Real-time user behavior monitoring method and system based on online car-hailing

Abstract

The invention discloses a real-time user behavior monitoring method and system based on online car-hailing. The real-time user behavior monitoring method comprises the following steps: acquiring position information, driving behavior data and service record data of a driver in real time; processing the position information, the driving behavior data and the service record data of the driver, and constructing a position information model, a driving behavior data model and a service record model; processing the position information model, the driving behavior data model and the service recording model by using a data processing algorithm to obtain processed driver behavior data; analyzing the processed driver behavior data by using a machine learning algorithm to obtain analyzed driver behavior data; and monitoring the analyzed driver behavior data in real time to obtain monitored driver behavior data. The driving behavior and the service attitude of the driver can be accurately evaluated, personalized training and improvement suggestions are provided for the driver, and the abnormal behavior of the driver in the service process can be identified.

Description

Technical Field

[0001] The present invention relates to a technical field of online car-hailing monitoring and management, and more specifically, to a real-time user behavior monitoring method and system based on online car-hailing. Background Art

[0002] With the continuous development of Internet technology, the Internet provides us with many things that can be done without leaving home, so online taxi booking (hereinafter referred to as online taxi booking) has emerged, which allows people to make reservations in advance and get on the car smoothly to complete their trip. The emergence of "online taxi booking" provides urban residents with a convenient and reservable taxi travel mode, effectively meeting the travel needs of some taxi passengers. At the same time, since private cars can apply to become online taxi owners, the utilization efficiency of motor vehicles has been improved to a certain extent.

[0003] With the rapid development of online ride-hailing services, the service quality and efficiency of drivers are crucial to the competitiveness of the entire industry. Traditional online ride-hailing platforms have certain deficiencies in driver behavior monitoring and analysis, such as untimely data collection, inaccurate service evaluation, and inefficient driver management. Therefore, there is an urgent need for an online ride-hailing platform that can monitor and analyze driver behavior data in real time, efficiently, and accurately to optimize service quality and enhance market competitiveness. Summary of the invention

[0004] In view of this, the present invention provides a real-time user behavior monitoring method based on online car-hailing, so as to solve the problems of untimely data collection, inaccurate service evaluation and inefficient driver management.

[0005] In a first aspect, the present application provides a real-time user behavior monitoring method based on online car-hailing, including: obtaining the driver's location information, driving behavior data, and service record data in real time;

[0006] Processing the driver's location information, the driving behavior data, and the service record data to construct a location information model, a driving behavior data model, and a service record model;

[0007] Using a data processing algorithm to process the location information model, the driving behavior data model, and the service record model to obtain processed driver behavior data;

[0008] Analyzing the processed driver behavior data using a machine learning algorithm to obtain analyzed driver behavior data; which includes: analyzing the processed driver behavior data using a classification algorithm and performing sentiment analysis on the passenger's text information; or, grouping the processed driver behavior data using a clustering algorithm to identify different behavior patterns;

[0009] The analyzed driver behavior data is monitored in real time to obtain monitored driver behavior data.

[0010] In the second aspect, the present application provides a real-time user behavior monitoring system based on online car-hailing, including:

[0011] Real-time data acquisition module, used to obtain the driver's location information, driving behavior data and service record data in real time;

[0012] A model definition module, connected to the real-time data acquisition module, for receiving the driver's location information, driving behavior data and service record data, processing the driver's location information, driving behavior data and service record data, and constructing a location information model, a driving behavior data model and a service record model;

[0013] a data processing and analysis module, connected to the model definition module, for receiving a location information model, a driving behavior data model and a service record model, processing the location information model, the driving behavior data model and the service record model using a data processing algorithm to obtain processed driver behavior data, and analyzing the processed driver behavior data using a machine learning algorithm to obtain analyzed driver behavior data, which includes: analyzing the processed driver behavior data using a classification algorithm and performing sentiment analysis on passenger text information; or grouping the processed driver behavior data using a clustering algorithm to identify different behavior patterns;

[0014] The data monitoring module is connected to the data processing and analysis module, and is used to receive the analyzed driver behavior data, monitor the analyzed driver behavior data in real time, and obtain the monitored driver behavior data.

[0015] Compared with the prior art, the real-time user behavior monitoring method based on online car-hailing provided by the present invention achieves at least the following beneficial effects:

[0016] The present invention provides a real-time user behavior monitoring method based on online car-hailing and a system thereof. The real-time user behavior monitoring method includes: obtaining the driver's location information, driving behavior data and service record data in real time; processing the driver's location information, driving behavior data and service record data to construct a location information model, a driving behavior data model and a service record model; processing the location information model, the driving behavior data model and the service record model using a data processing algorithm to obtain processed driver behavior data; analyzing the processed driver behavior data using a machine learning algorithm to obtain analyzed driver behavior data, which includes: analyzing the processed driver behavior data using a classification algorithm and performing sentiment analysis on passenger text information; or grouping the processed driver behavior data using a clustering algorithm to identify different behavior patterns; monitoring the analyzed driver behavior data in real time to obtain monitored driver behavior data. The above scheme can obtain and analyze driver behavior data in real time, help managers of online car-hailing platforms to adjust service strategies in a timely manner and optimize service efficiency; accurately evaluate the driver's driving behavior and service attitude in combination with a machine learning algorithm, provide personalized training and improvement suggestions for the driver, and also identify abnormal behavior of the driver during the service process.

