A vehicle valuation method and apparatus

Abstract

The application relates to a vehicle valuation method and device, and the method comprises the following steps: acquiring vehicle model information of a vehicle to be valued and attribute information related to valuation of the vehicle to be valued; for each vehicle model information, determining an attribute category corresponding to each attribute information and marking the attribute category of the attribute information to obtain marked attribute information; inputting the marked attribute information and the vehicle model information into a target vehicle valuation model; and obtaining a valuation result of the vehicle to be valued according to the marked attribute information, the vehicle model information, a valuation function and a weight of the vehicle to be valued. According to the scheme, from the perspective of the vehicle life cycle, influence factors of vehicle valuation are divided into stable factors and fluctuation factors, and then a valuation function corresponding to each factor is constructed, vehicle data in each dimension can be standardized, vehicle model information and attribute information are introduced, the real value of the vehicle can be better reflected, and the accuracy of valuation is improved.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a vehicle valuation method and apparatus. Background Technology

[0002] With changes in domestic consumption levels and attitudes, the used car market is gradually booming, with transaction volume steadily increasing. Providing accurate and reasonable valuations for used cars is becoming increasingly important. Currently, valuations rely heavily on appraisers. However, due to significant differences in appraisers' experience and abilities, the same car may receive different valuations from different appraisers. Furthermore, because individual appraisers cannot collect sufficient historical vehicle transaction data for reference when assessing the current condition of a vehicle, their understanding of the vehicle is not comprehensive enough, leading to inaccurate valuations.

[0003] This shows that existing used car valuation methods still suffer from the problem of lacking a unified valuation standard and being inaccurate in valuation. Summary of the Invention

[0004] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, a first aspect of this invention proposes a vehicle valuation method, the method comprising:

[0005] Obtain the attribute information of the vehicle to be valued;

[0006] Determine the attribute category corresponding to each attribute information, and the attribute category column is determined based on the degree of influence of the attribute information on the vehicle valuation over time;

[0007] The attribute information and the attribute category are input into the target vehicle valuation model; the valuation function in the target vehicle valuation model includes sub-valuation functions corresponding to each attribute information and weights corresponding to each attribute category; the target vehicle valuation model is constructed based on vehicle valuation data within a historical time period.

[0008] The valuation result of the vehicle to be valued is obtained based on the attribute information and the valuation function.

[0009] Optionally, the attribute categories include stable attributes and volatile attributes. The stable attribute is the attribute whose impact on the vehicle valuation over time is less than a first threshold, and the volatile attribute is the attribute whose impact on the vehicle valuation over time is greater than or equal to the first threshold.

[0010] Optionally, the target vehicle estimation model is trained using the following method:

[0011] Obtain sample vehicle valuation data within a preset historical time period. The sample vehicle valuation data includes: sample attribute information of the vehicle and the influence value of the sample attribute information on the vehicle valuation in each sub-time period. The preset historical time period includes multiple sub-time periods.

[0012] From the sample vehicle valuation data, the valuation data corresponding to each stable attribute and the valuation data corresponding to each volatility attribute are obtained respectively, resulting in multiple stable attribute valuation data and multiple volatility attribute valuation data.

[0013] Based on the influence values ​​in the stable attribute valuation data, a first initial valuation function is determined for each of the stable attributes; and based on the influence values ​​in the volatility attribute valuation data, a second initial valuation function is determined for each of the volatility attributes.

[0014] Using the sample vehicle estimation data, the first initial estimation function and the second initial estimation function are trained respectively to obtain the stable attribute initial model composed of the first initial estimation function of each stable attribute and the fluctuation attribute initial model composed of the second initial estimation function of each fluctuation attribute.

[0015] The target vehicle valuation model is obtained based on the pre-set weights of the initial model of the stable attribute and the initial model of the volatility attribute.

[0016] Optionally, determining the first initial estimation function corresponding to each stable attribute based on the influence value in the stable attribute estimation data includes:

[0017] Set the weights corresponding to each of the sub-time periods;

[0018] For each of the stable attributes, the weighted average of the influence value and the weight corresponding to each of the sub-time periods is calculated to obtain the weighted influence value;

[0019] The functional relationship between the vehicle age information and the weighted influence value is determined under various values ​​of the stable attribute to obtain the initial estimation function corresponding to the stable attribute.

[0020] Optionally, the sample attribute information includes optional configuration information, and before setting the weights corresponding to each of the sub-time periods, it further includes:

[0021] Cluster the optional configuration information of each of the sample vehicle valuation data to obtain multiple sets of similar optional configuration information;

[0022] Determining the functional relationship between the vehicle age information and the weighted influence value includes:

[0023] For each group of similar optional configuration information, the functional relationship between the vehicle age information and the weighted influence value is determined.

