Vehicle loss estimation method, device, computer equipment and storage medium

Abstract

The embodiments of the present application belong to the field of financial technology and relate to a vehicle loss estimation method, apparatus, computer equipment, and storage medium. The method comprises the following steps: identifying, classifying, and extracting features from sample data to obtain vehicle image features, vehicle description features, and vehicle information features, and calculating vehicle loss estimation pricing; combining vehicle image features, vehicle description features, and vehicle information features to obtain vehicle loss features, vehicle accident features, vehicle configuration features, and vehicle insurance policy features; inputting vehicle loss features, vehicle accident features, vehicle information features, vehicle insurance policy features, and vehicle loss estimation pricing into a preset learner for training, and performing integrated learning on the trained learner to obtain a vehicle loss estimation model; and inputting repair report data into the vehicle loss estimation model to obtain loss estimation data. The present application can effectively realize automated and efficient accurate estimation of vehicle losses.

Description

Technical Field

[0001] The present application relates to the field of financial technology, specifically to the field of vehicle insurance technology, and in particular to a vehicle loss estimation method, device, computer equipment and storage medium. Background Art

[0002] When a vehicle is in an accident, in order to shorten the claims process and effectively manage the internal claims time limit, it is necessary to estimate the extent of the vehicle's loss and compile the estimated data into a visual loss report to provide effective data basis for the subsequent vehicle claims process.

[0003] The loss estimation methods for vehicles in accidents generally include the following: professional personnel conduct on-site inspections of the vehicle's losses, and estimate the loss based on factors such as the vehicle's brand, model, year, mileage, and condition, combined with factors such as market price and repair costs; the repair shop appraises and repairs the vehicle, and estimates the loss based on factors such as repair labor costs, repair material costs and parts costs, combined with factors such as the vehicle's brand, model, year, mileage, and condition; utilize the vehicle valuation database to automatically calculate the vehicle's market value and loss estimate based on factors such as the vehicle's brand, model, year, mileage, and condition; utilize image recognition technology to analyze and process vehicle photos to extract key vehicle features such as license plate number, vehicle brand, model, color, degree of damage, etc., and estimate the loss based on factors such as market price and repair costs.

[0004] However, all of the aforementioned methods have drawbacks. Vehicle damage estimation is primarily performed by professionals or repair shops, requiring on-site inspections, consuming considerable human resources, and being cumbersome. Using a database to estimate vehicle damage can lead to inaccurate estimates due to mismatches between stored database data and updated data on the actual vehicle. Furthermore, using images to estimate vehicle damage is susceptible to image quality and image recognition accuracy, resulting in unstable estimates. Summary of the Invention

[0005] The purpose of the embodiments of the present application is to provide a vehicle loss estimation method, apparatus, computer equipment and storage medium to solve the problems of low efficiency, inaccuracy and high human resource consumption in vehicle loss estimation.

[0006] In order to solve the above technical problems, the embodiment of the present application provides a vehicle loss estimation method, which adopts the following technical solutions:

[0007] Acquiring sample data, identifying and classifying the sample data to obtain vehicle image data, vehicle description data, and vehicle information data, and calculating vehicle loss estimation pricing based on the vehicle information data and a preset calculation formula;

[0008] Performing feature extraction on the vehicle image data, the vehicle description data, and the vehicle information data respectively to obtain vehicle image features, vehicle description features, and vehicle information features;

[0009] Performing a first feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle loss feature;

[0010] Performing a second feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle accident feature;

[0011] Performing a third feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle configuration feature;

[0012] Performing a fourth feature combination on the vehicle description feature and the vehicle information feature to obtain a vehicle insurance policy feature;

[0013] Inputting the vehicle loss characteristics, the vehicle accident characteristics, the vehicle information characteristics, the vehicle insurance policy characteristics, and the vehicle loss estimation pricing into a preset learner for training, and performing ensemble learning on the trained learner to obtain a vehicle loss estimation model; and

[0014] Obtain repair report data, input the repair report data into the vehicle loss estimation model to perform loss estimation, and obtain loss estimation data.

[0015] Furthermore, the step of identifying and classifying the sample data to obtain vehicle image data, vehicle description data, and vehicle information data specifically includes:

[0016] Performing image recognition on the vehicle sample data, and using the recognized data as the vehicle image data;

[0017] Performing voice recognition and text recognition on the vehicle sample data, and collating the recognized data to obtain the vehicle description data; and

[0018] The data containing the corresponding field in the vehicle sample data is identified according to the preset field identification information, and the identified data is used as the vehicle information data.

