Transaction intention prediction method and device based on vehicle positioning, equipment and medium

By using electronic fences and intention prediction models in used car transactions, the problem of inefficient clue recognition caused by information asymmetry is solved, the recognition and conversion efficiency of used car transaction clues is improved, and more efficient data utilization and user experience are achieved .

CN118886531APending Publication Date: 2024-11-01PING AN INT FINANCIAL LEASING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410890668.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-03
Publication Date
2024-11-01

AI Technical Summary

Technical Problem

During the second-hand car transaction, due to the asymmetry of information between the buyer and the seller, the clue identification efficiency is low, the cost is high, the accuracy is poor, the timeliness is poor, the conversion rate is low, the user experience is poor, and the transaction intention cannot be effectively identified.

Method used

By setting up electronic fences at preset monitoring points, collecting vehicle identification information, obtaining vehicle data, and using the intention prediction model to perform classification prediction, the transaction intention of the target user is determined, and the efficiency of clue identification and conversion is improved.

Benefits of technology

Reduce data processing volume, improve the efficiency and accuracy of vehicle identification and data acquisition, open up a new clue entrance, improve the accuracy and timeliness of transaction clue identification, realize low-cost secondary utilization, and improve transactions efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118886531A_ABST
    Figure CN118886531A_ABST
Patent Text Reader

Abstract

The invention provides a transaction intention prediction method and device based on vehicle positioning, equipment and a medium, and relates to the field of data processing. According to the method, the electronic fence is arranged at the preset monitoring point position, the vehicle identification information is collected, and then the vehicle data of the target vehicle is obtained according to the vehicle identification information, so that the data processing amount is reduced, the vehicle identification efficiency and the vehicle data obtaining efficiency and accuracy are improved, secondary utilization of the existing data at low cost is realized, and the user experience is improved. And a new clue inlet is opened. Through the intention prediction model, the vehicle data is classified and predicted, and the transaction intention of the target user for the target vehicle is judged, so that the identification accuracy and timeliness of the transaction clues are improved, and the identification efficiency and conversion efficiency of the vehicle transaction clues are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a transaction intention prediction method, device, equipment and medium based on vehicle positioning. Background Art

[0003] In the process of used car transactions, due to the information asymmetry between buyers and sellers, a lot of manpower and material resources are consumed in terms of efficiency and conversion, and the results are not ideal, which is particularly evident in the following aspects: First, the cost of obtaining leads is getting higher and higher. The current mainstream traffic platforms have basically completed the monopoly on leads, and the pricing is increasing year by year. Second, the clues are not accurate, and various types of information are isolated and scattered. There is no effective verification and cleaning, resulting in situations where people and cars cannot be found, and the information of people and cars is inconsistent. Third, the timeliness of the clues is poor. After being circulated by many parties, the clues are obtained with a time lag, and the clues are no longer valid. Fourth, the conversion rate of leads is low, and a large number of invalid leads waste storage costs and outbound call manpower. Fifth, the user experience is poor. Due to limited targeted information, the language used when customer service communicates with users is too procedural and not personalized, and professional marketing cannot be carried out, which greatly reduces customer intentions.

[0004] Therefore, how to solve the current low efficiency of vehicle transaction clue identification has become a technical problem that needs to be solved urgently. Summary of the invention

[0005] The present application provides a transaction intention prediction method, device, equipment and medium based on vehicle positioning, aiming to improve the efficiency of vehicle transaction clue identification.

[0006] In a first aspect, the present application provides a transaction intention prediction method based on vehicle positioning, the transaction intention prediction method based on vehicle positioning comprising the following steps:

[0007] Collect vehicle identification information of target vehicles based on the electronic fence at preset monitoring points;

[0008] Based on the vehicle identification information, obtaining vehicle data of the target vehicle;

[0009] Based on the intention prediction model, the vehicle data is classified and predicted to determine the target user's predicted transaction intention for the target vehicle.