[0024] Optionally, determining the second initial valuation function corresponding to each volatility attribute based on the influence value in the volatility attribute valuation data includes:

[0025] For each of the aforementioned fluctuation attributes, the value of the fluctuation attribute is used as a feature vector, and the influence value corresponding to the fluctuation attribute is used as a target value. The preset algorithm model is trained to obtain the functional relationship between the fluctuation attribute and the influence value.

[0026] The initial valuation function corresponding to the volatility attribute is determined based on the aforementioned functional relationship.

[0027] Optionally, the fluctuation attribute includes vehicle condition information, which includes: accident information and maintenance information of the vehicle. Before training the preset algorithm model using the value of the fluctuation attribute as a feature vector and the influence value corresponding to the fluctuation attribute as a target value, the method further includes:

[0028] Obtain vehicles with the same vehicle age information from the sample vehicle valuation data to obtain a data set of multiple vehicles with the same vehicle age information;

[0029] Cluster the accident information of vehicles in the data set with the same vehicle age information to obtain multiple sets of similar accident information;

[0030] The vehicle repair information in the accident information set is clustered to obtain multiple sets of similar repair information;

[0031] Obtain the impact value of the vehicle in each group's maintenance information set.

[0032] Optionally, the step of using the value of the fluctuation attribute as a feature vector and the influence value corresponding to the fluctuation attribute as a target value to train a preset algorithm model includes:

[0033] For each set of maintenance information, the vehicle age information, the accident information, and the maintenance information are used as feature vectors, and the impact value of the vehicle in the set of maintenance information is used as the target value. The preset algorithm model is trained to obtain the functional relationship between the vehicle age information and the impact value corresponding to the set of maintenance information.

[0034] The functional relationship is a piecewise function with the vehicle age information as the independent variable and the influence value as the dependent variable. Each sub-segment of the piecewise function is a constant function, and the value of the constant function decreases as the vehicle age information increases.

[0035] Optionally, the fluctuation attribute includes mileage information. For each value of the mileage information, the functional relationship between the vehicle age information and the influence value corresponding to the mileage information value is: a linear function with the vehicle age information as the independent variable and the influence value as the dependent variable, and the slope of the linear function is negative.

[0036] Optionally, the fluctuation attribute is the monthly information of the vehicle transaction, and the functional relationship between the monthly information and the influence value is a broken line function with the monthly information as the independent variable and the influence value as the dependent variable.

[0037] Optionally, the attribute information includes at least the vehicle model information, optional configuration information, license plate information, usage information, and external environment information of the vehicle to be valued.

[0038] Optionally, the license plate information in the stable attributes includes registration information and license plate attribute information; the usage information in the stable attributes includes usage nature information and number of transfers; and the external environment information in the stable attributes includes national preferential policy information.

[0039] Optionally, the usage information in the fluctuation attribute includes vehicle age information, vehicle condition information, mileage information, and claim information, and the external policy information in the fluctuation attribute includes the month information of the vehicle transaction.

[0040] This invention also provides a vehicle valuation device, the device comprising:

[0041] The attribute information acquisition module is used to acquire the attribute information of the vehicle to be valued.

[0042] The attribute category determination module is used to determine the attribute category corresponding to each attribute information. The attribute category is determined based on the degree of influence of the attribute information on the vehicle valuation over time.

[0043] The data input module is used to input the attribute information and the attribute category into the target vehicle valuation model; the valuation function in the target vehicle valuation model includes sub-valuation functions corresponding to each attribute information and weights corresponding to the attribute category; the target vehicle valuation model is constructed based on vehicle valuation data within a historical time period.

[0044] The valuation module is used to obtain the valuation result of the vehicle to be valued based on the attribute information and the valuation function.

[0045] Optionally, the apparatus further includes a model training module, the model training module being used for:

[0046] Obtain sample vehicle valuation data within a preset historical time period. The sample vehicle valuation data includes: sample attribute information of the vehicle and the influence value of the sample attribute information on the vehicle valuation in each sub-time period. The preset historical time period includes multiple sub-time periods.

[0047] From the sample vehicle valuation data, the valuation data corresponding to each stable attribute and the valuation data corresponding to each volatility attribute are obtained respectively, resulting in multiple stable attribute valuation data and multiple volatility attribute valuation data.

[0048] Based on the influence values ​​in the stable attribute valuation data, a first initial valuation function is determined for each of the stable attributes; and based on the influence values ​​in the volatility attribute valuation data, a second initial valuation function is determined for each of the volatility attributes.

