hailing passenger information set respectively to obtain the online car-hailing driver data processing set and the online car-hailing passenger data processing set;

[0028] Step 3: Construct a heterogeneous agent model based on the online car-hailing driver data processing set and the online car-hailing passenger data processing set, and obtain the matching values ​​of the online car-hailing driver and the online car-hailing passenger under the same order;

[0029] Step 4: Collect feedback information from online ride-hailing passengers in the target order, obtain a feedback information set and calculate the satisfaction value of the online ride-hailing passengers;

[0030] Step 5: Perform evaluation operations based on the satisfaction value and matching value of the online car-hailing passengers to obtain different types of evaluation results.

[0031] Technical effects and advantages of the present invention:

[0032] The present invention can deeply understand the interactive behavior between online car-hailing drivers and passengers through heterogeneous agent models and behavioral analysis, so as to discover and solve problems in the service and improve service quality and user experience; by utilizing the evaluation module to evaluate passenger satisfaction value and matching value, the matching degree between online car-hailing drivers and passengers can be optimized, and the success rate and satisfaction of orders can be improved; by analyzing behavioral data and evaluation results, the resources of online car-hailing drivers and passengers can be more effectively allocated, and vehicle utilization and service efficiency can be improved.

[0033] Based on the passenger's preference and behavior analysis, the present invention can provide passengers with personalized online car-hailing matching recommendations, thereby enhancing user stickiness and satisfaction; through the results of the evaluation module, problems such as poor service quality or insufficient matching can be discovered and resolved in a timely manner, thereby reducing complaints and chargebacks caused thereby, and reducing cost risks; the quality of online car-hailing services and user experience can be optimized, thereby attracting more users and drivers to join the platform and enhancing market competitiveness; the present invention helps to improve the operation and management of the online car-hailing industry, improve service quality and user satisfaction, and thereby achieve sustainable development and commercial success. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0035] Figure 1 This is a schematic diagram of the online car-hailing behavior analysis system based on the heterogeneous agent model and SaaS in the present invention.

[0036] Figure 2 This is a schematic diagram of the online car-hailing behavior analysis method based on the heterogeneous agent model and SaaS in the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] Reference Figure 1 - Figure 2 The following examples are obtained:

[0039] Example 1

[0040] The online car-hailing behavior analysis system based on the heterogeneous agent model and SaaS includes an online car-hailing driver-side information collection module, an online car-hailing passenger-side information collection module, an information processing module, a heterogeneous agent model construction module, a service feedback module, a behavior analysis module, and an evaluation module; the online car-hailing driver-side information collection module, the online car-hailing passenger-side information collection module, the information processing module, the heterogeneous agent model construction module, the service feedback module, the behavior analysis module, and the evaluation module are communicated with each other.

[0041] The online car-hailing driver information collection module is used to collect the online car-hailing driver's information through the SaaS platform, obtain the online car-hailing driver information set and transmit it to the information processing module; collect the online car-hailing driver's information through the SaaS platform, which includes the driver's personal data, vehicle information, order records, order preferences and other information. The collected information can be used to build a heterogeneous agent model for the driver, as well as subsequent behavioral analysis and evaluation operations.

[0042] The online car-hailing passenger information collection module is used to collect the information of online car-hailing passengers through the SaaS platform, obtain the online car-hailing passenger information set and transmit it to the information processing module; the online car-hailing passenger information is collected through the SaaS platform, including the passenger's personal profile, riding preferences, historical order records, etc. This information is crucial for building a heterogeneous agent model of passengers and behavioral analysis.

[0043] The information processing module is used to process the online car-hailing driver information set and the online car-hailing passenger information set respectively to obtain the online car-hailing driver data processing set and the online car-hailing passenger data processing set, and then transmit the online car-hailing driver data processing set and the online car-hailing passenger data processing set together to the heterogeneous agent model construction module; this module is responsible for processing the collected driver and passenger information, including data cleaning, feature extraction, data standardization and other operations familiar to people in this field, so as to facilitate subsequent model construction and analysis.

