TOPN tenant online car-hailing order processing method based on tenant portrait

Abstract

The invention relates to a TOPN tenant online car-hailing order processing method based on tenant portraits, and relates to the technical field of data mining, basic information and historical order information of tenants are collected, pre-processed and analyzed, travel characteristics of the tenants are extracted, the travel characteristics of the tenants comprise driver characteristics, vehicle characteristics and historical order characteristics, and the driver characteristics, the vehicle characteristics and the historical order characteristics are extracted; establishing a tenant portrait model according to the extracted tenant travel characteristics, obtaining current online car-hailing order information, matching the current online car-hailing order information with the tenant portrait, determining the matching degree of each order, sorting the orders according to the matching degrees, obtaining a TOPN tenant online car-hailing order, allocating a driver to the TOPN tenant online car-hailing order, and notifying the driver to go to receive a passenger. By constructing detailed tenant portraits and accurately matching orders, the empty driving and invalid waiting time is reduced, and TOPN represents the number of the first N orders of each tenant.

Description

Technical Field

[0001] The present invention relates to the technical field of data mining, and more specifically, to a method for processing TOPN tenant online car-hailing orders based on tenant portraits. Background Art

[0002] In today's digital age, the online car-hailing industry is experiencing rapid development and widespread application. With the popularization of the mobile Internet and the increasing demand of consumers for convenient travel, online car-hailing platforms have become an important part of modern urban transportation. However, with the increase in the number of users and the soaring number of orders, traditional order processing methods have gradually revealed problems such as low efficiency and resource waste, which directly affect the improvement of service quality and user experience.

[0003] To address these challenges, online car-hailing platforms have begun to actively explore intelligent order processing methods based on tenant portraits. An online car-hailing service provider can be regarded as a tenant in the system, and there are multiple tenants in the online car-hailing aggregation platform that can provide services. A tenant portrait refers to a user model established by analyzing and mining information such as user behavior, preferences, and historical order data. The TOPN tenant order processing method based on tenant portraits can effectively optimize resource allocation and improve service accuracy, thereby achieving better service quality and user experience. Summary of the Invention

[0004] The present invention aims at the technical problems existing in the prior art, and provides a method for processing TOPN tenant online car-hailing orders based on tenant portraits to solve the problems raised in the above background art.

[0005] The technical solution of the present invention to solve the above technical problems is as follows: A method for processing TOPN tenant online car-hailing orders based on tenant portraits specifically includes the following steps:

[0006] Step 101, collect the basic information and historical order information of the tenant, and perform preprocessing and analysis to extract the travel characteristics of the tenant;

[0007] Step 102, establish a tenant portrait model according to the extracted travel characteristics of the tenant, obtain the current online car-hailing order information, match it with the tenant portrait, and determine the matching degree of each order;

[0008] Step 103, sort the orders according to the matching degree to obtain the TOPN tenant online car-hailing orders, assign drivers to the TOPN tenant online car-hailing orders, and notify the drivers to pick up passengers.

[0009] In a preferred embodiment, in the step 101, collecting the basic information and historical order information of the tenant, and performing preprocessing and analysis to extract the travel characteristics of the tenant, the specific steps are as follows:

[0010] Step A1, Data collection: Obtain the basic information and historical order information of the tenant from the database of the online car-hailing platform. The basic information of the tenant includes driver information and vehicle information. Among them, the driver information includes gender, age, and whether there is a child seat. The vehicle information includes whether the vehicle is idle, the luxury type, and the cleanliness of the vehicle. The historical order information includes order time, starting location, ending location, and riding frequency;

[0011] Step A2, Data preprocessing: Clean the collected data, including removing duplicate data, handling missing values and outliers. According to the collected information, extract the travel characteristics of the tenant. The travel characteristics include driver characteristics, vehicle characteristics, and historical order characteristics. For each characteristic, perform standardization processing and feature encoding. Further, it includes the following steps:

[0012] Step A201, Binary encoding for the extracted gender, whether there is a child seat, and whether it is idle. Use {G i , C i , F i} ∈ (0, 1) to represent gender, child seat, and idle. Among them, G0 represents male, G1 represents female, C0 represents no child seat, C1 represents having a child seat, F0 represents not idle, and F1 represents idle;

[0013] Step A202, Bin the extracted age and convert it into a categorical variable. Divide the age into m intervals and use A j ∈ (0, 1,..., m - 1) to represent age; use L j ∈ (0, 1,..., k - 1) to represent the luxury type, where k is the number of categories of the luxury type; use D j ∈ (0, 1, 2) to represent the cleanliness level, and divide it into low, medium, and high levels; use T j ∈ (0, 1,..., 23) to represent the order time, and the extracted time is in 24-hour format.