[0049] Using the sample vehicle estimation data, the first initial estimation function and the second initial estimation function are trained respectively to obtain the stable attribute initial model composed of the first initial estimation function of each stable attribute and the fluctuation attribute initial model composed of the second initial estimation function of each fluctuation attribute.

[0050] The target vehicle valuation model is obtained based on the pre-set weights of the initial model of the stable attribute and the initial model of the volatility attribute.

[0051] Optionally, the model training module is specifically used for:

[0052] Set the weights corresponding to each of the sub-time periods;

[0053] For each of the stable attributes, the weighted average of the influence value and the weight corresponding to each of the sub-time periods is calculated to obtain the weighted influence value;

[0054] The functional relationship between the vehicle age information and the weighted influence value is determined under various values ​​of the stable attribute to obtain the initial estimation function corresponding to the stable attribute.

[0055] Optionally, the model training module is specifically used for:

[0056] Cluster the optional configuration information of each of the sample vehicle valuation data to obtain multiple sets of similar optional configuration information;

[0057] Determining the functional relationship between the vehicle age information and the weighted influence value includes:

[0058] For each group of similar optional configuration information, the functional relationship between the vehicle age information and the weighted influence value is determined.

[0059] Optionally, the model training module is specifically used for:

[0060] For each of the aforementioned fluctuation attributes, the value of the fluctuation attribute is used as a feature vector, and the influence value corresponding to the fluctuation attribute is used as a target value. The preset algorithm model is trained to obtain the functional relationship between the fluctuation attribute and the influence value.

[0061] The initial valuation function corresponding to the volatility attribute is determined based on the aforementioned functional relationship.

[0062] Optionally, the model training module is specifically used for:

[0063] Obtain vehicles with the same vehicle age information from the sample vehicle valuation data to obtain a data set of multiple vehicles with the same vehicle age information;

[0064] Cluster the accident information of vehicles in the data set with the same vehicle age information to obtain multiple sets of similar accident information;

[0065] The vehicle repair information in the accident information set is clustered to obtain multiple sets of similar repair information;

[0066] Obtain the impact value of the vehicle in each group's maintenance information set.

[0067] Optionally, the model training module is specifically used for:

[0068] For each set of maintenance information, the vehicle age information, the accident information, and the maintenance information are used as feature vectors, and the impact value of the vehicle in the set of maintenance information is used as the target value. The preset algorithm model is trained to obtain the functional relationship between the vehicle age information and the impact value corresponding to the set of maintenance information.

[0069] The functional relationship is a piecewise function with the vehicle age information as the independent variable and the influence value as the dependent variable. Each sub-segment of the piecewise function is a constant function, and the value of the constant function decreases as the vehicle age information increases.

[0070] A third aspect of the present invention provides an electronic device comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set, or an instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the vehicle valuation method as described in the first aspect.

[0071] A fourth aspect of the present invention provides a computer-readable storage medium storing at least one instruction, at least one program, a code set, or an instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the vehicle valuation method as described in the first aspect.

[0072] According to specific embodiments provided by the present invention, the present invention has the following technical effects:

[0073] The vehicle valuation method provided in this invention involves: acquiring attribute information of the vehicle to be valued; determining the attribute category corresponding to each attribute information, with the attribute category determined based on the degree of influence of the attribute information on vehicle valuation over time; inputting the attribute information and attribute category into a target vehicle valuation model; the valuation function in the target vehicle valuation model includes sub-valuation functions corresponding to each attribute information and weights corresponding to the attribute categories; the target vehicle valuation model is constructed based on vehicle valuation data within a historical time period; and the valuation result of the vehicle to be valued is obtained based on the attribute information and valuation function. This solution constructs a vehicle valuation model based on historical data, classifying the influencing factors of vehicle valuation into stable and fluctuating factors from the perspective of the vehicle lifecycle, and then constructing valuation functions corresponding to each factor. This standardizes data from various dimensions of the vehicle and incorporates vehicle model and attribute information, better reflecting the true value of the vehicle and improving the accuracy of the valuation.

[0074] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0075] To more clearly illustrate the technical solutions of the present invention, the accompanying drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0076] Figure 1 A flowchart illustrating the steps of a vehicle valuation method provided in an embodiment of the present invention;

[0077] Figure 2 A flowchart illustrating the steps of training a target vehicle valuation model according to an embodiment of the present invention;

[0078] Figure 3 This is a schematic diagram illustrating the functional relationship between vehicle age and weighted influence value corresponding to the vehicle registration information provided in this embodiment of the invention.