[0017] Of course, any product implementing the present invention does not necessarily need to achieve all of the technical effects described above at the same time.

[0018] Further features and advantages of the present invention will become apparent from the following detailed description of exemplary embodiments of the present invention with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

[0020] Figure 1 It is one of the flow charts of the real-time user behavior monitoring method based on online car-hailing provided by the present invention;

[0021] Figure 2 This is the second flow chart of the real-time user behavior monitoring method based on online car-hailing provided by the present invention;

[0022] Figure 3 It is a structural schematic diagram of the real-time user behavior monitoring system based on online car-hailing provided by the present invention. DETAILED DESCRIPTION

[0023] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless otherwise specifically stated.

[0024] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0025] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0026] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0027] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0028] Embodiment 1

[0029] See also Figure 1 and Figure 2 As shown, Figure 1 It is one of the flow charts of the real-time user behavior monitoring method based on online car-hailing provided by the present invention; Figure 2 This is the second flow chart of the real-time user behavior monitoring method based on online car-hailing provided by the present invention; this embodiment provides a real-time user behavior monitoring method based on online car-hailing, including:

[0030] Step 101: Acquire the driver's location information, driving behavior data, and service record data in real time;

[0031] Specifically, based on map information, the driver's location information and driving behavior data can be obtained in real time. Specifically, the driver's geographic location data is collected in real time through third-party map technology in the online car-hailing driver-side App (Application). Through third-party map technology, the online car-hailing driver-side App can provide more accurate and efficient navigation services, thereby improving user experience and safety; the driver's location information includes longitude (indicating the driver's specific geographic location), latitude (indicating the driver's specific geographic location), speed (indicating the driver's current speed), driving direction (indicating the driver's current driving direction) and timestamp (recording the time of data collection).

[0032] The third-party map SDK (Software Development Kit) in the online ride-hailing driver app collects information such as acceleration (indicating the degree to which the driver accelerates in a short period of time), braking (indicating the degree to which the driver brakes in a short period of time), turning, speed, speeding (indicating the situation where the driver exceeds the prescribed speed), fatigue driving (indicating the situation where the driver drives for a long time) and timestamps, mainly to provide more accurate navigation services and driving safety warnings. Driving behavior data includes acceleration, braking, turning, speed, speeding, fatigue driving and timestamps.

[0033] The collection of the above information is crucial to improving user experience and ensuring driving safety. For example, by real-time monitoring of the vehicle's acceleration, braking and turning conditions, the navigation route can be optimized to avoid congestion and dangerous sections, thereby improving driving efficiency and safety.

[0034] Obtain service record data based on the detailed information of each service. Online ride-hailing platforms usually automatically record the detailed information of each service, including the start and end time of the trip (the App will automatically record the time when the passenger gets on and off the car), the distance of the trip (the App will record the starting and ending points of the passenger and calculate the distance of the trip), etc., and allow passengers and drivers to evaluate the service. This information helps improve service quality and management.

[0035] Step 102: Process the driver's location information, driving behavior data, and service record data to construct a location information model, a driving behavior data model, and a service record model;

[0036] Specifically, see Figure 2 As shown, step 1021, using the message queue to receive the driver's location information, driving behavior data and service record data, that is, using the message queue to receive the real-time data stream, to ensure that the driver's location information, driving behavior data and service record data can be quickly transmitted without loss, and the above message queue can be multiple message queues.

[0037] Step 1022: clean the driver's location information, driving behavior data, and service record data to obtain cleaned driver's location information, driving behavior data, and service record data. The above scheme can use a filter algorithm to clean the driver's location information, driving behavior data, and service record data to remove duplicate, erroneous, and irrelevant data, that is, it can remove data points that do not meet the requirements or affect data quality, thereby ensuring the accuracy of subsequent analysis.

[0038] Step 1023: Convert the cleaned driver's location information, driving behavior data, and service record data into a unified format by using data standardization and formatting processing;

[0039] The above data standardization is the process of converting data from different sources and formats into a unified format and standard. When processing driver location information, driving behavior data, and service record data, data standardization can help us ensure the uniformity and consistency of the data, which is convenient for subsequent analysis and processing.

[0040] The above-mentioned data formatting process refers to organizing and converting data into a predetermined format for easy storage, transmission and analysis. For the driver's location information, driving behavior data and service record data, data formatting process may include the following aspects:

[0041] Position information formatting

[0042] Timestamp standardization: Standardize the timestamps of all location information to the ISO 8601 format (for example: 2023-10-05T14:48:00Z).

[0043] Coordinate system unification: Unify the coordinate system of all location information to WGS 84 (latitude and longitude format).

[0044] Data field standardization: Ensure that each location information record contains the following fields: timestamp, longitude, latitude, speed, direction, etc.