[0014] In a preferred embodiment, in step 102, according to the extracted travel characteristics of the tenant, establish a tenant portrait, obtain the current online car-hailing order information, and match it with the tenant portrait to determine the matching degree of each order. The specific steps are as follows:

[0015] Step B1, Construct a feature matrix: For each feature, construct its feature matrix among the tenants. The driver feature matrix is represented as: X d = [x d1 x d2 ...x dn , where x dnRepresent the gender, age, and child seat information of the nth driver; the vehicle feature matrix is represented as: X v =[x v1 x v2 ...x vp , where x vp represents the idle status, luxury type, and cleanliness of the pth vehicle;

[0016] Step B2. Establish a tenant profile: Combine the constructed driver feature matrix and vehicle feature matrix into a comprehensive feature matrix X tenant =[X d X v . Standardize the combined feature matrix, use the standardized feature matrix to train a machine learning model, and based on the trained model, predict each tenant to generate a tenant profile;

[0017] Step B3. Calculate the matching degree: Obtain the current online car-hailing order information, match it with the tenant profile, determine the matching degree of each order, and represent the order feature vector as Use cosine similarity to measure the matching degree between two vectors as follows:

[0018]

[0019] where Matching(η) is the matching degree, Q1 and Q2 are the driver features and vehicle features respectively, O η represents the order frequency, V(η) represents the vehicle selection tendency, and η represents the tenant.

[0020] In a preferred embodiment, in the step B2 of establishing a tenant profile, using the standardized feature matrix to train a machine learning model and predicting each tenant based on the trained model to generate a tenant profile further includes the following steps:

[0021] B201. Feature matrix standardization: Standardize the combined feature matrix X tenant to ensure that the mean of each feature is 0 and the standard deviation is 1, improving the stability and effect of the model. The specific calculation formula is as follows:

[0022]

[0023] where X tenant is the combined one, μ X and σ X are the mean and standard deviation of the feature matrix respectively, and X' is the standardized feature matrix;

[0024] Step B202, Model Training: Use the standardized feature matrix X' and the target variable to train a decision tree model. The target variable is the classification label of the tenant portrait. Select the optimal feature division point by maximizing the information gain, split the data into subsets according to the selected features, and repeat the splitting process for each subset until the minimum sample number is satisfied;

[0025] Step B203, Portrait Generation: According to the trained model, predict each tenant to generate a tenant portrait, predict the order frequency and vehicle selection tendency of the tenant, and represent the feature set as {α1, α2,..., α τ}. Each leaf node L of the decision tree corresponds to a category C. The decision tree model is as follows:

[0026]

[0027] Among them, f(x) is the prediction result of the model for the sample x, C r is the category of the leaf node, and R r is a set of conditions for dividing features;

[0028] The order frequency prediction is: O η = f O (x η ), where O η represents the order frequency, η represents the tenant, x η is the feature vector of the tenant, and f O is the function of the trained decision tree model for predicting the order frequency;

[0029] The vehicle selection tendency prediction is: V(η) = f V (x η ), where V(η) represents the vehicle selection tendency, and f V is the function of the trained decision tree model for predicting the vehicle selection tendency;

[0030] In summary, the tenant portrait is represented by a vector as: Among them, O η represents the order frequency, and V(η) represents the vehicle selection tendency.

[0031] In a preferred embodiment, in step 103, sort the orders according to the matching degree to obtain the TOPN tenant online car-hailing orders, assign drivers to the TOPN tenant online car-hailing orders, and notify the drivers to pick up passengers. The specific steps are as follows:

[0032] Step C1. Screening TOPN: Traverse each order, calculate its matching degree with each tenant, and create a list to store each order and its calculated matching degree. For each tenant, rank each order based on the calculated matching degree, sort the orders in descending order of the matching degree, and extract the top N orders from the sorted list to form the TOPN tenant online car-hailing orders.