[0010] In a second aspect, the present application further provides a transaction intention prediction device based on vehicle positioning, the transaction intention prediction device based on vehicle positioning comprising:

[0011] An information collection module is used to collect vehicle identification information of a target vehicle based on an electronic fence at a preset monitoring point;

[0012] A data acquisition module, used to acquire vehicle data of the target vehicle based on the vehicle identification information;

[0013] The intention prediction module is used to perform classification prediction processing on the vehicle data based on the intention prediction model to determine the target user's predicted transaction intention for the target vehicle.

[0014] In a third aspect, the present application also provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the steps of the transaction intention prediction method based on vehicle positioning as described above are implemented.

[0015] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the transaction intention prediction method based on vehicle positioning as described above are implemented.

[0016] The present application provides a transaction intention prediction method, device, equipment and medium based on vehicle positioning. The method of the present application includes an electronic fence based on a preset monitoring point to collect vehicle identification information of the target vehicle; based on the vehicle identification information, obtain the vehicle data of the target vehicle; based on the intention prediction model, classify and predict the vehicle data to determine the target user's predicted transaction intention for the target vehicle. Through the above method, the present application sets an electronic fence at a preset monitoring point to collect vehicle identification information, and then obtains the vehicle data of the target vehicle based on the vehicle identification information, thereby reducing the amount of data processing, improving the vehicle identification efficiency and the acquisition efficiency and accuracy of vehicle data, and realizing low-cost secondary utilization of existing data, opening up a new clue entrance. Through the intention prediction model, the vehicle data is classified and predicted to determine the target user's transaction intention for the target vehicle, thereby improving the recognition accuracy and timeliness of transaction clues, and then improving the recognition efficiency and conversion efficiency of vehicle transaction clues. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flowchart of a first embodiment of a transaction intention prediction method based on vehicle positioning provided by the present application;

[0019] Figure 2A schematic diagram of network communication connection of a vehicle transaction intention clue identification system provided in an embodiment of the present application;

[0020] Figure 3 A flow chart of a second embodiment of a transaction intention prediction method based on vehicle positioning provided by the present application;

[0021] Figure 4 A schematic flow chart of a third embodiment of a transaction intention prediction method based on vehicle positioning provided by the present application;

[0022] Figure 5 It is a structural schematic diagram of a first embodiment of a transaction intention prediction device based on vehicle positioning provided by the present application;

[0023] Figure 6 It is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application.

[0024] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0026] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0027] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0028] Please refer to Figure 1 , Figure 1 A flowchart of a first embodiment of a transaction intention prediction method based on vehicle positioning provided in this application.

[0029] like Figure 1 As shown, the transaction intention prediction method based on vehicle positioning includes steps S101 to S103.

[0030] S101, collecting vehicle identification information of a target vehicle based on an electronic fence at a preset monitoring point;

[0031] In one embodiment, the address information of vehicle trading related places such as second-hand car trading markets, evaluation outlets, 4S stores, etc. is collected, the electronic fence data is maintained by point, and the shape boundary, size, alarm threshold, etc. of the electronic fence are configured. The data collected by the electronic fence can be uniformly controlled and collected through the centralized control center. In addition, the centralized control center can support the functions of adding, viewing, modifying, and deleting the electronic fence.

[0032] Among them, the electronic fence is the most advanced perimeter anti-theft alarm system, which consists of an electronic fence host and a front-end detection fence. The electronic fence host generates and receives high-voltage pulse signals, and can generate alarm signals when the front-end detection fence is in a state of touching the network, short circuit, or open circuit, and send intrusion signals to the security alarm center; the front-end detection fence is a tangible perimeter composed of components such as poles and metal wires. Multi-level networking can be achieved through the control keyboard or control software.

[0033] In one embodiment, when the target vehicle enters or leaves the monitoring range of the electronic fence, if the alarm triggering condition is met, an alarm and notification are triggered, the alarm time is recorded, and the downstream terminal is notified.

[0034] In one embodiment, the alarm trigger condition may be the length of time that the target vehicle stays within the electronic fence, that is, the duration from the target vehicle entering the electronic fence to leaving the electronic fence. A time threshold may be set, such as 30 minutes, that is, if the target vehicle stays for equal to or greater than 30 minutes, an alarm and notification are triggered.