[0045] The above-mentioned ISO 8601 is an international standard, and its full name is "Data storage and exchange formats, information exchange, date and time representation methods". This standard was developed by the International Organization for Standardization (ISO) to provide a unified format for representing dates and times so that data can be exchanged and recorded worldwide. The purpose of ISO 8601 is to reduce misunderstandings caused by regional customs and format differences, while enhancing the convenience and consistency of date and time data exchange between computer systems.

[0046] Driving behavior data formatting

[0047] Timestamp standardization: Also standardize the timestamps of all driving behavior data to the ISO 8601 format.

[0048] Behavior type standardization: uniformly encode driving behaviors (such as acceleration, deceleration, and turning) to ensure that each behavior type has a unique identifier.

[0049] Data field standardization: Ensure that each driving behavior record contains the following fields: timestamp, behavior type, behavior intensity (such as acceleration value) and duration, etc.

[0050] Service record data format

[0051] Timestamp standardization: Standardize the timestamps of all service record data to the ISO 8601 format.

[0052] Standardization of service types: uniformly encode service types (such as order acceptance, order completion, order cancellation, etc.) to ensure that each service type has a unique identifier.

[0053] Data field standardization: Ensure that each service record contains the following fields: timestamp, service type, service ID (Identification), driver ID, passenger ID, start location, end location, etc.

[0054] Through the above-mentioned standardization and formatting processing, it can be ensured that the cleaned driver location information, driving behavior data and service record data have consistent formats and standards. All data will be integrated into a unified format to facilitate subsequent analysis and application. This information involves integrating different types of data (such as location information, driving behavior data and service record data) into a database table to ensure that each field has a clear definition and purpose.

[0055] Step 1024: Serialize the driver's location information, driving behavior data, and service record data converted into a unified format into a structured format, and construct a location information model, a driving behavior data model, and a service record model;

[0056] The driver's location information, driving behavior data, and service record data converted into a unified format are serialized into a structured format through a serialization algorithm.

[0057] The elements of the driver's location information include longitude, latitude, speed, driving direction and timestamp. The above elements together constitute the driver's location information model, which is used to describe the driver's location status at a certain moment;

[0058] The elements of driving behavior data include acceleration, braking, speeding, fatigue driving and timestamp. The above elements together constitute a driving behavior data model, which is used to describe the driver's driving behavior over a period of time.

[0059] The elements of service record data include order ID, vehicle ID, passenger name, departure place, destination, driver information, vehicle information, service time and service fee. The above elements together constitute the service record model, which is used to record the passenger's riding experience. The online car-hailing platform can better understand user needs, optimize service processes, and provide more convenient, reasonably priced, safe and reliable services.

[0060] The above-mentioned location information model, driving behavior data model and service record model can ensure the consistency of data structure.

[0061] Step 103: Process the location information model, the driving behavior data model, and the service record model using a data processing algorithm to obtain processed driver behavior data;

[0062] The above data processing algorithm includes a time window algorithm and a programming algorithm. The programming algorithm may be a MapReduce algorithm, which is a distributed algorithm for processing and generating large data sets.

[0063] The above time window algorithm is mainly used to process data streams with time attributes. It is used to process time-related data streams, such as order processing and route planning. The MapReduce algorithm is used to process and generate large data sets. It breaks down large-scale data processing tasks into multiple small, independent subtasks that can be processed in parallel on a large number of computers. The MapReduce algorithm consists of two main phases: the Map phase and the Reduce phase. The Map phase is responsible for the initial processing of the data, and the Reduce phase is responsible for merging the intermediate results output by the Map phase to generate the final result data.

[0064] The location information model, driving behavior data model and service record model are processed using data processing algorithms to obtain the processed driver behavior data including:

[0065] Use the time window algorithm to perform real-time aggregation analysis on the location information model, driving behavior data model, and service record model to obtain statistical information within a preset time interval;

[0066] The statistical information within a preset time interval is processed in parallel using a programmed algorithm.

[0067] The above-mentioned time window algorithm divides the continuous data stream into certain time intervals, and aggregates and analyzes the location information model, driving behavior data model and service record model in each time window, so as to obtain statistical information within the preset time interval, such as the order volume, order completion rate, average waiting time, order confirmation information, driver arrival notification, trip start reminder, trip end notification and evaluation and feedback invitation from 7-9 am on weekdays.

[0068] The MapReduce algorithm is used to parallelize the statistical information within the preset time interval and perform parallel calculations on the data in each time window, which can greatly improve the processing efficiency.

[0069] The above scheme first uses the time window algorithm to filter out the statistical information within the preset time interval that needs further processing, and then uses the MapReduce algorithm to process and analyze the statistical information within the preset time interval on a large scale, effectively improving data processing efficiency and accuracy.

[0070] It should be noted that the data processing algorithm also includes data indexing. For example, when building an index, you can use common query fields to build an index to improve query efficiency. Before using the time window algorithm to perform real-time aggregation analysis on the location information model, driving behavior data model, and service record model, you should first build an index. The location information model, driving behavior data model, and service record model have their own query fields. Specifically,

[0071] The location information model usually includes fields such as vehicle ID, timestamp, latitude and longitude, speed, direction, etc. To improve query efficiency, you can build indexes for the following fields:

[0072] Vehicle ID: Used to quickly find the location information of a specific vehicle.