[0033] Step C2. Assigning drivers: Assign drivers to each order according to the drivers' preferences and the TOPN orders, and send notifications to each driver assigned to an order, informing them of the pick-up location and time.

[0034] The beneficial effects of the present invention are as follows: Collect the basic information and historical order information of tenants, perform preprocessing and analysis, and extract the travel characteristics of tenants. The travel characteristics of the tenants include driver characteristics, vehicle characteristics, and historical order characteristics. According to the extracted travel characteristics of tenants, establish a tenant portrait model, obtain the current online car-hailing order information, match it with the tenant portrait, determine the matching degree of each order, sort the orders according to the matching degree to obtain the TOPN tenant online car-hailing orders, assign drivers to the TOPN tenant online car-hailing orders, and notify the drivers to pick up passengers. Among them, the "TOPN" for each tenant represents the number of the top N orders of each tenant. According to the historical behavior and preferences of tenants, construct a detailed tenant portrait, accurately match orders, improve the fit between orders and tenants, and by assigning the most matching orders, the order receiving rate and income of drivers can be increased, and the operation cost can be further optimized. Description of the Drawings

[0035] Figure 1 It is a flowchart of the method of the present invention. Detailed Embodiments

[0036] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present application.

[0037] In the description of the present application, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present application, "a plurality" means two or more, unless otherwise specifically defined.

[0038] In the description of the present application, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in this application is not necessarily construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to implement and use the present invention. In the following description, details are set forth for purposes of explanation. It should be understood that those of ordinary skill in the art can recognize that the present invention can be implemented without these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but rather to be consistent with the broadest scope that conforms to the principles and features disclosed in this application.

[0039] Embodiment 1

[0040] This embodiment provides a method for processing TOPN tenant online car-hailing orders based on tenant portraits as shown in Figure 1 the following, which specifically includes the following steps:

[0041] Step 101: Collect the basic information and historical order information of the tenant, and perform preprocessing and analysis to extract the travel characteristics of the tenant;

[0042] Step 102: According to the extracted travel characteristics of the tenant, establish a tenant portrait model, obtain the current online car-hailing order information, match it with the tenant portrait, and determine the matching degree of each order;

[0043] Step 103: Sort the orders according to the matching degree to obtain the TOPN tenant online car-hailing orders, assign drivers to the TOPN tenant online car-hailing orders, and notify the drivers to pick up passengers.

[0044] Preferably, in step 101, the basic information and historical order information of the tenant are collected, and preprocessing and analysis are performed to extract the travel characteristics of the tenant. The specific steps are as follows:

[0045] Step A1: Data collection: Obtain the basic information and historical order information of the tenant from the database of the online car-hailing platform. The basic information of the tenant includes driver information and vehicle information. Among them, the driver information includes gender, age, and whether there is a child seat, and the vehicle information includes whether the vehicle is idle, the luxury type of the vehicle, and the cleanliness. The historical order information includes order time, starting location, ending location, and riding frequency;

[0046] Step A2, Data Preprocessing: Clean the collected data, including removing duplicate data, handling missing values and outliers, ensuring the integrity and accuracy of the data, and avoiding analysis distortion caused by data quality problems. According to the collected information, extract the travel characteristics of tenants, including driver characteristics and vehicle characteristics. For each characteristic, perform standardization processing to ensure that the numerical ranges of different characteristics are consistent, and perform feature encoding to integrate the basic information and historical order information of tenants into a unified data set, which further includes the following steps:

[0047] Step A201, Binary Encoding for Extracted Gender, Presence of Child Seat, and Availability: Use {G i , C i , F i} ∈ (0, 1) to represent gender, child seat, and availability, where G0 represents male, G1 represents female, C0 represents no child seat, C1 represents having a child seat, F0 represents not available, and F1 represents available;

[0048] Step A202, Binning the Extracted Age and Converting to Categorical Variable: Divide the age into m intervals and convert it into a categorical variable. Use A j ∈ (0, 1,..., m - 1) to represent age; Use L j ∈ (0, 1,..., k - 1) to represent the luxury type, where k is the number of categories of the luxury type; Use D j ∈ (0, 1, 2) to represent the cleanliness level, which is divided into low, medium, and high grades; Use T j ∈ (0, 1,..., 23) to represent the order time, and the extracted time is in 24-hour format.