[0044] The heterogeneous agent model construction module is used to construct a heterogeneous agent model based on the online car-hailing driver data processing set and the online car-hailing passenger data processing set, obtain the matching values ​​of the online car-hailing driver and the online car-hailing passenger under the same order, and output the matching values ​​as output results to the evaluation module; this module constructs a heterogeneous agent model based on the processed driver and passenger information. This model describes the interactive behavior between drivers and passengers, such as the interactive correlation between the driver's order acceptance strategy and the passenger's travel preferences. This model can help understand and predict the degree of adaptation of the interactive behavior between different types of agents, so that more suitable orders can be arranged to ensure the user experience of the taxi platform.

[0045] The service feedback module is used to collect feedback information from online car-hailing passengers in the target order, obtain a feedback information set and transmit it to the behavior analysis module; it is used to collect passengers' feedback information on the service, such as ratings, comments, etc. The present invention adopts the form of a questionnaire survey, and a preset number of feedback items with high platform attention are used as the content of the questionnaire survey, and then the item scores are collected in the form of score assignment, with a full score of 100. These feedback information are very important for evaluating service quality and improving services.

[0046] The behavior analysis module is used to calculate the satisfaction value of the online car-hailing passengers based on the feedback information set and transmit it to the evaluation module; perform statistical analysis on the scoring data, and then transmit the final satisfaction value of the online car-hailing passengers to the evaluation module.

[0047] The evaluation module is used to perform evaluation operations based on the satisfaction value and matching value of online ride-hailing passengers to obtain different types of evaluation results.

[0048] Among them, the online car-hailing passenger data processing set consists of mapping data set one and non-mapping data set one, and the online car-hailing driver data processing set consists of mapping data set two and non-mapping data set two. Each passenger data in mapping data set one corresponds to a unique driver data in mapping data set two; mapping data set one contains the data of online car-hailing passengers, and mapping data set two contains the data of online car-hailing drivers. Each passenger data in mapping data set one has a corresponding unique driver data in mapping data set two. This one-to-one correspondence can be used to establish a connection between passengers and drivers. For example, passengers in mapping data set one prefer short distance and fast time in navigation, so passengers with short distance fall into non-mapping data set one, while drivers in mapping data set two prefer to drive orders. If the completion speed is fast, then fast time and fast order completion speed constitute a mapping relationship, and the tendency data of the two can be represented as vectors later. Non-mapped dataset 1 and non-mapped dataset 2 contain passenger and driver data that cannot be directly mapped to each other's dataset. These data may not be able to directly establish a corresponding relationship between the two datasets for various reasons. If it contains data whose project content is not public to protect user privacy, this data will fall into non-mapped dataset 1 or non-mapped dataset 2, and the mapped dataset 1 obtained each time is different. If the user chooses to hide his priority for fast time in navigation, this content will also fall into non-mapped dataset 1. Similarly, the item that the driver tends to complete the order quickly will not appear in the mapped dataset 2.

[0049] The heterogeneous agent model construction module is used to construct a heterogeneous agent model based on the online car-hailing driver data processing set and the online car-hailing passenger data processing set. The matching values ​​of online car-hailing drivers and online car-hailing passengers under the same order are:

[0050] Each passenger data item in the mapping data set 1 is mapped to a vector in the preset vector space and marked as vector Ai, and each driver data item in the mapping data set 2 is mapped to a vector in the preset vector space and marked as vector Bi. When i of vector Ai and vector Bi is the same, the two are in a corresponding relationship. For each pair of corresponding passenger and driver data, it can be represented in the vector space, which is convenient for subsequent similarity calculation and matching;

[0051] Perform similarity screening on the corresponding vectors Ai and vectors Bi, the total number of vectors that reach the preset similarity threshold is marked as m, and the total number of vectors Ai is marked as n;

[0052] The total number of passenger data items in the non-mapped dataset 1 is marked as Ci, and the total number of passenger data items in the non-mapped dataset 2 is marked as Si. The matching values ​​of online car-hailing drivers and online car-hailing passengers under the same order are calculated. f1 and f2 are both preset proportional coefficients. f1 represents the proportion of passenger data in mapping data set 1 in the matching value calculation. f2 represents the proportion of driver data in mapping data set 2 in the matching value calculation. PPi is the matching value of online car-hailing drivers and passengers under the same order. The matching value PPi represents the degree of matching between online car-hailing drivers and passengers under the same order. The higher the value, the higher the degree of matching.