[0073] Timestamp: used to query the historical location of the vehicle in chronological order.

[0074] Latitude and longitude: used for spatial queries, such as finding vehicles in a specific area.

[0075] The driving behavior data model may include fields such as vehicle ID, timestamp, acceleration, braking force, turning angle, etc. In order to analyze driving behavior, you can build indexes on the following fields:

[0076] Vehicle ID: Used to find driving behavior data for a specific vehicle.

[0077] Timestamp: Used to analyze driving behavior in chronological order.

[0078] Acceleration, braking force, turning angle: used for querying and analyzing specific driving behaviors.

[0079] The service record model usually includes fields such as order ID, vehicle ID, passenger ID, start time, end time, travel distance, service fee, etc. In order to quickly query and count service records, you can build indexes for the following fields:

[0080] Order ID: Used to find details of a specific order.

[0081] Vehicle ID: Used to find all service records for a specific vehicle.

[0082] Passenger ID: Used to find all orders for a specific passenger.

[0083] Start time, end time: used to query service records by time range.

[0084] Based on the index construction, real-time aggregation analysis (such as time window algorithm) is performed to support real-time analysis and decision-making.

[0085] Real-time location aggregation: Based on the vehicle ID and timestamp, the average speed, driving trajectory, etc. of the vehicle are calculated in real time.

[0086] Driving behavior aggregation: Analyze the vehicle's driving habits in real time based on the vehicle ID and timestamp, such as the frequency of sudden acceleration and sudden braking.

[0087] Service record aggregation: Based on vehicle ID, passenger ID and time range, real-time statistics of vehicle operation efficiency, passenger frequency, etc.

[0088] Step 104: Analyze the processed driver behavior data using a machine learning algorithm to obtain the analyzed driver behavior data, and use the machine learning algorithm to accurately evaluate the driver's driving behavior and service attitude, and provide personalized training and improvement suggestions for the driver.

[0089] Specifically, machine learning algorithms include classification algorithms and clustering algorithms;

[0090] Analyze the processed driver behavior data using classification algorithms and perform sentiment analysis on the passenger text messages; or

[0091] Clustering algorithms are used to group the processed driver behavior data and identify different behavior patterns.

[0092] The above solution builds a classification model for the processed driver behavior data, which can determine whether the driver meets the predetermined service quality standards based on the data. This analysis not only helps to improve the passenger's riding experience, but also improves the platform's operating efficiency and service quality.

[0093] The processed driver behavior data (driving speed, acceleration, braking frequency, passenger evaluation and complaint records) are processed and analyzed using classification algorithms. Classification algorithms can classify drivers according to their behavioral characteristics. For example, drivers can be divided into two categories: "high-quality service" and "non-high-quality service". By training a classification model, the classification model can predict the category of the driver's service quality based on the driver's behavior data. For example, using algorithms such as decision trees, logistic regression, and support vector machines (SVM), these algorithms can learn classification rules based on historical data and then predict and classify new data.

[0094] Based on the results of the classification model, drivers can be evaluated and ranked to identify which drivers have met the service quality standards and which need improvement. This evaluation result can be used to optimize driver selection criteria, improve training effectiveness, and adjust incentives to improve overall service quality and passenger satisfaction.

[0095] An important application area is to use classification algorithms to analyze the driver's behavior data to determine whether they meet the service quality standards. By collecting and analyzing the driver's driving behavior, service attitude and other related data, a classification model can be constructed, which can determine whether the driver meets the predetermined service quality standards based on these data. This analysis not only helps to improve the passenger's riding experience, but also improves the platform's operating efficiency and service quality.

[0096] The processed driver behavior data (number of orders, order time, driving distance, service time and service evaluation) are clustered using the K-means clustering algorithm, such as clustering similar driver behaviors together. The K-means algorithm iteratively distributes data points into K clusters, each of which is represented by its average value. The K-means clustering algorithm is a commonly used unsupervised learning method.

[0097] The above clustering further includes: randomly selecting K initial centroids, then calculating the distance of each data point to each centroid, assigning it to the nearest centroid to form a cluster. Subsequently, recalculating the centroid of each cluster, repeating this process until the centroid no longer changes or changes very little, and convergence is achieved.

[0098] After clustering is completed, the characteristics of each cluster are analyzed, such as order frequency, service time, user evaluation, etc., to identify different driver behavior patterns. For example, some clusters may represent efficient service providers who respond quickly and complete orders, while other clusters may represent inefficient service providers who may spend more time on the road or wait longer.

[0099] Classification algorithms are used to analyze the processed driver behavior data (such as order acceptance speed, service attitude, driving route and other dimensions) to identify the driver's specific behavior patterns.

[0100] Clustering algorithms are used to divide drivers into different groups or "clusters" based on processed driver behavior data. Drivers within each cluster have similar behaviors, while drivers between different clusters show different behavior patterns. Based on the clustering results, the platform can develop personalized training plans for drivers in each cluster. For example, for drivers with poor performance, special training courses can be provided to improve their service quality and efficiency. For drivers with excellent performance, further incentives or promotion opportunities can be provided.