[0049] Preferably, in step 102, according to the extracted travel characteristics of tenants, establish a tenant profile, obtain the current online car-hailing order information, and match it with the tenant profile to determine the matching degree of each order. The specific steps are as follows:

[0050] Step B1, Construct Feature Matrix: For each characteristic, construct its feature matrix among tenants. The driver feature matrix is represented as: X d = [x d1 x d2 ...x dn , where x dn represents the gender, age, and child seat information of the nth driver; The vehicle feature matrix is represented as: X v = [x v1 x v2 ...x vp , where x vp represents the availability status, luxury type, and cleanliness level of the pth vehicle;

[0051] Step B2. Establish tenant portraits: Combine the constructed driver feature matrix and vehicle feature matrix into a comprehensive feature matrix X tenant =[X d X v . Standardize the combined feature matrix, use the standardized feature matrix to train a machine learning model, and predict each tenant according to the trained model to generate tenant portraits. It further includes the following steps:

[0052] B201. Feature matrix standardization: Standardize the combined feature matrix X tenant to ensure that the mean of each feature is 0 and the standard deviation is 1, improving the stability and performance of the model. The specific calculation formula is as follows:

[0053]

[0054] where X tenant is the combined one, μ X and σ X are the mean and standard deviation of the feature matrix respectively, and X' is the standardized feature matrix;

[0055] Step B202. Model training: Use the standardized feature matrix X' and the target variable to train a decision tree model. The target variable is the classification label of the tenant portrait. Select the optimal feature splitting point by maximizing the information gain, split the data into subsets according to the selected feature, and repeat the splitting process for each subset until the minimum sample number is met. It further includes the following steps:

[0056] Step S1. Calculate the maximum information gain of each feature, select the feature with the largest information gain for splitting. The dataset consists of the feature matrix and the target variable. For node λ, the information entropy is where l q is the probability of class q, and r is the total number of classes; for the subset δ after splitting by feature α, the conditional entropy is: where |δ w | is the number of samples in subset δ w , |δ| is the number of samples in the current node δ, and the information gain is the entropy reduction amount on node δ, denoted as IG(δ,α)=H(δ)-H(δ|α);

[0057] Step S2. Recursive splitting: Recursively apply the same process to each subset until all samples belong to the same class;

[0058] Step B203. Portrait generation: Predict each tenant according to the trained model to generate tenant portraits, predict the order frequency and vehicle selection tendency of the tenant, and use {α1,α2,...,α for the feature setτ} indicates that each leaf node L of the decision tree corresponds to a category C, and the decision tree model is as follows:

[0059]

[0060] Among them, f(x) is the prediction result of the model for the sample x, and C r is the category of the leaf node, and R r is a set of conditions for dividing features;

[0061] The order frequency is predicted as: O η = f O (x η ), where O η represents the order frequency, η represents the tenant, x η is the feature vector of the tenant, and f O is the function of the trained decision tree model for predicting the order frequency;

[0062] The vehicle selection tendency is predicted as: V(η) = f V (x η ), where V(η) represents the vehicle selection tendency, and f V is the function of the trained decision tree model for predicting the vehicle selection tendency;

[0063] In summary, the tenant portrait is represented by a vector as: Among them, O η represents the order frequency, and V(η) represents the vehicle selection tendency;

[0064] Step B3, calculate the matching degree: Obtain the current online car-hailing order information, match it with the tenant portrait, determine the matching degree of each order, and represent the order feature vector as Use the cosine similarity to measure the matching degree between two vectors as follows:

[0065]

[0066] Among them, Matching(η) is the matching degree, Q1 and Q2 are the driver feature and the vehicle feature respectively, O η represents the order frequency, V(η) represents the vehicle selection tendency, and η represents the tenant.

[0067] Preferably, in step 103, the orders are sorted according to the matching degree to obtain the TOPN tenant online car-hailing orders, assign drivers to the TOPN tenant online car-hailing orders, and notify the drivers to pick up passengers. By sorting the orders according to the matching degree, the orders that best meet the tenant's needs are processed first, enabling the drivers to receive orders that are more in line with the route and time arrangements, reducing the empty driving and ineffective waiting time. The specific steps are as follows:

[0068] Step C1, screening TOPN: Traverse each order, calculate its matching degree with each tenant, and create a list to store each order and its calculated matching degree. For each tenant, rank each order based on the calculated matching degree, sort the orders in descending order of the matching degree, and extract the top N orders from the sorted list to form the TOPN tenant online car-hailing orders. By analyzing the matching degree data between the orders and the tenants, the utilization efficiency of drivers and vehicles is maximized;

[0069] Step C2, driver assignment: Assign drivers to each order according to the drivers' preferences and the TOPN orders, and send notifications to each driver assigned to an order, informing them of the pick-up location and time.