[0035] In one embodiment, the alarm triggering condition may also be that as long as the target vehicle is detected to enter or leave the range of the electronic fence, the alarm and notification are triggered. The alarm may also be triggered in other ways, which are not specifically limited in the embodiments of the present application.

[0036] In one embodiment, when the electronic fence triggers an alarm, the vehicle identification information of the target vehicle, such as vehicle type, vehicle brand, vehicle appearance, license plate number, driver, etc., is collected through an image acquisition device (such as a camera, etc.).

[0037] S102, acquiring vehicle data of the target vehicle based on the vehicle identification information;

[0038] In one embodiment, the vehicle data includes vehicle positioning data and vehicle user information, and the vehicle positioning data includes point positioning information and driving trajectory data.

[0039] In one embodiment, positioning equipment such as GPS can be installed on the vehicle, including wired and wireless GPS tools, and debugged to ensure that satellite signals can be received. By connecting with the GPS provider, the driving trajectory data of the target vehicle can be obtained.

[0040] In one embodiment, a GPS device on a vehicle is used to collect driving trajectory data, including the latitude and longitude of the vehicle, a timestamp, etc.

[0041] In one embodiment, the point location information of the target vehicle is determined by collecting the point information corresponding to the electronic fence of the vehicle identification information of the target vehicle, recording the time the vehicle stays at each point, and calculating it through the alarm of entering and leaving the electronic fence.

[0042] In one embodiment, the vehicle user information of the target vehicle, including the owner's identity, contact information, frame number, insurance information, etc., is obtained through vehicle type information such as license plate number through channels such as 4S stores and insurance companies.

[0043] S103: Based on the intention prediction model, classify and predict the vehicle data to determine the target user's predicted transaction intention for the target vehicle.

[0044] In one embodiment, if Figure 2 As shown, Figure 2 A schematic diagram of the network communication connection of a vehicle transaction intention clue identification system provided in the embodiment of the present application. Through the series connection and data intercommunication between systems, a real-time alarm can be issued through the electronic fence when the target vehicle enters the parking lot, and vehicle identification information can be collected, including the vehicle VIN code, license plate number, entry time, etc.

[0045] In one embodiment, vehicle data, namely vehicle positioning data and vehicle user information, such as user address, vehicle historical driving trajectory, mileage, lease period, evaluation record, etc., are obtained based on vehicle identification information, and a comprehensive score is given based on key factors related to transaction intention. According to the scoring results, clues are classified and predicted, invalid clues are eliminated, and valid clues are classified by level, city, etc., and pushed to intelligent customer service for outbound calls to confirm the final intention.

[0046] Furthermore, based on the data processing algorithm, the vehicle data is preprocessed to obtain standard data; based on the feature extraction algorithm, feature data in the standard data is extracted to obtain transaction intention features; based on the intention prediction model, the transaction intention features are classified and predicted to determine the target user's predicted transaction intention for the target vehicle.

[0047] In one embodiment, a data processing algorithm is used to clean, standardize, and format the collected vehicle data to eliminate noise and inconsistency and obtain standard data. A feature extraction algorithm is applied to extract features related to transaction intention from the standard data, such as vehicle usage frequency, driving distance, user behavior pattern, vehicle age, maintenance information, etc. The data can be comprehensively scored based on key factors related to transaction intention, such as vehicle condition, user satisfaction, market trends, etc. The scoring system can include weight allocation to reflect the impact of different factors on transaction intention. Based on the scoring results, the leads are classified and predicted to distinguish high-intention, medium-intention, and low-intention users to eliminate invalid leads and retain potential transaction leads.

[0048] This embodiment firstly reuses the existing data at a low cost, opening up new clue entrances; secondly, it uses technologies such as electronic fences, large models, and AI smart outbound calls to minimize labor costs; finally, after connecting various systems, it improves the accuracy and timeliness of data, effectively improves clue conversion, and brings into play new data value.