[0101] Through personalized training methods, online ride-hailing platforms can not only improve service quality and better meet the needs of passengers, but also help improve drivers' job satisfaction and income levels.

[0102] Step 105: Monitor the analyzed driver behavior data in real time to obtain monitored driver behavior data.

[0103] Specifically, in the above scheme, the video surveillance system can be used to monitor the analyzed driver behavior data in real time. The real-time monitoring content may include the driver's abnormal behavior, such as smoking, making phone calls, not wearing seat belts and yawning. Of course, it is also possible to identify whether the driver has abnormal behavior by real-time monitoring of the driver's facial information, human body movement posture, cigarette recognition, vehicle trajectory data, location information and speed information, etc. The monitored driver behavior data is obtained based on the real-time monitoring content, and the driver's abnormal behavior during the service process is identified based on the monitored driver behavior data, and an alarm message is sent according to the risk level of the abnormal behavior.

[0104] It can be seen from the above embodiments that the real-time user behavior monitoring method based on online car-hailing provided in this embodiment achieves at least the following beneficial effects:

[0105] The real-time user behavior monitoring method based on online car-hailing provided in this embodiment includes: obtaining the driver's location information, driving behavior data and service record data in real time; processing the driver's location information, driving behavior data and service record data to construct a location information model, a driving behavior data model and a service record model; processing the location information model, the driving behavior data model and the service record model using a data processing algorithm to obtain processed driver behavior data; analyzing the processed driver behavior data using a machine learning algorithm to obtain analyzed driver behavior data, which includes: analyzing the processed driver behavior data using a classification algorithm and performing sentiment analysis on the passenger's text information; or, grouping the processed driver behavior data using a clustering algorithm to identify different behavior patterns; monitoring the analyzed driver behavior data in real time to obtain monitored driver behavior data. The above scheme can obtain and analyze driver behavior data in real time, helping managers of online car-hailing platforms to adjust service strategies in a timely manner and optimize service efficiency; combining machine learning algorithms to accurately evaluate drivers' driving behavior and service attitudes, provide drivers with personalized training and improvement suggestions, and also identify abnormal behaviors of drivers during the service process.

[0106] Optionally, the processed driver behavior data is grouped using a clustering algorithm, and after different behavior patterns are identified, the following steps are further included:

[0107] Identify preferences of analyzed driver behavior data based on different behavior patterns and changes over time.

[0108] The above different behavior patterns include order frequency, driving route, service evaluation and service time. The order frequency, driving route, service evaluation and service time are analyzed over time to identify the preferences of the driver behavior data after analysis. Specifically,

[0109] Order frequency analysis: By recording the number of orders a driver receives each day, you can understand the driver's work activity. The order frequency may be higher during peak hours and lower during off-peak hours, which helps to understand the driver's busy hours and rest time.

[0110] Driving route analysis: The driver's driving route can be tracked through GPS (Global Positioning System) data or Beidou data. Frequently driven routes may be the driver's frequently used routes, or routes with high passenger demand. By analyzing this data, the driver's service area and preferred routes can be discovered.

[0111] Service evaluation analysis: Comprehensive assessment of service quality, driver service attitude, vehicle condition, safety measures and user satisfaction, so as to continuously improve service quality and meet user needs.

[0112] Service time analysis: By recording the start and end time of the driver's service, you can understand the driver's working hours and rest time. This helps analyze the driver's work efficiency and life balance.

[0113] Behavioral preference identification: Through comprehensive analysis of the above data, it is possible to identify the behavioral preferences of drivers. For example, some drivers may prefer to accept orders during specific time periods, or prefer to serve passengers in specific areas. These preferences may be related to factors such as the driver's personal situation, home address, and familiarity with a specific area.

[0114] By adopting the above solution, online ride-hailing platforms can better understand drivers’ behavior patterns, thereby optimizing dispatch algorithms and improving service efficiency. They can also provide drivers with more reasonable work schedules and suggestions to improve work efficiency and passenger satisfaction. In addition, this data analysis can also help the platform identify bottlenecks and problems in the service, thereby improving service quality and efficiency.

[0115] In an optional embodiment, after processing the driver's location information, driving behavior data, and service record data to construct the location information model, the driving behavior data model, and the service record model, the following further comprises:

[0116] Encrypting the processed driver behavior data to obtain encrypted driver behavior data;

[0117] Verification is performed based on the encrypted driver behavior data to obtain verified driver behavior data;

[0118] Permission management is performed based on the verified driver behavior data to obtain the managed driver behavior data.

[0119] The above encryption ensures the security of the processed driver behavior data during transmission and storage. Specifically, transmission encryption focuses on the protection of data during transmission, preventing data from being intercepted or tampered with during transmission, and ensuring that data is safely delivered from the sender to the receiver. Storage encryption focuses on protecting the data stored in the system, converting the processed driver behavior data through encryption algorithms to prevent unauthorized access and data leakage.