[0019] Furthermore, the steps of respectively extracting features from the vehicle image data, the vehicle description data, and the vehicle information data to obtain vehicle image features, vehicle description features, and vehicle information features, and calculating vehicle estimated loss pricing based on the vehicle information data, specifically include:

[0020] Performing data cleaning on the vehicle image data, the vehicle description data, and the vehicle information data to obtain valid image data, valid description data, and valid information data;

[0021] Identifying a vehicle image contained in the valid image data, and performing feature extraction on the vehicle image to obtain the vehicle image feature;

[0022] Identifying the voice information and / or text information of the effective description data, and performing feature extraction on the voice information and / or text information to obtain the vehicle description feature; and

[0023] Identify the information fields contained in the valid information data, perform feature extraction on the information data corresponding to the information fields, and obtain the vehicle information features.

[0024] Furthermore, the vehicle image features include vehicle damage image features, accident scene image features, and vehicle configuration image features; the vehicle description features include vehicle damage condition features, vehicle accident condition features, vehicle configuration condition features, and vehicle insurance policy condition features; the vehicle information features include vehicle damage information features, vehicle accident information features, vehicle configuration information features, and vehicle insurance policy information features; and the step of performing a first feature combination on the vehicle image features, the vehicle description features, and the vehicle information features to obtain vehicle loss features specifically includes the following steps:

[0025] Performing feature intersection on the vehicle damage image feature, the vehicle damage condition feature, and the vehicle damage information feature to obtain the vehicle loss feature;

[0026] The step of performing a second feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle accident feature specifically includes the following steps:

[0027] Performing feature intersection on the accident scene image feature, the vehicle accident condition feature, and the vehicle accident information feature to obtain the vehicle accident feature;

[0028] The step of performing a third feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain the vehicle configuration feature specifically includes the following steps:

[0029] Performing feature intersection on the vehicle configuration image feature, the vehicle configuration status feature, and the vehicle configuration information feature to obtain the vehicle configuration feature; and

[0030] The step of performing a fourth feature combination on the vehicle description feature and the vehicle information feature to obtain the vehicle insurance policy feature specifically includes the following steps:

[0031] Perform feature intersection on the vehicle insurance policy status feature and the vehicle insurance policy information feature to obtain the vehicle insurance policy feature.

[0032] Furthermore, the step of inputting the vehicle loss characteristics, the vehicle accident characteristics, the vehicle information characteristics, the vehicle insurance policy characteristics, and the vehicle loss estimation pricing into a preset learner for training, and performing ensemble learning on the trained learner to obtain a vehicle loss estimation model specifically includes:

[0033] Obtaining feature identifiers of the vehicle loss feature, the vehicle accident feature, the vehicle information feature, and the vehicle insurance policy feature;

[0034] Allocating the vehicle loss feature, the vehicle accident feature, the vehicle information feature, and the vehicle insurance policy feature to corresponding loss feature learners, accident feature learners, information feature learners, and insurance policy feature learners according to the feature identifiers;

[0035] Inputting the vehicle loss estimation pricing into the loss feature learner, the accident feature learner, the information feature learner, and the policy feature learner for training;

[0036] Performing integrated learning on the loss feature learner, the accident feature learner, the information feature learner, and the policy feature learner according to a preset integrated learning rule to obtain an initial loss estimation model; and

[0037] The accuracy of the initial loss estimation model is evaluated, and the initial loss estimation model is adjusted according to the evaluation result to obtain the vehicle loss estimation model.

[0038] Furthermore, the step of evaluating the accuracy of the initial loss estimation model and adjusting the initial loss estimation model according to the evaluation result to obtain the vehicle loss estimation model specifically includes:

[0039] Obtaining a preset model evaluation rule, and evaluating the initial loss estimation model according to the model evaluation rule to obtain model accuracy;

[0040] Comparing the model accuracy with a preset standard value to determine whether the model accuracy meets the standard value;

[0041] If the model accuracy meets the standard value, using the initial loss estimation model as the vehicle loss estimation model; and

[0042] If the model accuracy does not meet the standard value, the loss feature learner, the accident feature learner, the information feature learner, and the policy feature learner are iteratively adjusted and re-integrated and learned to obtain an optimized loss estimation model until the model accuracy of the optimized loss estimation model meets the standard value. After the model accuracy of the optimized loss estimation model meets the standard value, the optimized loss estimation model is used as the vehicle loss estimation model.

[0043] Furthermore, after the step of acquiring the repair report data and inputting the repair report data into the vehicle loss estimation model to perform loss estimation to obtain loss estimation data, the following steps are also included:

[0044] Obtaining preset field identification information, identifying the loss estimation data according to the field identification information, and obtaining data key fields;

[0045] Obtaining a preset report text template, identifying the report text template according to the field identification information, and obtaining text key fields; and

[0046] The loss estimation data is correspondingly filled into the report text template according to the data key fields and the text key fields to generate a model loss estimation report.