[0053] The similarity screening of the corresponding vectors Ai and Bi refers to:

[0054] First, obtain a set of corresponding vector data: vector Ai and vector Bi, and then use cosine similarity to measure the similarity between vector Ai and vector Bi, and use cosine similarity to measure the similarity between vector Ai and vector Bi. Cosine similarity is a commonly used similarity measurement method used to measure the angle between two vectors. The closer the value is to 1, the higher the similarity. The similarity value is marked as XS i, and it is obtained by the following calculation: The preset similarity threshold is marked as XX. If the similarity value XS i is greater than or equal to the preset similarity threshold mark XX, it is judged that the vector data of vector Ai and vector Bi have passed the screening, indicating that the degree of match between the passenger and the driver is high. If the similarity value XS i is less than the preset similarity threshold XX, it is judged that the vector data of vector Ai and vector Bi have not passed the screening, indicating that the degree of match between the passenger and the driver has not met the expected standard.

[0055] The behavior analysis module is used to calculate the satisfaction value of online ride-hailing passengers based on the feedback information set, which refers to:

[0056] Get all the positive feedback data items in the feedback information set and mark them as ZXi, and all the negative feedback data items and mark them as FXi, and then calculate the satisfaction value of the online car-hailing passengers as follows: wi is the preset proportional coefficient corresponding to the positive feedback data item ZXi. A weighted calculation is performed according to the preset proportional coefficient wi, which indicates the contribution of these positive feedback data to the satisfaction value. ri is the preset proportional coefficient corresponding to the negative feedback data item FXi. A weighted calculation is performed according to the preset proportional coefficient ri, which indicates the negative impact of these negative feedback data on the satisfaction value. MYi is the satisfaction value of the online car-hailing passengers, which can quantify the impact of feedback information on passenger satisfaction, help evaluate the quality of online car-hailing services, and provide a reference for improving services.

[0057] The evaluation module is used to perform evaluation operations based on the satisfaction value and matching value of online ride-hailing passengers. Different types of evaluation results are:

[0058] Calculate the satisfaction value MYi and matching value PPi of the online car-hailing passengers: PG1i=k1·MYi+k2·PPi,

[0059] PG1 i is the first evaluation value, k1 and k2 are both preset proportional coefficients, and the sum of the two is one. If the first evaluation value PG1 i is less than the preset first evaluation threshold, the evaluation type 1 is generated; if the first evaluation value PG1 i is less than the preset first evaluation threshold, the evaluation type 1 is generated; If i is greater than or equal to the preset first evaluation threshold, the following calculation is continued: PG2i=k3·(MYi-Q1)+k4·(PPi-Q2), PG2i is the second evaluation value, k3 and k4 are both preset proportional coefficients, and the sum of the two is one, Q1 is the preset standard value corresponding to the satisfaction value MYi of the online car-hailing passenger, Q2 is the preset standard value corresponding to the matching value PPi, if the second evaluation value PG2i is less than the preset second evaluation threshold, then evaluation type two is generated; if the second evaluation value PG2i is greater than or equal to the preset second evaluation threshold, then evaluation type three is generated. Through this evaluation process, the online car-hailing service can be evaluated at different levels according to the comprehensive situation of the satisfaction value and the matching value, and corresponding improvements or adjustments can be made accordingly. The generated evaluation types reflect the different levels of evaluation of the quality and matching degree of the online car-hailing service. Evaluation type one indicates that the service quality is low and needs further improvement and optimization; evaluation type two indicates that the service quality is average, but it can be compensated by improving the matching degree, and improvements need to be made to the matching degree. Type three means that both the service quality and matching degree are high, which is a relatively ideal situation. The generated evaluation results can be used to guide the operation strategy and service optimization of online car-hailing service providers. For example, for the evaluation type one, the service quality can be improved by improving the service process and improving the quality of the target drivers in the order. For the evaluation type two, the matching order results can be optimized, and it is recommended to improve the quality of the target drivers in the order to improve the personalization and matching degree of the service. For the evaluation type three, encouragement and praise can be given, such as the platform's incentive behaviors such as ranking model orders. Through the expansion and in-depth understanding of the evaluation module, the data and information provided by the online car-hailing behavior analysis system can be better utilized to provide effective support and guidance for the improvement and optimization of online car-hailing services. For example, when matching vehicles for target users, priority is given to matching drivers with a higher matching degree than the previous time. For example, when matching a driver with a matching degree lower than the previous order or the historical average order, a reminder is sent to the driver to improve the driver's service awareness in order to ensure the completion quality of the matched order.