[0120] In this embodiment, encryption protocols such as TLS / SSL (Transport Layer Security / Secure Sockets Layer) are used to ensure the security of the processed driver behavior data during transmission. Encryption algorithms such as AES (Advanced Encryption Standard) are used to encrypt the stored processed driver behavior data to protect the processed driver behavior data from unauthorized access.

[0121] In this embodiment, transmission encryption can be performed first to ensure the security of the processed driver behavior data from the starting point to the end point, and then storage encryption can be performed to protect the security of the processed driver behavior data stored in the system. This approach can maximize the integrity and confidentiality of user data.

[0122] The above verification based on the encrypted driver behavior data is called identity verification, which ensures that the identity information of passengers and drivers is authentic and reliable, thereby ensuring the security and reliability of platform services. It can be understood as identity verification based on the encrypted driver behavior data, ensuring that only users who have passed the identity verification can access the data.

[0123] The above-mentioned permission management is performed based on the verified driver behavior data. In this embodiment, the permission management involves two main permission management models: role-based access control (RBAC) and attribute-based access control (ABAC), in which:

[0124] The RBAC model manages permissions by defining user roles. Each role has a specific set of permissions, and users are assigned to one or more roles, thereby inheriting the corresponding permissions. The RBAC model can be used to manage permissions for different users, such as the permissions allocation for driver users and managers.

[0125] The ABAC model dynamically determines access rights through user attributes, environment attributes, and resource attributes. Compared with the RBAC model, ABAC is more flexible and can dynamically adjust permissions based on actual conditions, but its implementation complexity is relatively high. In online car-hailing applications, the ABAC model can be used to implement more fine-grained permission control, such as dynamically allocating or adjusting permissions based on the user's real-time location, service status, and other attributes.

[0126] In this embodiment, either role-based access control (RBAC) or attribute-based access control (ABAC) can be used. Alternatively, a combination of role-based access control (RBAC) and attribute-based access control (ABAC) can be used according to actual conditions to achieve more flexible and fine-grained access control in the online car-hailing platform. For example, different permissions can be assigned to different roles (such as administrators, drivers, and passengers), and permissions can be further refined based on the attributes of these roles (such as geographic location, user level). In this way, not only can the security of the platform and the protection of data be ensured, but also a personalized service experience can be provided.

[0127] In an optional embodiment, the analyzed driver behavior data is monitored in real time, and the monitored driver behavior data includes:

[0128] A real-time data dashboard is established based on the analyzed driver behavior data, and the real-time statistics of driver behavior are displayed on the real-time data dashboard. The real-time statistics of driver behavior include time data and violation data;

[0129] When the video surveillance system detects abnormal or illegal behavior of the driver, the online ride-hailing platform automatically sends alarm data to the manager.

[0130] The above real-time data dashboard can display the analyzed driver behavior data. The time data includes the average response time and the average travel time. The average response time refers to the average time from the passenger's request for a ride to the driver's acceptance of the order. It reflects the efficiency of the driver resource allocation of the online ride-hailing platform and the driver's enthusiasm for responding to orders. The average travel time refers to the average time it takes for passengers to get from the pick-up point to the drop-off point. It is affected by many factors, such as traffic conditions, travel distance, and driver driving habits. Violation data includes the number of violations, drivers' refusal to pick up passengers, detours, and fatigue driving.

[0131] The real-time data dashboard displays the average response time, average travel time, violation data, number of online drivers, order response rate, mileage and duration, and passenger evaluations. Specifically, the real-time data dashboard can display the number of online drivers within a certain period of time in real time, helping managers understand the operating status and scheduling needs of the platform.

[0132] Average response time: This data can be used to understand the driver's order acceptance situation in a timely manner so that corresponding measures can be taken. For example, when the average response time suddenly becomes longer, the platform can adjust the incentive mechanism (such as giving drivers more subsidies) to increase the driver's enthusiasm for accepting orders, or optimize the order allocation algorithm to more reasonably allocate orders to nearby drivers.

[0133] Average travel time: This data can help optimize route planning suggestions by providing real-time information on average travel time. If the average travel time in a certain area is found to be abnormally long, there may be a traffic bottleneck in that area. The platform can consider adjusting navigation suggestions or reminding drivers to avoid that area.

[0134] Violation data, which can show the number of violations and allow the platform to detect abnormal situations and handle them in a timely manner.

[0135] Order response rate: By counting the timeliness of drivers accepting orders, we can evaluate the driver’s service quality and the platform’s response efficiency.

[0136] Mileage and duration: The real-time data dashboard can display the driver’s mileage and duration over a certain period of time, which helps analyze the driver’s work intensity and efficiency.

[0137] Passenger evaluation: The driver's behavior and service quality can be reflected through passengers' evaluation. The real-time data dashboard can display these evaluation data in real time to help the platform monitor service quality.

[0138] The above abnormal driver behaviors include but are not limited to smoking, talking on the phone, not wearing seat belts and yawning.

[0139] When the video surveillance system detects abnormal driver behavior or violation data, the online ride-hailing platform can warn, train or punish the driver, identify potential safety hazards and violations, and take timely preventive measures to ensure the safety of passengers and drivers.