[0047] In order to solve the above technical problems, the embodiment of the present application further provides a vehicle loss estimation device, which adopts the following technical solution:

[0048] A data classification module is used to obtain sample data, identify and classify the sample data to obtain vehicle image data, vehicle description data, and vehicle information data, and calculate vehicle loss estimation pricing based on the vehicle information data and a preset calculation formula;

[0049] A feature extraction module is used to extract features from the vehicle image data, the vehicle description data, and the vehicle information data to obtain vehicle image features, vehicle description features, and vehicle information features;

[0050] A first feature combination module is used to perform a first feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle loss feature;

[0051] A second feature combination module is used to perform a second feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle accident feature;

[0052] a third feature combination module, configured to perform a third feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle configuration feature;

[0053] a fourth feature combination module, configured to perform a fourth feature combination on the vehicle description feature and the vehicle information feature to obtain a vehicle insurance policy feature;

[0054] a model training module, configured to input the vehicle loss characteristics, the vehicle accident characteristics, the vehicle information characteristics, the vehicle insurance policy characteristics, and the vehicle loss estimation pricing into a preset learner for training, and perform ensemble learning on the trained learner to obtain a vehicle loss estimation model; and

[0055] The model prediction module is used to obtain repair report data, input the repair report data into the vehicle loss estimation model to perform loss estimation, and obtain loss estimation data.

[0056] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0057] A computer device includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the vehicle loss estimation method as described in any one of the above are implemented.

[0058] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0059] A computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of any of the above vehicle loss estimation methods.

[0060] Compared with the prior art, the embodiments of the present application have the following major beneficial effects: This embodiment identifies and classifies sample data and extracts features from the classified data, thereby obtaining vehicle image features based on accident scene images, vehicle description features based on voice data and text data recorded based on the accident situation, and vehicle information features based on the basic configuration information of the vehicle. By combining the vehicle image features, vehicle description features, and vehicle information features, vehicle loss features representing the vehicle loss situation, vehicle accident features representing the accident situation, vehicle configuration features representing the vehicle configuration situation, and vehicle policy features representing the vehicle insurance policy situation are obtained. By inputting the vehicle loss features, vehicle accident features, vehicle information features, vehicle insurance policy features, and vehicle loss estimation pricing calculated based on the vehicle information data into a preset learner for training, and then performing integrated learning on the trained learner, a vehicle loss estimation model that accurately estimates vehicle losses is obtained. The acquired repair report data is then input into the vehicle loss estimation model to obtain loss estimation data, thereby achieving automated and efficient accurate estimation of vehicle losses. This embodiment can be applied to the field of auto insurance. The above method can accurately estimate the vehicle loss at the time of an accident, thereby providing effective data for subsequent auto insurance claims and vehicle repairs. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the solutions in this application, a brief introduction will be given below to the drawings required for use in the description of the embodiments of this application. Obviously, the drawings described below are some embodiments of this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0062] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0063] Figure 2 A flow chart of an embodiment of a vehicle loss estimation method according to the present application;

[0064] Figure 3 yes Figure 2 A flowchart of a specific implementation of step S10;

[0065] Figure 4 yes Figure 2 A flowchart of a specific implementation of step S20;

[0066] Figure 5 yes Figure 2 A flowchart of a specific implementation of step S70;

[0067] Figure 6 yes Figure 5A flowchart of a specific implementation of step S705;

[0068] Figure 7 is a schematic structural diagram of an embodiment of a vehicle loss estimation device according to the present application;

[0069] Figure 8 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0070] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art to which this application belongs. The terms used in the specification of the application are for the purpose of describing specific embodiments only and are not intended to limit this application. The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0071] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0072] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0073] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. Network 104 is a medium for providing communication links between terminal devices 101, 102, 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0074] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0075] Terminal devices 101, 102, and 103 can be various electronic devices with display screens and support web browsing, including but not limited to smartphones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV), laptop computers, desktop computers, etc.

[0076] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0077] It should be noted that the vehicle loss estimation method provided in the embodiment of the present application is generally executed by a server, and accordingly, the vehicle loss estimation device is generally set in the server.