[0049] This embodiment provides a transaction intention prediction method based on vehicle positioning. The method of this application sets up an electronic fence at a preset monitoring point to collect vehicle identification information, and then obtains the vehicle data of the target vehicle based on the vehicle identification information, thereby reducing the amount of data processing, improving the efficiency of vehicle identification and the efficiency and accuracy of vehicle data acquisition, and realizing low-cost secondary utilization of existing data, opening up a new clue entry. Through the intention prediction model, the vehicle data is classified and predicted, and the target user's transaction intention for the target vehicle is judged, thereby improving the recognition accuracy and timeliness of transaction clues, and then improving the recognition efficiency and conversion efficiency of vehicle transaction clues.

[0050] Please refer to Figure 3 , Figure 3 This is a flow chart of the second embodiment of the transaction intention prediction method based on vehicle positioning provided in this application.

[0051] In this embodiment, Figure 3 As shown, based on the above Figure 1 In the illustrated embodiment, after step S103, the following steps are further included:

[0052] S201, when the predicted transaction intention is a positive transaction intention, based on a preset intelligent dialogue instruction, conducting a dialogue with the target user to generate user dialogue data;

[0053] In one embodiment, positive transaction intention means that the predicted result is that the target user has a high transaction intention for the target vehicle, such as reaching a medium transaction intention or a high transaction intention.

[0054] In one embodiment, the transaction intention can also be represented by data, such as presetting a transaction intention probability threshold, and evaluating the probability of the target user's predicted transaction intention through the intention prediction model. When the predicted probability is greater than or equal to the transaction intention probability threshold, it means that the predicted transaction intention is a positive transaction intention, that is, the target user intends to conduct a vehicle transaction, which is an effective clue. On the contrary, when the predicted probability is less than the transaction intention probability threshold, it means that the predicted transaction intention is a negative transaction intention, that is, the target user is currently unlikely to conduct a vehicle transaction, and the transaction success rate is very low, which is an invalid clue.

[0055] In one embodiment, according to the prediction results of the intention prediction model, the effective leads are sorted by city, region, predicted transaction success rate, etc., to generate a lead list. Through intelligent dialogue instructions, AI intelligent outbound calls are set up with scheduled tasks and dialogues every day according to the received lead list, and the results are confirmed according to the customer's answer, and the user dialogue data such as the call time, whether the call is connected, the call duration, and the call content are recorded.

[0056] S202, obtaining the target user's real transaction intention for the target vehicle based on the user conversation data;

[0057] In one embodiment, the real transaction intention of the target user is identified based on the call content in the user conversation data, marked and fed back to the intention prediction model to perform regression verification on the prediction results of the intention prediction model.

[0058] In one embodiment, key sentences and keywords related to transaction intentions in user conversation data can be extracted through text recognition, such as key words such as "considering changing cars", "looking at cars", "not considering it yet", or context relevance recognition can be performed based on words such as "yes", "good", "no", and "no", so as to extract key sentences and analyze the real transaction intentions of the target users.

[0059] S203: Based on the comparison result between the actual transaction intention and the predicted transaction intention, regression verification is performed on the intention prediction model to evaluate the prediction accuracy of the intention prediction model.

[0060] In one embodiment, user conversation data and real transaction intentions can be fed back into the intention prediction model, and the real transaction intentions and predicted transaction intentions can be compared to evaluate the prediction accuracy of the model. According to the regression verification results, the model parameters are adjusted to optimize the intention prediction model and improve the prediction accuracy of the model.

[0061] Understandably, the purpose of regression validation is to ensure that the predicted results of the model in actual applications are as close as possible to the actual results, thereby improving the credibility and practicality of the model. Through this process, the shortcomings of the model can be identified and measures can be taken to improve it.

[0062] Furthermore, when the actual transaction intention is inconsistent with the predicted transaction intention, the intention prediction model is iteratively trained based on the actual transaction intention and the vehicle data to obtain an optimized intention prediction model.