[0070] It should be noted that in the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0071] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program code.

[0072] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded computers, or other programmable data processing devices to generate a machine, so that the instructions executed by the processors of the computer or other programmable data processing devices generate means for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0073] These computer program instructions can also be stored in a computer-readable memory capable of guiding a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0074] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are executed on the computer or other programmable apparatus to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one process or a plurality of processes and / or blocks Figure 1 one process or a plurality of processes and / or blocks Figure 1 or steps for implementing the functions specified in a plurality of blocks.

[0075] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to cover the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0076] It is obvious that those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A method for processing TOPN tenant online car-hailing orders based on tenant portraits, characterized in that Specifically, it includes the following steps: Step 101: Collect the basic information and historical order information of the tenant, preprocess and analyze them, and extract the travel characteristics of the tenant; Step 102: According to the extracted travel characteristics of the tenant, establish a tenant portrait model, obtain the current online car-hailing order information, match it with the tenant portrait, and determine the matching degree of each order; Step 103: Sort the orders according to the matching degree to obtain the TOPN tenant online car-hailing orders, assign drivers to the TOPN tenant online car-hailing orders, and notify the drivers to pick up passengers.

2. The TOPN tenant online car-hailing order processing method based on tenant portraits according to claim 1, wherein: In step 101, the basic information and historical order information of the tenant are collected, preprocessed and analyzed to extract the travel characteristics of the tenant. The specific steps are as follows: Step A1: Data collection: Obtain the basic information and historical order information of the tenant from the database of the online car-hailing platform. The basic information of the tenant includes driver information and vehicle information. Among them, the driver information includes gender, age, and whether there is a child seat. The vehicle information includes whether the vehicle is idle, the luxury type and cleanliness of the vehicle. The historical order information includes order time, starting location, ending location, and riding frequency; Step A2: Data preprocessing: Clean the collected data, including removing duplicate data, handling missing values and outliers. According to the collected information, extract the travel characteristics of the tenant. The travel characteristics include driver characteristics, vehicle characteristics, and historical order characteristics, and perform feature encoding.

3. The TOPN tenant online car-hailing order processing method based on tenant portraits according to claim 2, wherein, In step A2 data preprocessing, according to the collected information, extract driver characteristics, vehicle characteristics, and historical order characteristics, and perform feature encoding, which further includes the following steps: Step A201: Binary code the extracted gender, presence of child seat, and availability, using {G i , C i , F i} ∈ (0, 1) to represent gender, child seat, and availability, where G0 represents male, G1 represents female, C0 represents no child seat, C1 represents having a child seat, F0 represents not available, and F1 represents available; Step A202: Bin the extracted age, convert it into a categorical variable, divide the age into m intervals, and use A j ∈(0, 1, ..., m - 1) to represent age; use L j ∈(0, 1, ..., k - 1) to represent the luxury type, where k is the number of categories of the luxury type; use D j ∈(0, 1, 2) to represent the cleanliness level, which is divided into low, medium, and high grades; use T j ∈(0, 1, ..., 23) to represent the order time, and the extracted time is in 24-hour format.