[0079] Figure 4This is a schematic diagram illustrating the functional relationship between vehicle age and weighted influence value corresponding to license plate attribute information provided in an embodiment of the present invention.

[0080] Figure 5 This is a schematic diagram illustrating the functional relationship between vehicle age and weighted influence value corresponding to usage property information provided in this embodiment of the invention;

[0081] Figure 6 This is a schematic diagram illustrating the functional relationship between vehicle age and weighted influence value corresponding to the number of ownership transfers provided in this embodiment of the invention.

[0082] Figure 7 This is a schematic diagram of the initial estimation function corresponding to the vehicle age provided in an embodiment of the present invention;

[0083] Figure 8 This is a schematic diagram illustrating the functional relationship between vehicle condition and influence values ​​provided in an embodiment of the present invention.

[0084] Figure 9 This is a schematic diagram illustrating the functional relationship between mileage and impact value provided in an embodiment of the present invention;

[0085] Figure 10 A schematic diagram illustrating the functional relationship between the month of a vehicle transaction and its impact value, provided in an embodiment of the present invention;

[0086] Figure 11 This is a structural block diagram of a vehicle valuation device provided in an embodiment of the present invention. Detailed Implementation

[0087] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0088] This specification provides the operational steps for the methods described in the embodiments or flowcharts, but may include more or fewer operational steps based on conventional or non-inventive labor. In actual system or server product execution, the methods shown in the embodiments or drawings may be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment).

[0089] Figure 1 A flowchart illustrating the steps of a vehicle valuation method provided in an embodiment of the present invention. The method includes the following steps:

[0090] Step 101: Obtain the attribute information of the vehicle to be valued.

[0091] The life cycle of a car generally includes the sales stage, order stage, production stage, delivery stage, usage stage, resale stage, after-sales stage, and scrapping stage.

[0092] In the sales process, sales staff introduce vehicle functions, features, optional equipment, and national and corporate incentive policies to customers. In the order process, customers choose the model, color, and other optional configurations according to their preferences. Dealers also consider offering incentives based on sales performance and national policies, generating orders accordingly. In the production process, manufacturers produce vehicles according to customer orders. Issues with the precision of the manufacturing process and the quality of parts at this stage can lead to vehicle returns and claims after delivery, lowering the vehicle's valuation. During delivery, damage to the vehicle due to logistics or warehousing can result in returns or repairs. In the usage phase, customer driving habits and normal driving wear and tear, leading to simple maintenance or repairs, as well as accidents, all negatively impact the vehicle's valuation. In the resale phase, customers may resell the vehicle due to personal circumstances, affecting its valuation; the more frequent this happens, the greater the impact. The after-sales process generally involves routine maintenance and accident repairs; different after-sales services have varying impacts on the vehicle's valuation. In the scrapping stage, the estimated value of a scrapped vehicle has dropped to its lowest point, and at this time, selling the parts separately may be more valuable than the whole vehicle.

[0093] Throughout a vehicle's lifecycle, different factors influence its valuation. Attribute information about the vehicle being valued can be obtained from each stage of its lifecycle; this attribute information represents the factors affecting the vehicle's valuation.

[0094] In one possible implementation, the attribute information includes at least the vehicle model information, optional configuration information, license plate information, usage information, and external environment information of the vehicle to be valued.

[0095] Specifically, vehicle model information refers to the vehicle's license plate and model number, which is a major factor affecting the vehicle's valuation. Optional configuration information refers to features that can be directly selected by the buyer according to their preferences or needs when ordering the vehicle. Examples include adding navigation, leather seats, a power sunroof, and vehicle stability control. Adding optional configurations can improve comfort and safety.

[0096] License plate information refers to whether a vehicle has been registered.

[0097] Usage information refers to information related to vehicle use, such as whether the vehicle is used by an individual or a company, the number of times the vehicle has been transferred, the mileage driven, basic vehicle condition information, and vehicle claim information.

[0098] External environmental information refers to external information that is unrelated to the vehicle itself but affects the vehicle's valuation, such as the month the vehicle is currently being sold or the national preferential policies in effect at the time of the sale.

[0099] Step 102: Determine the attribute category corresponding to each attribute information. The attribute category is determined based on the degree of influence of the attribute information on the vehicle valuation over time.

[0100] Throughout a vehicle's lifecycle, various factors influence its valuation, and the degree of influence of different attributes on valuation changes over time. Based on the degree of influence of each attribute on vehicle valuation over time, the attributes are categorized. Attributes with a greater degree of influence on vehicle valuation over time are grouped into one category, and those with a smaller degree of influence are grouped into another, resulting in the attribute categories corresponding to the attribute information.