[0060] Example 2

[0061] The online car-hailing behavior analysis method based on the heterogeneous agent model and SaaS includes the following steps:

[0062] Step 1: Collect the information of online car-hailing drivers through the SaaS platform to obtain the online car-hailing driver information set, and collect the information of online car-hailing passengers through the SaaS platform to obtain the online car-hailing passenger information set;

[0063] Step 2: Process the online car-hailing driver information set and the online car-hailing passenger information set respectively to obtain the online car-hailing driver data processing set and the online car-hailing passenger data processing set;

[0064] Step 3: Construct a heterogeneous agent model based on the online car-hailing driver data processing set and the online car-hailing passenger data processing set, and obtain the matching values ​​of the online car-hailing driver and the online car-hailing passenger under the same order;

[0065] Step 4: Collect feedback information from online ride-hailing passengers in the target order, obtain a feedback information set and calculate the satisfaction value of the online ride-hailing passengers;

[0066] Step 5: Perform evaluation operations based on the satisfaction value and matching value of the online car-hailing passengers to obtain different types of evaluation results.

[0067] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0068] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0069] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0070] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0071] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. The online car-hailing behavior analysis system based on heterogeneous agent model and SaaS is characterized by: It includes the online car-hailing driver-side information collection module, the online car-hailing passenger-side information collection module, the information processing module, the heterogeneous agent model construction module, the service feedback module, the behavior analysis module, and the evaluation module; The online car-hailing driver information collection module is used to collect the online car-hailing driver information through the SaaS platform, obtain the online car-hailing driver information set and transmit it to the information processing module; The online car-hailing passenger terminal information collection module is used to collect the information of online car-hailing passengers through the SaaS platform, obtain the online car-hailing passenger information set and transmit it to the information processing module; The information processing module is used to process the online car-hailing driver information set and the online car-hailing passenger information set respectively to obtain the online car-hailing driver data processing set and the online car-hailing passenger data processing set, and then transmit the online car-hailing driver data processing set and the online car-hailing passenger data processing set to the heterogeneous agent model construction module; The heterogeneous agent model construction module is used to construct a heterogeneous agent model based on the online car-hailing driver data processing set and the online car-hailing passenger data processing set, obtain the matching values ​​of the online car-hailing driver and the online car-hailing passenger under the same order, and output the matching values ​​as output results to the evaluation module; The service feedback module is used to collect feedback information from online ride-hailing passengers in the target order, obtain feedback information sets and transmit them to the behavior analysis module; The behavior analysis module is used to calculate the satisfaction value of online ride-hailing passengers based on the feedback information set and transmit it to the evaluation module; The evaluation module is used to perform evaluation operations based on the satisfaction value and matching value of online ride-hailing passengers to obtain different types of evaluation results; The online car-hailing passenger data processing set consists of mapping data set 1 and non-mapping data set 1, and the online car-hailing driver data processing set consists of mapping data set 2 and non-mapping data set 2; Each passenger data item in the mapping data set 1 corresponds to a unique driver data item in the mapping data set 2; The heterogeneous agent model construction module is used to construct a heterogeneous agent model based on the online car-hailing driver data processing set and the online car-hailing passenger data processing set. The matching values ​​of online car-hailing drivers and online car-hailing passengers under the same order are: Each passenger data item in the mapping data set 1 is mapped to a vector in a preset vector space and marked as vector Ai, and each driver data item in the mapping data set 2 is mapped to a vector in a preset vector space and marked as vector Bi. When i of vector Ai and vector Bi is the same, the two are in a corresponding relationship; Perform similarity screening on the corresponding vectors Ai and Bi, the total number of vectors that reach the preset similarity threshold is marked as m, and the total number of vectors Ai is marked as n; The total number of passenger data items in the non-mapped dataset 1 is marked as Ci, and the total number of passenger data items in the non-mapped dataset 2 is marked as Si. The matching values ​​of online car-hailing drivers and online car-hailing passengers under the same order are calculated. f1 and f2 are both preset proportional coefficients, and PPi is the matching value between online car-hailing drivers and online car-hailing passengers under the same order.