[0140] In an optional solution, the real-time user behavior monitoring method based on online car-hailing further includes:

[0141] Use machine learning algorithms to predict historical driver behavior data and obtain prediction results;

[0142] Based on the prediction results and real-time driver behavior data, drivers are dynamically adjusted to the predicted area.

[0143] The machine learning algorithms in the above scheme include linear regression, decision tree, random forest, gradient boosting machine and neural network.

[0144] Choose an appropriate machine learning algorithm such as linear regression, decision tree, random forest, gradient boosting machine, or neural network.

[0145] The historical driver behavior data is divided into a training set and a test set. The model is trained using the training set data, and the model performance is evaluated on the test set.

[0146] The accuracy of model predictions was evaluated using appropriate metrics such as mean square error, mean absolute error, etc. Model parameters were further optimized through cross-validation to ensure that the model had good generalization ability.

[0147] Dynamic adjustment strategy development: Based on the forecast results, design a dynamic adjustment of drivers to the forecast area. For example, when the demand in the forecast area will increase, more drivers will be dispatched to the forecast area in advance; otherwise, the number of drivers going to the forecast area will be reduced to avoid wasting resources.

[0148] This embodiment uses historical driver behavior data and prediction models to predict future passenger demand (such as the number of orders, time, and location) and driver supply (such as the number of online drivers, their location, and available time) so as to adjust resources in advance. According to the prediction results and real-time data (such as driver location and passenger demand distribution), drivers are dynamically adjusted to the predicted area to ensure service coverage and work efficiency.

[0149] Through the above steps, not only the accuracy of the prediction can be improved, but also the dispatch efficiency can be effectively improved, ensuring that passenger needs are met in a timely manner while maximizing the driver's work efficiency.

[0150] In an optional embodiment, the analyzed driver behavior data is monitored in real time, and after obtaining the monitored driver behavior data, the following steps are further included:

[0151] The monitored driver behavior data is tracked in real time to obtain the tracked driver behavior data;

[0152] Based on the tracked driver behavior data and feedback mechanism, optimized driver behavior data is obtained.

[0153] In this solution, cameras or other sensors installed on online ride-hailing vehicles can be used to collect driver behavior data in real time and upload it to the online ride-hailing platform in real time. The online ride-hailing platform analyzes the uploaded driver behavior data and evaluates whether the driver behavior data complies with regulations, such as whether there is abnormal driving behavior (such as smoking, yawning, deliberately driving slowly, taking detours, etc.). Once abnormal behavior is detected, the online ride-hailing platform can immediately issue a warning or prompt message to the driver to urge him to correct his behavior.

[0154] Online ride-hailing platforms should establish effective feedback channels so that drivers can receive timely evaluation information about their behavior, including positive encouragement and negative warnings. Based on the performance differences of each driver, online ride-hailing platforms can provide personalized training to help drivers improve bad driving habits and improve service quality. By establishing reward mechanisms, such as excellent service awards and safe driving awards, drivers can be encouraged to improve their service attitudes and driving behaviors. Online ride-hailing platforms should continuously collect user satisfaction feedback on their services, and constantly adjust and optimize monitoring systems and technologies to better adapt to market changes and service needs.

[0155] In the above scheme, the driver behavior data after monitoring is tracked in real time. Real-time effect tracking is a key link, which involves continuous monitoring of the implementation of service strategies to ensure that they can effectively achieve the predetermined goals, thereby providing passengers with a safer, more convenient and efficient travel experience. Subsequently, based on the feedback mechanism and the driver behavior data after tracking, the service strategy is continuously adjusted and optimized, which can improve the driver service level, enhance the trust and loyalty of passengers to the platform, and gain market advantages.

[0156] Embodiment 2

[0157] See also Figure 3 As shown, Figure 3 : is a structural diagram of a real-time user behavior monitoring system based on online car-hailing provided by the present invention; this embodiment provides a real-time user behavior monitoring system based on online car-hailing, including:

[0158] A real-time data acquisition module 100 is used to acquire the driver's location information, driving behavior data and service record data in real time;

[0159] The model definition module 200 is connected to the real-time data acquisition module 100, and is used to receive the driver's location information, driving behavior data, and service record data, process the driver's location information, driving behavior data, and service record data, and construct a location information model, a driving behavior data model, and a service record model;

[0160] The data processing and analysis module 300 is connected to the model definition module 200, and is used to receive the location information model, the driving behavior data model, and the service record model, and use the data processing algorithm to process the location information model, the driving behavior data model, and the service record model to obtain processed driver behavior data, and use the machine learning algorithm to analyze the processed driver behavior data to obtain analyzed driver behavior data, which includes: using a classification algorithm to analyze the processed driver behavior data, and performing sentiment analysis on the text information of the passenger; or using a clustering algorithm to group the processed driver behavior data and identify different behavior patterns;

[0161] The data monitoring module 400 is connected to the data processing and analysis module 300, and is used to receive the analyzed driver behavior data, monitor the analyzed driver behavior data in real time, and obtain the monitored driver behavior data.