[0078] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0079] Continue to refer Figure 2 , shows a flow chart of an embodiment of the method for system safety monitoring calculation according to the present application. The vehicle loss estimation method includes the following steps:

[0080] Step S10, acquiring sample data, identifying and classifying the sample data to obtain vehicle image data, vehicle description data, and vehicle information data, and calculating vehicle loss estimation pricing based on the vehicle information data and a preset calculation formula;

[0081] In this embodiment, vehicle image data is provided by on-site vehicle images captured and entered into the system database. This image can be an image file captured by the vehicle owner or insurance agent and entered into the system database, or it can be an image captured from a video file captured by relevant road section surveillance. Vehicle description data is a voice and text file entered by the vehicle owner or insurance agent describing the accident scene and circumstances. Vehicle information data includes the corresponding vehicle configuration information obtained from the vehicle insurance policy and vehicle accident information entered into the database. This information is filled in by insurance agents after on-site inspections and saved in the system database. By calculating the estimated loss pricing of the vehicle, it is possible to effectively obtain the estimated pricing corresponding to the vehicle image data, vehicle description data, and vehicle information data, providing data support for subsequent learner training.

[0082] Step S20, performing feature extraction on the vehicle image data, the vehicle description data, and the vehicle information data to obtain vehicle image features, vehicle description features, and vehicle information features;

[0083] In this embodiment, the vehicle information data includes data such as the vehicle's brand, model, year, mileage, and vehicle condition. By processing the above data through the vehicle valuation database, the value and repair cost of the vehicle can be effectively calculated, and the vehicle's loss can be further valued to obtain the vehicle loss estimated pricing.

[0084] Step S30, performing a first feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle loss feature;

[0085] Step S40, performing a second feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle accident feature;

[0086] Step S50, performing a third feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle configuration feature;

[0087] Step S60, performing a fourth feature combination on the vehicle description feature and the vehicle information feature to obtain a vehicle insurance policy feature;

[0088] In this embodiment, the vehicle loss characteristics represent the characteristics corresponding to the vehicle loss situation, including the vehicle loss location, vehicle loss parts, vehicle loss degree, etc. The vehicle accident characteristics represent the characteristics corresponding to the vehicle accident situation, including the vehicle accident location, vehicle accident method, the situation of both parties in the vehicle accident, etc. The vehicle configuration characteristics represent the characteristics corresponding to the vehicle configuration situation, including vehicle brand, vehicle model, vehicle configuration parts, etc.

[0089] Step S70: inputting the vehicle loss characteristics, the vehicle accident characteristics, the vehicle information characteristics, the vehicle insurance policy characteristics, and the vehicle loss estimation pricing into a preset learner for training, and performing ensemble learning on the trained learner to obtain a vehicle loss estimation model; and

[0090] In this embodiment, ensemble learning may use a random forest composed of several decision trees to obtain learning results by classifying vehicle loss features, vehicle accident features, vehicle information features, and vehicle insurance policy features multiple times.

[0091] Step S80: Acquire repair report data, input the repair report data into the vehicle loss estimation model to perform loss estimation, and obtain loss estimation data.

[0092] This embodiment identifies and classifies sample data and extracts features from the classified data to obtain vehicle image features based on accident scene images, vehicle description features based on voice and text data recorded from the accident, and vehicle information features based on the vehicle's basic configuration information. By combining the vehicle image features, vehicle description features, and vehicle information features, it obtains vehicle loss features that characterize the vehicle's damage situation, vehicle accident features that characterize the accident situation, vehicle configuration features that characterize the vehicle's configuration, and vehicle policy features that characterize the vehicle's insurance policy situation. By inputting the vehicle loss features, vehicle accident features, vehicle information features, vehicle policy features, and vehicle loss estimation pricing calculated based on the vehicle information data into a pre-set learner for training, and then performing ensemble learning on the trained learner, a vehicle loss estimation model that accurately estimates vehicle losses is obtained. Acquired repair report data is then input into the vehicle loss estimation model to obtain loss estimation data, thereby achieving automated and efficient accurate vehicle loss estimation. This embodiment can be applied to the field of auto insurance. The above method can accurately estimate vehicle losses at the time of an accident, providing effective data for subsequent auto insurance claims and vehicle repairs.

[0093] Continue to refer Figure 3 In some optional implementations of this embodiment, step S10 includes the following steps:

[0094] Step S101, performing image recognition on the vehicle sample data, and using the recognized data as the vehicle image data;

[0095] Step S102, performing voice recognition and text recognition on the vehicle sample data, and arranging the recognized data to obtain the vehicle description data; and

[0096] Step S103: identifying data containing corresponding fields in the vehicle sample data according to preset field identification information, and using the identified data as the vehicle information data.

[0097] In this embodiment, the field identification information is an SQL statement used to retrieve specific characters and fields. This SQL statement is used to search for data within the vehicle sample data that contains specific characters and fields. For example, if vehicle information data includes a vehicle's make, model, year, mileage, and condition, the specific characters are set to include the brand name, model name, year, and mileage units, and the fields are set to describe the vehicle's condition, such as vehicle size, vehicle model, and vehicle condition.