2. The online car-hailing behavior analysis system based on heterogeneous agent model and SaaS according to claim 1 is characterized in that: The similarity screening of the corresponding vectors Ai and Bi refers to: First, obtain a set of vector data in a corresponding relationship: vector Ai and vector Bi, and then use cosine similarity to measure the similarity between vector Ai and vector Bi. The similarity value is marked as XSi, which is obtained by the following calculation: The preset similarity threshold is marked as XX. If the similarity value XSi is greater than or equal to the preset similarity threshold XX, it is judged that the vector data of vector Ai and vector Bi have passed the screening. If the similarity value XSi is less than the preset similarity threshold XX, it is judged that the vector data of vector Ai and vector Bi have not passed the screening.

3. The online car-hailing behavior analysis system based on heterogeneous agent model and SaaS according to claim 2 is characterized in that: The behavior analysis module is used to calculate the satisfaction value of online ride-hailing passengers based on the feedback information set, which refers to: Get all the positive feedback data items in the feedback information set and mark them as ZXi, and all the negative feedback data items and mark them as FXi, and then calculate the satisfaction value of the online car-hailing passengers as follows: wi is the preset proportional coefficient corresponding to the positive feedback data item ZXi, ri is the preset proportional coefficient corresponding to the negative feedback data item FXi, and MYi is the satisfaction value of the online car-hailing passengers.

4. The online car-hailing behavior analysis system based on heterogeneous agent model and SaaS according to claim 3 is characterized in that: The evaluation module is used to perform evaluation operations based on the satisfaction value and matching value of online ride-hailing passengers. Different types of evaluation results are: Calculate the satisfaction value MYi and matching value PPi of the online car-hailing passengers: PG1i=k1·MYi+k2·PPi, PG1i is the first evaluation value, k1 and k2 are both preset proportional coefficients, and the sum of the two is one. If the first evaluation value PG1i is less than the preset first evaluation threshold, evaluation type one is generated; if the first evaluation value PG1i is greater than or equal to the preset first evaluation threshold, the following calculation is continued: PG2i=k3·(MYi-Q1)+k4·(PPi-Q2), PG2i is the second evaluation value, k3 and k4 are both preset proportional coefficients, and the sum of the two is one, Q1 is the preset standard value corresponding to the satisfaction value MYi of the online car-hailing passenger, Q2 is the preset standard value corresponding to the matching value PPi, if the second evaluation value PG2i is less than the preset second evaluation threshold, evaluation type two is generated; if the second evaluation value PG2i is greater than or equal to the preset second evaluation threshold, evaluation type three is generated.

5. A method for analyzing online car-hailing behavior based on a heterogeneous agent model and SaaS according to any one of claims 1 to 4, characterized in that: The following steps are involved: Step 1: Collect the information of online car-hailing drivers through the SaaS platform to obtain the online car-hailing driver information set, and collect the information of online car-hailing passengers through the SaaS platform to obtain the online car-hailing passenger information set; Step 2: Process the online car-hailing driver information set and the online car-hailing passenger information set respectively to obtain the online car-hailing driver data processing set and the online car-hailing passenger data processing set; Step 3: Construct a heterogeneous agent model based on the online car-hailing driver data processing set and the online car-hailing passenger data processing set, and obtain the matching values ​​of the online car-hailing driver and the online car-hailing passenger under the same order; Step 4: Collect feedback information from online ride-hailing passengers in the target order, obtain a feedback information set and calculate the satisfaction value of the online ride-hailing passengers; Step 5: Perform evaluation operations based on the satisfaction value and matching value of the online car-hailing passengers to obtain different types of evaluation results.

Images