[0063] In one embodiment, if the target user's actual transaction intention is inconsistent with the predicted transaction intention of the intention prediction model, it means that the prediction result of the intention prediction model is inaccurate. Therefore, the user conversation data and the actual transaction intention can be fed back to the intention prediction model, and the intention prediction model can be adjusted and optimized by combining the previously collected vehicle data.

[0064] In one embodiment, the intention prediction model may be optimized and trained based on the collected vehicle data, model prediction results and other data at regular intervals to form a closed loop of continuous improvement, so that the intention prediction model always maintains high-precision prediction capabilities.

[0065] This embodiment conducts intelligent dialogue with clues with high transaction intentions, identifies the target user's real transaction intention for the target vehicle through intelligent dialogue with the target user, and then verifies the prediction accuracy of the intention prediction model. And by comparing the real transaction intention with the predicted transaction intention, the intention prediction model is regressively verified, and the intention prediction model is optimized and adjusted according to the verification results to improve the credibility and prediction accuracy of the intention prediction model.

[0066] Please refer to Figure 4 , Figure 4 This is a flow chart of the third embodiment of the transaction intention prediction method based on vehicle positioning provided in this application.

[0067] In this embodiment, Figure 4 As shown, based on the above Figure 1 In the illustrated embodiment, before step S103, the following steps are further included:

[0068] S301, obtaining historical stock data;

[0069] In one embodiment, a GPS device on the vehicle is used to collect vehicle driving trajectory data, including the vehicle's latitude and longitude, timestamp, etc. The vehicle's stay time at each point is calculated through the vehicle's entry and exit electronic fence alarm mechanism.

[0070] In one embodiment, historical vehicle information of a large number of vehicles may be collected as historical stock data.

[0071] S302: preprocess the historical stock data based on a data preprocessing algorithm to obtain model training data;

[0072] In one embodiment, data preprocessing includes, but is not limited to, cleaning data, removing noise, outliers, and duplicate data, etc., to ensure data quality and accuracy. Necessary conversion and standardization are performed on the data to facilitate subsequent model training.

[0073] In one embodiment, features related to transaction intention are extracted from standardized data, such as the customer's lease start time, remaining lease period, vehicle mileage, vehicle brand value retention rate, customer gender, spatiotemporal features of driving trajectory (such as time, date, etc.) and stay time features.

[0074] In one embodiment, different features may be combined or converted to obtain richer information.

[0075] S303: Based on the model training data, perform iterative optimization training on the initial prediction model to obtain the intention prediction model.

[0076] In one embodiment, a deep learning architecture suitable for processing sequence data and spatiotemporal data can be selected, a model structure that can integrate multiple features and predict transaction intentions can be designed, and multimodal learning methods can be considered to integrate data from different sources.

[0077] In one embodiment, for the constructed model structure, a reward function that can reflect the transaction intention can also be defined. For example, when the vehicle stays at a certain point for a long time and the area is related to the transaction intention, a higher reward is given; in other cases, a lower reward or penalty is given.

[0078] Among them, the reward function is a core concept in reinforcement learning (RL), which is used to quantify the immediate return obtained by the agent after performing an action. The reward function R(s,a) defines the immediate reward that can be obtained by performing action a in a given state s. The reward function can clearly guide the intention prediction model to reach the target state or perform the expected behavior. The reward function provides a clear goal or direction for the intention prediction model, guiding the intention prediction model to take actions in the environment to achieve the desired results.

[0079] Furthermore, based on the intention results corresponding to the historical stock data, the model training data is labeled to obtain labeled data; based on the labeled data, the initial prediction model is trained to obtain the intention prediction model; based on the test set data, the capability of the intention prediction model is evaluated, and the intention prediction model is iteratively optimized according to the evaluation results until the prediction accuracy of the intention prediction model is greater than or equal to a preset accuracy threshold.

[0080] In one embodiment, the historical stock data is annotated with intention results and used for model training. The annotated data includes the vehicle's GPS data, stay time, and corresponding transaction intention labels. During the training process, the prediction performance and generalization ability of the intention prediction model are improved by adjusting model parameters, optimizing algorithms, learning rates, and other strategies.