[0098] Continue to refer Figure 4 In some optional implementations of this embodiment, step S20 includes the following steps:

[0099] Step S201, performing data cleaning on the vehicle image data, the vehicle description data, and the vehicle information data to obtain valid image data, valid description data, and valid information data;

[0100] In this embodiment, data cleaning includes data deduplication, data consistency check, data missing detection, data missing value processing, etc.

[0101] Step S202, identifying a vehicle image contained in the valid image data, and performing feature extraction on the vehicle image to obtain the vehicle image features;

[0102] Step S203, identifying the voice information and / or text information of the effective description data, and performing feature extraction on the voice information and / or text information to obtain the vehicle description feature; and

[0103] Step S204: identifying the information fields contained in the valid information data, performing feature extraction on the information data corresponding to the information fields, and obtaining the vehicle information features.

[0104] This embodiment performs data cleaning on vehicle image data, vehicle description data, and vehicle information data to obtain non-repetitive and easily calculated valid image data, valid description data, and valid information data, and then performs feature extraction on the valid image data, valid description data, and valid information data, respectively, to effectively obtain vehicle image features, vehicle description features, and vehicle information features that can accurately characterize the features of the valid image data, valid description data, and valid information data.

[0105] In some optional implementations of this embodiment, step S30 includes the following steps:

[0106] Perform feature intersection on the vehicle damage image feature, the vehicle damage condition feature, and the vehicle damage information feature to obtain the vehicle loss feature.

[0107] In some optional implementations of this embodiment, step S40 includes the following steps:

[0108] Perform feature intersection on the accident scene image features, the vehicle accident condition features, and the vehicle accident information features to obtain the vehicle accident features.

[0109] In some optional implementations of this embodiment, step S50 includes the following steps:

[0110] Perform feature intersection on the vehicle configuration image feature, the vehicle configuration status feature, and the vehicle configuration information feature to obtain the vehicle configuration feature.

[0111] In some optional implementations of this embodiment, step S60 includes the following steps:

[0112] Perform feature intersection on the vehicle insurance policy status feature and the vehicle insurance policy information feature to obtain the vehicle insurance policy feature.

[0113] In this embodiment, the combination of related features is achieved through feature crossover. Feature crossover enables features of different dimensions to be effectively combined to obtain feature data that can better characterize the features of the object.

[0114] Continue to refer Figure 5 In some optional implementations of this embodiment, step S70 includes the following steps:

[0115] Step S701, obtaining characteristic identifiers of the vehicle loss characteristics, the vehicle accident characteristics, the vehicle information characteristics, and the vehicle insurance policy characteristics;

[0116] In this embodiment, the feature identifier is an identification mark added to different features. The identification mark can be used to quickly obtain the category to which the feature belongs, so that the feature can be conveniently and efficiently distinguished into the corresponding learner.

[0117] Step S702: assigning the vehicle loss feature, the vehicle accident feature, the vehicle information feature, and the vehicle insurance policy feature to corresponding loss feature learners, accident feature learners, information feature learners, and insurance policy feature learners according to the feature identifiers;

[0118] Step S703: inputting the vehicle loss estimation pricing into the loss feature learner, the accident feature learner, the information feature learner, and the policy feature learner for training;

[0119] Step S704, performing integrated learning on the loss feature learner, the accident feature learner, the information feature learner, and the policy feature learner according to a preset integrated learning rule to obtain an initial loss estimation model; and

[0120] Step S705 , evaluating the accuracy of the initial loss estimation model, and adjusting the initial loss estimation model according to the evaluation result to obtain the vehicle loss estimation model.

[0121] In this embodiment, by adjusting the initial estimation model, the accuracy of the model can be effectively improved, and flexible adjustments can be made so that the model can meet prediction requirements in different situations.

[0122] Continue to refer Figure 6 In some optional implementations of this embodiment, step S705 includes the following steps:

[0123] Step S7051: obtaining a preset model evaluation rule, and evaluating the initial loss estimation model according to the model evaluation rule to obtain the model accuracy;

[0124] In this embodiment, the model evaluation rule obtains the similarity of the estimated results output by the model, evaluates the model accuracy based on a pre-stored accuracy relationship mapping table, and outputs the model accuracy value to facilitate subsequent comparison with the standard value. For example, if the similarity of the estimated results output by the model is 90%, 85%, and 95%, respectively, the average value of 90% can be taken, and the model accuracy value corresponding to 90% can be found in the accuracy relationship mapping table. The above model evaluation rules can be adjusted accordingly based on actual circumstances.