[0162] Although some specific embodiments of the present invention have been described in detail by way of example, it will be appreciated by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It will be appreciated by those skilled in the art that the above embodiments may be modified without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. A real-time user behavior monitoring method based on online car-hailing, characterized in that: include: Obtain driver location information, driving behavior data, and service record data in real time; Processing the driver's location information, the driving behavior data, and the service record data to construct a location information model, a driving behavior data model, and a service record model; Using a data processing algorithm to process the location information model, the driving behavior data model, and the service record model to obtain processed driver behavior data; Analyzing the processed driver behavior data using a machine learning algorithm to obtain analyzed driver behavior data; It includes: using a classification algorithm to analyze the processed driver behavior data and performing sentiment analysis on the passenger's text information; or using a clustering algorithm to group the processed driver behavior data and identify different behavior patterns; The analyzed driver behavior data is monitored in real time to obtain monitored driver behavior data.

2. The real-time user behavior monitoring method based on online car-hailing according to claim 1 is characterized in that: The method of grouping the processed driver behavior data using a clustering algorithm and identifying different behavior patterns also includes: According to the different behavior patterns and changes over time, the preferences of the analyzed driver behavior data are identified.

3. The real-time user behavior monitoring method based on online car-hailing according to claim 1 is characterized in that: The method of using a data processing algorithm to process the location information model, the driving behavior data model and the service record model to obtain the processed driver behavior data includes: A time window algorithm is used to perform aggregate analysis on the location information model, the driving behavior data model and the service record model to obtain statistical information within a preset time interval; and a programming algorithm is used to perform parallel processing on the statistical information within the preset time interval.

4. The real-time user behavior monitoring method based on online car-hailing according to claim 3 is characterized in that: The method of using a time window algorithm to aggregate and analyze the location information model, the driving behavior data model, and the service record model to obtain statistical information within a preset time interval includes: Indexes are constructed for different fields in the location information model, the driving behavior data model, and the service record model.

5. The real-time user behavior monitoring method based on online car-hailing according to claim 1 is characterized in that: After the processing of the driver's location information, the driving behavior data and the service record data to construct the location information model, the driving behavior data model and the service record model, the further step includes: encrypting the processed driver behavior data to obtain encrypted driver behavior data; Verification is performed based on the encrypted driver behavior data to obtain verified driver behavior data; Permission management is performed based on the verified driver behavior data to obtain the managed driver behavior data.

6. The real-time user behavior monitoring method based on online car-hailing according to claim 1 is characterized in that: The driver behavior data after analysis is monitored in real time, and the driver behavior data after monitoring includes: Establishing a real-time data dashboard based on the analyzed driver behavior data, and displaying real-time statistical data of driver behavior through the real-time data dashboard, wherein the real-time statistical data of driver behavior includes time data and violation behavior data; When the video surveillance system detects abnormal or illegal behavior of the driver, the online ride-hailing platform sends alarm data to the manager.

7. The real-time user behavior monitoring method based on online car-hailing according to claim 1 is characterized in that: Also includes: Use machine learning algorithms to predict historical driver behavior data and obtain prediction results; Based on the prediction results and real-time driver behavior data, the driver is dynamically adjusted to the prediction area.

8. The real-time user behavior monitoring method based on online car-hailing according to claim 1 is characterized in that: The analyzing driver behavior data is monitored in real time to obtain the monitored driver behavior data, and further includes: Tracking the monitored driver behavior data in real time to obtain tracked driver behavior data; Optimized driver behavior data is obtained based on the tracked driver behavior data and feedback mechanism.

9. The real-time user behavior monitoring method based on online car-hailing according to claim 1 is characterized in that: The driver's location information, the driving behavior data and the service record data are processed to construct a location information model, a driving behavior data model and a service record model, including: Using a message queue to receive the driver's location information, the driving behavior data, and the service record data; Cleaning the driver's location information, the driving behavior data, and the service record data to obtain cleaned driver's location information, the driving behavior data, and the service record data; Converting the cleaned driver's location information, the driving behavior data, and the service record data into a unified format by using data standardization and formatting processing; The driver's location information, the driving behavior data, and the service record data converted into a unified format are serialized into a structured format, and the location information model, the driving behavior data model, and the service record model are constructed.

10. A real-time user behavior monitoring system based on online car-hailing, characterized in that: include: Real-time data acquisition module, used to obtain the driver's location information, driving behavior data and service record data in real time; A model definition module, connected to the real-time data acquisition module, for receiving the driver's location information, driving behavior data and service record data, processing the driver's location information, driving behavior data and service record data, and constructing a location information model, a driving behavior data model and a service record model; a data processing and analysis module, connected to the model definition module, for receiving a location information model, a driving behavior data model and a service record model, processing the location information model, the driving behavior data model and the service record model using a data processing algorithm to obtain processed driver behavior data, and analyzing the processed driver behavior data using a machine learning algorithm to obtain analyzed driver behavior data, which includes: analyzing the processed driver behavior data using a classification algorithm and performing sentiment analysis on passenger text information; or grouping the processed driver behavior data using a clustering algorithm to identify different behavior patterns; The data monitoring module is connected to the data processing and analysis module, and is used to receive the analyzed driver behavior data, monitor the analyzed driver behavior data in real time, and obtain the monitored driver behavior data.

Images