4. A method for processing TOPN tenant online car-hailing orders based on tenant portraits according to claim 1, characterized in that, In step 102, according to the extracted travel characteristics of the tenant, establish a tenant portrait, obtain the current online car-hailing order information, match it with the tenant portrait, and determine the matching degree of each order. The specific steps are as follows: Step B1. Construct feature matrices: For each feature, construct its feature matrix in the tenant. The driver feature matrix is expressed as: X d = [x d1 x d2 ...x dn , where x dn represents the gender, age, and child seat information of the nth driver; the vehicle feature matrix is expressed as: X v = [x v1 x v2 ...x vp , where x vp represents the idle status, luxury type, and cleanliness of the pth vehicle. Step B2. Establish tenant portraits: Merge the constructed driver feature matrix and vehicle feature matrix into a comprehensive feature matrix X tenant = [X d X v . Standardize the merged feature matrix, use the standardized feature matrix to train a machine learning model, and predict each tenant according to the trained model to generate tenant portraits; Step B3. Calculate the matching degree: Obtain the current online car-hailing order information, match it with the tenant portrait, determine the matching degree of each order, and represent the order feature vector as Use cosine similarity to measure the matching degree between two vectors as follows: Among them, Matching(η) is the matching degree, Q1 and Q2 are the driver characteristics and vehicle characteristics respectively, and O η represents the order frequency, V(η) represents the vehicle selection tendency, and η represents the tenant.

5. The method for processing TOPN tenant online car-hailing orders based on tenant portraits according to claim 4, wherein, In step B2 of establishing the tenant portrait, use the standardized feature matrix to train the machine learning model, and according to the trained model, predict each tenant to generate a tenant portrait, which further includes the following steps: B201. Feature matrix standardization: Standardize the merged feature matrix X tenant to ensure that the mean of each feature is 0 and the standard deviation is 1, improving the stability and performance of the model. The specific calculation formula is as follows: Among them, X tenant is the merged one, μ X and σ X are the mean and standard deviation of the feature matrix respectively, and X' is the standardized feature matrix; Step B202: Model training: Use the standardized feature matrix X' and the target variable to train the decision tree model. The target variable is the classification label of the tenant portrait. Select the optimal feature division point by maximizing the information gain, split the data into subsets according to the selected feature, and repeat the splitting process for each subset until the minimum sample quantity is met; Step B203, Image Generation: According to the trained model, predict for each tenant to generate a tenant image, predict the order frequency and vehicle selection tendency of the tenant, and represent the feature set as {α1, α2,..., α τ}, and each leaf node L of the decision tree corresponds to a category C. The decision tree model is as follows: where f(x) is the prediction result of the model for the sample x, C r is the class of the leaf node, and R r is a set of conditions for dividing features; The predicted order frequency is: O η = f O (x η ), where O η represents the order frequency, η represents the tenant, and x η is the feature vector of the tenant, and f O is the function of the trained decision tree model for predicting the order frequency; The vehicle selection tendency prediction is: V(η) = f V (x η ), where V(η) represents the vehicle selection tendency, and f V is a function of the trained decision tree model for predicting the vehicle selection tendency; In summary, the tenant profile is represented by a vector as follows: where O η represents the order frequency, and V(η) represents the vehicle selection tendency.

6. The TOPN tenant online car-hailing order processing method based on tenant portraits according to claim 5, characterized in that, In step B202 of model training, select the optimal feature division point by maximizing the information gain, split the data into subsets according to the selected feature, and repeat the splitting process for each subset until the minimum sample quantity is met, which further includes the following steps: Step S1: Calculate the maximum information gain of each feature, and select the feature with the maximum information gain for partitioning. The dataset consists of a feature matrix and a target variable. For node λ, the information entropy is where l q is the probability of class q, and r is the total number of classes; for the subset δ after partitioning by feature α, the conditional entropy is: where |δ w | is the number of samples in subset δ w , |δ| is the number of samples in the current node δ, and the information gain is the amount of entropy reduction on node δ, denoted as IG(δ,α) = H(δ) - H(δ|α); Step S2: Recursive partitioning: Recursively apply the same process to each subset until all samples belong to the same category.

7. A method for processing TOPN tenant online car-hailing orders based on tenant portraits according to claim 1, characterized in that In step 103, sort the orders according to the matching degree to obtain the TOPN tenant online car-hailing orders, assign drivers to the TOPN tenant online car-hailing orders, and notify the drivers to pick up passengers. The specific steps are as follows: Step C1. Screening TOPN: Traverse each order, calculate its matching degree with each tenant, and create a list to store each order and its calculated matching degree. For each tenant, rank each order based on the calculated matching degree, sort the orders in descending order of the matching degree, and extract the top N orders from the sorted list to form the TOPN tenant online car-hailing orders. Step C2. Assigning drivers: Assign drivers to each order according to the drivers' preferences and the TOPN orders, and send notifications to each driver assigned to an order, informing them of the pick-up location and time.

Images