[0125] Step S7052: Compare the model accuracy with a preset standard value to determine whether the model accuracy meets the standard value;

[0126] In this embodiment, the preset standard value is 90, and the model accuracy is a score consistent with the standard value unit. By comparing the model accuracy with the preset standard value, it is determined whether the model meets the preset requirements after adjustment. The above preset standard value can be adjusted accordingly according to actual conditions.

[0127] Step S7053: If the model accuracy meets the standard value, the initial loss estimation model is used as the vehicle loss estimation model; and

[0128] Step S7054: If the model accuracy does not meet the standard value, the loss feature learner, the accident feature learner, the information feature learner, and the policy feature learner are iteratively adjusted and re-integrated to obtain an optimized loss estimation model until the model accuracy of the optimized loss estimation model meets the standard value. After the model accuracy of the optimized loss estimation model meets the standard value, the optimized loss estimation model is used as the vehicle loss estimation model.

[0129] This embodiment effectively optimizes and adjusts the initial loss estimation model by iteratively adjusting the learner to ensure that the vehicle loss estimation model ultimately outputted can meet the set accuracy requirements.

[0130] In some optional implementations of this embodiment, after step S50, the following steps are further included:

[0131] Obtaining preset field identification information, identifying the loss estimation data according to the field identification information, and obtaining data key fields;

[0132] Obtaining a preset report text template, identifying the report text template according to the field identification information, and obtaining text key fields; and

[0133] The loss estimation data is correspondingly filled into the report text template according to the data key fields and the text key fields to generate a model loss estimation report.

[0134] This embodiment obtains field identification information to identify the loss estimate data and the report text template, thereby obtaining data key fields and text key fields. Based on the corresponding relationship between the data key fields and the text key fields, the loss estimate data is entered into the corresponding positions in the report text template to generate a model loss estimate report, facilitating subsequent review and adjustment. For example, if the data key fields are loss extent, loss location, and loss pricing, and the corresponding text key fields are also loss extent, loss location, and loss pricing, the loss extent, loss location, and loss pricing in the loss estimate data are entered into the corresponding positions in the report text template.

[0135] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware via computer-readable instructions. The computer-readable instructions can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes in the above-described method embodiments. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0136] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0137] Further references Figure 7 , as a response to the above Figure 1 The present application provides an embodiment of a business vehicle loss estimation device, which is similar to the embodiment of the present invention. Figure 1 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0138] like Figure 7 As shown, the vehicle loss estimation device 900 of this embodiment includes: a data classification module 901, a feature extraction module 902, a first feature combination module 903, a second feature combination module 904, a third feature combination module 905, a fourth feature combination module 906, a model training module 907, and a model prediction module 908.

[0139] The data classification module 901 is used to obtain sample data, identify and classify the sample data to obtain vehicle image data, vehicle description data, and vehicle information data, and calculate the vehicle loss estimation pricing based on the vehicle information data and a preset calculation formula;

[0140] A feature extraction module 902 is used to extract features from the vehicle image data, the vehicle description data, and the vehicle information data to obtain vehicle image features, vehicle description features, and vehicle information features;

[0141] A first feature combination module 903 is configured to perform a first feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle loss feature;

[0142] A second feature combination module 904 is configured to perform a second feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle accident feature;

[0143] A third feature combination module 905 is configured to perform a third feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle configuration feature;

[0144] A fourth feature combination module 906 is configured to perform a fourth feature combination on the vehicle description feature and the vehicle information feature to obtain a vehicle insurance policy feature;

[0145] A model training module 907 is configured to input the vehicle loss characteristics, the vehicle accident characteristics, the vehicle information characteristics, the vehicle insurance policy characteristics, and the vehicle loss estimation pricing into a preset learner for training, and perform integrated learning on the trained learner to obtain a vehicle loss estimation model; and

[0146] The model prediction module 908 is used to obtain repair report data, input the repair report data into the vehicle loss estimation model to perform loss estimation, and obtain loss estimation data.

[0147] By adopting the above-mentioned device, this embodiment can effectively realize accurate estimation of vehicle losses in an automated and efficient manner.

[0148] To solve the above technical problems, the present application also provides a computer device. Figure 8 , Figure 8 This is a basic structural block diagram of the computer device in this embodiment.

[0149] The computer device 2 includes a memory 21, a processor 22, and a network interface 23 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 2 with components 21-23, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0150] The computer device may be a desktop computer, notebook computer, PDA, cloud server, etc. The computer device may interact with the user via a keyboard, mouse, remote control, touchpad, or voice control device.