[0081] In one embodiment, the trained model is evaluated using a test set, including indicators such as accuracy, recall, F1 score, etc. The model is tuned based on the evaluation results, including adjusting the model structure, increasing the number or types of features, improving the reward function, etc.

[0082] In one embodiment, model evaluation is a process of evaluating model performance through test indicators. After the test preparation is completed, the trained model is tested using the selected test samples and the corresponding evaluation indicators are calculated.

[0083] Among them, the accuracy rate is an indicator to measure the accuracy of model prediction, and its calculation method is the ratio of correctly predicted samples to the total number of samples. The recall rate is an indicator to measure the completeness of the model, and its calculation method is the ratio of correctly predicted positive samples to all positive samples. The F1 value is the harmonic mean of the accuracy rate and the recall rate, which is used to comprehensively evaluate the performance of the model.

[0084] See also Figure 5 , Figure 5 It is a structural schematic diagram of a first embodiment of a transaction intention prediction device based on vehicle positioning provided in the present application, and the transaction intention prediction device based on vehicle positioning is used to execute the aforementioned transaction intention prediction method based on vehicle positioning.

[0085] like Figure 5 As shown, the transaction intention prediction device 400 based on vehicle positioning includes: an information collection module 401, a data acquisition module 402 and an intention prediction module 403.

[0086] The information collection module 401 is used to collect vehicle identification information of the target vehicle based on the electronic fence of the preset monitoring point;

[0087] A data acquisition module 402 is used to acquire vehicle data of the target vehicle based on the vehicle identification information;

[0088] The intention prediction module 403 is used to perform classification prediction processing on the vehicle data based on the intention prediction model to determine the target user's predicted transaction intention for the target vehicle.

[0089] In one embodiment, the intention prediction module 403 includes:

[0090] A standard data acquisition unit, used for preprocessing the vehicle data based on a data processing algorithm to obtain standard data;

[0091] A feature extraction unit, used to extract feature data from the standard data based on a feature extraction algorithm to obtain transaction intention features;

[0092] The intention prediction unit is used to classify and predict the transaction intention features based on the intention prediction model, and determine the predicted transaction intention of the target user for the target vehicle.

[0093] In one embodiment, the transaction intention prediction device 400 based on vehicle positioning further includes a result verification module, including:

[0094] An intelligent dialogue unit, configured to conduct a dialogue with the target user based on a preset intelligent dialogue instruction to generate user dialogue data when the predicted transaction intention is a positive transaction intention;

[0095] A real transaction intention identification unit, used for obtaining the real transaction intention of the target user for the target vehicle based on the user conversation data;

[0096] The model regression verification unit is used to perform regression verification on the intention prediction model based on the comparison result of the actual transaction intention and the predicted transaction intention, so as to evaluate the prediction accuracy of the intention prediction model.

[0097] In one embodiment, the model regression verification unit includes:

[0098] The model iteration optimization subunit is used to iteratively train the intention prediction model based on the actual transaction intention and the vehicle data to obtain an optimized intention prediction model when the actual transaction intention is inconsistent with the predicted transaction intention.

[0099] In one embodiment, the vehicle positioning-based transaction intention prediction device 400 further includes a model training module, including:

[0100] A historical stock data acquisition unit, used to acquire historical stock data;

[0101] A data preprocessing unit, used to preprocess the historical stock data based on a data preprocessing algorithm to obtain model training data;

[0102] The model training unit is used to iteratively optimize the initial prediction model based on the model training data to obtain the intention prediction model.

[0103] In one embodiment, the model training unit includes:

[0104] A data labeling subunit, used to label the model training data based on the intended results corresponding to the historical stock data to obtain labeled data;

[0105] A model training subunit, used to train the initial prediction model based on the labeled data to obtain the intention prediction model;

[0106] The model iteration optimization subunit is used to evaluate the capability of the intention prediction model based on the test set data, and iteratively optimize the intention prediction model according to the evaluation results until the prediction accuracy of the intention prediction model is greater than or equal to a preset accuracy threshold.