[0151] The memory 21 includes at least one type of readable storage medium, including flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic storage, a magnetic disk, an optical disk, etc. In some embodiments, the memory 21 may be an internal storage unit of the computer device 2, such as the hard disk or memory of the computer device 2. In other embodiments, the memory 21 may also be an external storage device of the computer device 2, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device 2. Of course, the memory 21 may also include both the internal storage unit of the computer device 2 and its external storage device. In this embodiment, the memory 21 is generally used to store the operating system and various application software installed on the computer device 2, such as computer-readable instructions for the vehicle loss estimation method. In addition, the memory 21 can also be used to temporarily store various types of data that have been output or are to be output.

[0152] In some embodiments, the processor 22 can be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 22 is generally used to control the overall operation of the computer device 2. In this embodiment, the processor 22 is used to execute computer-readable instructions stored in the memory 21 or process data, such as computer-readable instructions for executing the vehicle loss estimation method.

[0153] The network interface 23 may include a wireless network interface or a wired network interface. The network interface 23 is generally used to establish a communication connection between the computer device 2 and other electronic devices.

[0154] This embodiment, by adopting the above-mentioned computer equipment, can effectively realize accurate estimation of vehicle losses in an automated and efficient manner.

[0155] The present application also provides another embodiment, namely, providing a computer-readable storage medium, which stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the vehicle loss estimation method as described above.

[0156] This embodiment, by adopting the above-mentioned computer-readable storage medium, can effectively realize accurate estimation of vehicle losses in an automated and efficient manner.

[0157] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0158] Obviously, the embodiments described above are only some of the embodiments of the present application, rather than all of the embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions described in the aforementioned specific embodiments, or to make equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the present application specification and the accompanying drawings, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present application.

Claims

1. A vehicle loss estimation method, characterized in that: The steps include: Acquiring sample data, identifying and classifying the sample data to obtain vehicle image data, vehicle description data, and vehicle information data, and calculating vehicle loss estimation pricing based on the vehicle information data and a preset calculation formula; Performing feature extraction on the vehicle image data, the vehicle description data, and the vehicle information data respectively to obtain vehicle image features, vehicle description features, and vehicle information features; Performing a first feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle loss feature; Performing a second feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle accident feature; Performing a third feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle configuration feature; Performing a fourth feature combination on the vehicle description feature and the vehicle information feature to obtain a vehicle insurance policy feature; Inputting the vehicle loss characteristics, the vehicle accident characteristics, the vehicle information characteristics, the vehicle insurance policy characteristics, and the vehicle loss estimation pricing into a preset learner for training, and performing ensemble learning on the trained learner to obtain a vehicle loss estimation model; and Obtain repair report data, input the repair report data into the vehicle loss estimation model to perform loss estimation, and obtain loss estimation data.

2. The vehicle loss estimation method according to claim 1, characterized in that: The step of identifying and classifying the sample data to obtain vehicle image data, vehicle description data, and vehicle information data specifically includes: Performing image recognition on the vehicle sample data, and using the recognized data as the vehicle image data; Performing voice recognition and text recognition on the vehicle sample data, and collating the recognized data to obtain the vehicle description data; and The data containing the corresponding field in the vehicle sample data is identified according to the preset field identification information, and the identified data is used as the vehicle information data.

3. The vehicle loss estimation method according to claim 1, characterized in that: The step of respectively extracting features from the vehicle image data, the vehicle description data, and the vehicle information data to obtain vehicle image features, vehicle description features, and vehicle information features specifically includes: Performing data cleaning on the vehicle image data, the vehicle description data, and the vehicle information data to obtain valid image data, valid description data, and valid information data; Identifying a vehicle image contained in the valid image data, and performing feature extraction on the vehicle image to obtain the vehicle image feature; Identifying the voice information and / or text information of the effective description data, and performing feature extraction on the voice information and / or text information to obtain the vehicle description feature; and Identify the information fields contained in the valid information data, perform feature extraction on the information data corresponding to the information fields, and obtain the vehicle information features.

4. The vehicle loss estimation method according to claim 1, characterized in that: The vehicle image features include vehicle damage image features, accident scene image features, and vehicle configuration image features; the vehicle description features include vehicle damage condition features, vehicle accident condition features, vehicle configuration condition features, and vehicle insurance policy condition features; the vehicle information features include vehicle damage information features, vehicle accident information features, vehicle configuration information features, and vehicle insurance policy information features; and the step of performing a first feature combination on the vehicle image features, the vehicle description features, and the vehicle information features to obtain vehicle loss features specifically includes the following steps: Performing feature intersection on the vehicle damage image feature, the vehicle damage condition feature, and the vehicle damage information feature to obtain the vehicle loss feature; The step of performing a second feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle accident feature specifically includes the following steps: Performing feature intersection on the accident scene image feature, the vehicle accident condition feature, and the vehicle accident information feature to obtain the vehicle accident feature; The step of performing a third feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain the vehicle configuration feature specifically includes the following steps: Performing feature intersection on the vehicle configuration image feature, the vehicle configuration status feature, and the vehicle configuration information feature to obtain the vehicle configuration feature; and The step of performing a fourth feature combination on the vehicle description feature and the vehicle information feature to obtain the vehicle insurance policy feature specifically includes the following steps: Perform feature intersection on the vehicle insurance policy status feature and the vehicle insurance policy information feature to obtain the vehicle insurance policy feature.

5. The vehicle loss estimation method according to claim 1, characterized in that: The step of inputting the vehicle loss characteristics, the vehicle accident characteristics, the vehicle information characteristics, the vehicle insurance policy characteristics, and the vehicle loss estimation pricing into a preset learner for training, and performing ensemble learning on the trained learner to obtain a vehicle loss estimation model specifically includes: Obtaining feature identifiers of the vehicle loss feature, the vehicle accident feature, the vehicle information feature, and the vehicle insurance policy feature; Allocating the vehicle loss feature, the vehicle accident feature, the vehicle information feature, and the vehicle insurance policy feature to corresponding loss feature learners, accident feature learners, information feature learners, and insurance policy feature learners according to the feature identifiers; Inputting the vehicle loss estimation pricing into the loss feature learner, the accident feature learner, the information feature learner, and the policy feature learner for training; Performing integrated learning on the loss feature learner, the accident feature learner, the information feature learner, and the policy feature learner according to a preset integrated learning rule to obtain an initial loss estimation model; and The accuracy of the initial loss estimation model is evaluated, and the initial loss estimation model is adjusted according to the evaluation result to obtain the vehicle loss estimation model.

6. The vehicle loss estimation method according to claim 5, characterized in that: The step of evaluating the accuracy of the initial loss estimation model and adjusting the initial loss estimation model according to the evaluation result to obtain the vehicle loss estimation model specifically includes: Obtaining a preset model evaluation rule, and evaluating the initial loss estimation model according to the model evaluation rule to obtain model accuracy; Comparing the model accuracy with a preset standard value to determine whether the model accuracy meets the standard value; If the model accuracy meets the standard value, using the initial loss estimation model as the vehicle loss estimation model; and If the model accuracy does not meet the standard value, the loss feature learner, the accident feature learner, the information feature learner, and the policy feature learner are iteratively adjusted and re-integrated and learned to obtain an optimized loss estimation model until the model accuracy of the optimized loss estimation model meets the standard value. After the model accuracy of the optimized loss estimation model meets the standard value, the optimized loss estimation model is used as the vehicle loss estimation model.

7. The vehicle loss estimation method according to claim 1, characterized in that: After the steps of acquiring the repair report data and inputting the repair report data into the vehicle loss estimation model to perform loss estimation to obtain loss estimation data, the following steps are also included: Obtaining preset field identification information, identifying the loss estimation data according to the field identification information, and obtaining data key fields; Obtaining a preset report text template, identifying the report text template according to the field identification information, and obtaining text key fields; and The loss estimation data is correspondingly filled into the report text template according to the data key fields and the text key fields to generate a model loss estimation report.

8. A vehicle loss estimation device, characterized in that: include: A data classification module is used to obtain sample data, identify and classify the sample data to obtain vehicle image data, vehicle description data, and vehicle information data, and calculate vehicle loss estimation pricing based on the vehicle information data and a preset calculation formula; A feature extraction module is used to extract features from the vehicle image data, the vehicle description data, and the vehicle information data to obtain vehicle image features, vehicle description features, and vehicle information features; A first feature combination module is used to perform a first feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle loss feature; A second feature combination module is used to perform a second feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle accident feature; a third feature combination module, configured to perform a third feature combination on the vehicle image feature, the vehicle description feature, and the vehicle information feature to obtain a vehicle configuration feature; a fourth feature combination module, configured to perform a fourth feature combination on the vehicle description feature and the vehicle information feature to obtain a vehicle insurance policy feature; a model training module, configured to input the vehicle loss characteristics, the vehicle accident characteristics, the vehicle information characteristics, the vehicle insurance policy characteristics, and the vehicle loss estimation pricing into a preset learner for training, and perform ensemble learning on the trained learner to obtain a vehicle loss estimation model; and The model prediction module is used to obtain repair report data, input the repair report data into the vehicle loss estimation model to perform loss estimation, and obtain loss estimation data.

9. A computer device comprising a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the vehicle loss estimation method according to any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the vehicle loss estimation method according to any one of claims 1 to 7.

Images