[0107] In one embodiment, the vehicle data includes vehicle positioning data and vehicle user information, and the vehicle positioning data includes point positioning information and driving trajectory data.

[0108] It should be noted that technical personnel in the relevant field can clearly understand that, for the convenience and conciseness of description, the specific working process of the above-described device and each module can refer to the corresponding process in the aforementioned transaction intention prediction method based on vehicle positioning embodiment, and will not be repeated here.

[0109] The apparatus provided in the above embodiment may be implemented in the form of a computer program. The computer program may be Figure 6 Runs on the computer device shown.

[0110] See also Figure 6 , Figure 6 1 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device may be a server.

[0111] See also Figure 6 The computer device includes a processor, a memory and a network interface connected through a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0112] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any transaction intention prediction method based on vehicle positioning.

[0113] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0114] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any transaction intention prediction method based on vehicle positioning.

[0115] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0116] It should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0117] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0118] Collect vehicle identification information of target vehicles based on the electronic fence at preset monitoring points;

[0119] Based on the vehicle identification information, obtaining vehicle data of the target vehicle;

[0120] Based on the intention prediction model, the vehicle data is classified and predicted to determine the target user's predicted transaction intention for the target vehicle.

[0121] In one embodiment, when the processor implements the intention prediction model to classify and predict the vehicle data to determine the target user's predicted transaction intention for the target vehicle, it is used to implement:

[0122] Based on a data processing algorithm, preprocessing the vehicle data to obtain standard data;

[0123] Based on the feature extraction algorithm, feature data is extracted from the standard data to obtain transaction intention features;

[0124] Based on the intention prediction model, the transaction intention features are classified and predicted to determine the target user's predicted transaction intention for the target vehicle.

[0125] In one embodiment, after implementing the intention prediction model, performing classification prediction processing on the vehicle data, and determining the target user's predicted transaction intention for the target vehicle, the processor is further used to implement:

[0126] When the predicted transaction intention is a positive transaction intention, based on a preset intelligent dialogue instruction, a dialogue is conducted with the target user to generate user dialogue data;

[0127] Based on the user conversation data, obtaining the target user's real transaction intention for the target vehicle;

[0128] Based on the comparison results between the actual transaction intention and the predicted transaction intention, the intention prediction model is subjected to regression verification to evaluate the prediction accuracy of the intention prediction model.

[0129] In one embodiment, when the processor implements the regression verification of the intention prediction model based on the comparison result of the actual transaction intention and the predicted transaction intention to evaluate the prediction accuracy of the intention prediction model, it is used to implement:

[0130] When the actual transaction intention is inconsistent with the predicted transaction intention, the intention prediction model is iteratively trained based on the actual transaction intention and the vehicle data to obtain an optimized intention prediction model.

[0131] In one embodiment, before implementing the intention prediction model to classify and predict the vehicle data to determine the target user's predicted transaction intention for the target vehicle, the processor is further configured to implement:

[0132] Get historical inventory data;

[0133] Based on the data preprocessing algorithm, the historical stock data is preprocessed to obtain model training data;

[0134] Based on the model training data, the initial prediction model is iteratively optimized and trained to obtain the intention prediction model.

[0135] In one embodiment, when the processor implements the iterative optimization training of the initial prediction model based on the model training data to obtain the intention prediction model, it is used to implement:

[0136] Based on the intended results corresponding to the historical stock data, the model training data is labeled to obtain labeled data;

[0137] Based on the labeled data, training the initial prediction model to obtain the intention prediction model;

[0138] Based on the test set data, the capability of the intention prediction model is evaluated, and the intention prediction model is iteratively optimized according to the evaluation results until the prediction accuracy of the intention prediction model is greater than or equal to a preset accuracy threshold.

[0139] In one embodiment, the vehicle data includes vehicle positioning data and vehicle user information, and the vehicle positioning data includes point positioning information and driving trajectory data.

[0140] A computer-readable storage medium is also provided in an embodiment of the present application, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and the processor executes the program instructions to implement any one of the transaction intention prediction methods based on vehicle positioning provided in the embodiments of the present application.

[0141] The computer-readable storage medium may be an internal storage unit of the computer device described in the above embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (SmartMediaCard, SMC), a secure digital (SecureDigital, SD) card, a flash card (FlashCard), etc. equipped on the computer device.

[0142] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A transaction intention prediction method based on vehicle positioning, characterized in that: The method comprises: Collect vehicle identification information of target vehicles based on the electronic fence at preset monitoring points; Based on the vehicle identification information, obtaining vehicle data of the target vehicle; Based on the intention prediction model, the vehicle data is classified and predicted to determine the target user's predicted transaction intention for the target vehicle.

2. The transaction intention prediction method based on vehicle positioning according to claim 1 is characterized in that: The method of performing classification prediction processing on the vehicle data based on the intention prediction model to determine the predicted transaction intention of the target user for the target vehicle includes: Based on a data processing algorithm, preprocessing the vehicle data to obtain standard data; Based on the feature extraction algorithm, feature data is extracted from the standard data to obtain transaction intention features; Based on the intention prediction model, the transaction intention features are classified and predicted to determine the target user's predicted transaction intention for the target vehicle.

3. The transaction intention prediction method based on vehicle positioning according to claim 1 is characterized in that: After the vehicle data is classified and predicted based on the intention prediction model to determine the target user's predicted transaction intention for the target vehicle, the method further includes: When the predicted transaction intention is a positive transaction intention, based on a preset intelligent dialogue instruction, a dialogue is conducted with the target user to generate user dialogue data; Based on the user conversation data, obtaining the target user's real transaction intention for the target vehicle; Based on the comparison results between the actual transaction intention and the predicted transaction intention, the intention prediction model is subjected to regression verification to evaluate the prediction accuracy of the intention prediction model.

4. The transaction intention prediction method based on vehicle positioning according to claim 3 is characterized in that: The step of performing regression verification on the intention prediction model based on the comparison result between the actual transaction intention and the predicted transaction intention to evaluate the prediction accuracy of the intention prediction model includes: When the actual transaction intention is inconsistent with the predicted transaction intention, the intention prediction model is iteratively trained based on the actual transaction intention and the vehicle data to obtain an optimized intention prediction model.

5. The transaction intention prediction method based on vehicle positioning according to claim 1 is characterized in that: Before the vehicle data is classified and predicted based on the intention prediction model to determine the target user's predicted transaction intention for the target vehicle, the method further includes: Get historical inventory data; Based on the data preprocessing algorithm, the historical stock data is preprocessed to obtain model training data; Based on the model training data, the initial prediction model is iteratively optimized and trained to obtain the intention prediction model.

6. The transaction intention prediction method based on vehicle positioning according to claim 5 is characterized in that: The iterative optimization training of the initial prediction model based on the model training data to obtain the intention prediction model includes: Based on the intended results corresponding to the historical stock data, the model training data is labeled to obtain labeled data; Based on the labeled data, training the initial prediction model to obtain the intention prediction model; Based on the test set data, the capability of the intention prediction model is evaluated, and the intention prediction model is iteratively optimized according to the evaluation results until the prediction accuracy of the intention prediction model is greater than or equal to a preset accuracy threshold.

7. The transaction intention prediction method based on vehicle positioning according to claim 1 is characterized in that: The vehicle data includes vehicle positioning data and vehicle user information, and the vehicle positioning data includes point positioning information and driving trajectory data.

8. A transaction intention prediction device based on vehicle positioning, characterized in that: The transaction intention prediction device based on vehicle positioning includes: An information collection module is used to collect vehicle identification information of a target vehicle based on an electronic fence at a preset monitoring point; A data acquisition module, used to acquire vehicle data of the target vehicle based on the vehicle identification information; The intention prediction module is used to perform classification prediction processing on the vehicle data based on the intention prediction model to determine the target user's predicted transaction intention for the target vehicle.

9. A computer device, characterized in that: The computer device includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the steps of the transaction intention prediction method based on vehicle positioning as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the transaction intention prediction method based on vehicle positioning as described in any one of claims 1 to 7 are implemented.