[0101] For example, attributes such as registration information, license plate attributes, and usage information have little impact on valuation over time and are grouped into one category; attributes such as vehicle age, vehicle condition, mileage, and claims have a greater impact on valuation over time and are grouped into another category.

[0102] Step 103: Input the attribute information and the attribute category into the target vehicle valuation model; the valuation function in the target vehicle valuation model includes the sub-valuation function corresponding to each attribute information and the weight corresponding to the attribute category; the target vehicle valuation model is constructed based on vehicle valuation data within a historical time period.

[0103] By combining vehicle order data, vehicle attribute information, vehicle valuation data, after-sales data, etc., an initial valuation model is constructed from the dimension of attribute categories. The initial valuation model is trained according to the impact of each attribute category on the valuation, and the target vehicle valuation model is obtained.

[0104] The target vehicle valuation model includes valuation functions for each vehicle model. Each valuation function is a weighted sum of the sub-valuation functions corresponding to each attribute information and the weights corresponding to the attribute categories to which they belong.

[0105] It is important to note that vehicle valuation is significantly affected by the vehicle's brand and model. A corresponding target vehicle valuation model can be built for each model. When valuing a vehicle, the vehicle's attribute information and attribute category can be input into the target vehicle valuation model corresponding to the model.

[0106] In one possible implementation, the attribute category includes stable attributes and volatile attributes, wherein the stable attributes are those whose impact on the vehicle valuation over time is less than a first threshold, and the volatile attributes are those whose impact on the vehicle valuation over time is greater than or equal to the first threshold.

[0107] Specifically, the attribute information is divided into attribute categories of stable attributes and volatility attributes, and the attribute categories are labeled to facilitate the subsequent construction of valuation functions for stable attributes and volatility attributes respectively.

[0108] Furthermore, the target vehicle valuation model can be expressed by equation (1) as follows:

[0109]

[0110] Among them, w i S represents the weight of the i-th stable attribute. i w represents the valuation corresponding to the i-th volatility attribute. j R represents the weight of the j-th fluctuation attribute. j Let N represent the valuation corresponding to the j-th volatility attribute, N represent the number of stable attributes, and M represent the number of volatility attributes.

[0111] In one possible implementation, the license plate information in the stable attributes includes registration information and license plate attribute information; the usage information in the stable attributes includes usage nature information and number of transfers; and the external environment information in the stable attributes includes national preferential policy information.

[0112] Stable attributes are those that have a relatively small impact on vehicle valuation over time. Attributes such as registration information, license plate attributes, usage information, number of ownership transfers, and national preferential policy information have a relatively small impact on vehicle valuation as vehicle age changes; therefore, these attributes are classified as stable attributes.

[0113] In one possible implementation, the usage information in the fluctuation attribute includes vehicle age, vehicle condition, mileage, and claims, and the external policy information in the fluctuation attribute includes the month of the vehicle transaction.

[0114] Volatility attributes are those that have a significant impact on valuation over time. Vehicle age, vehicle condition, mileage, claims, and the month the vehicle was sold all have a significant impact on vehicle valuation over time; therefore, these attributes are categorized as volatility attributes.

[0115] Step 104: Obtain the valuation result of the vehicle to be valued based on the attribute information and the valuation function.

[0116] Each vehicle model has a corresponding target estimation function, and each target estimation function includes multiple sub-estimation functions corresponding to multiple attribute information. First, the labeled attribute information is substituted into the corresponding sub-estimation function to obtain the sub-estimation result. For example, the attribute information includes license plate attribute information; substituting the license plate attribute information into the first sub-estimation function corresponding to the license plate attribute information yields the first sub-estimation result for the license plate attribute information. The attribute information also includes mileage information; substituting the mileage information into the second sub-estimation function corresponding to the mileage information yields the second sub-estimation result for the mileage information.

[0117] In this way, multiple sub-valuation results corresponding to multiple labeled attribute information are obtained.

[0118] The valuation result is obtained by weighting and summing multiple sub-valuation results with the weights corresponding to the attribute category. For example, if the weight of the stability attribute is 0.4 and the weight of the volatility attribute is 0.6, then the first sub-valuation result is multiplied by the weight of the stability attribute (0.4), and the second sub-valuation result is multiplied by the weight of the volatility attribute (0.6). The sum of these multiplications gives the vehicle valuation result.

[0119] The weights of stable and volatile attributes can be preset based on experience. By setting different weights for stable and volatile attributes, the influence of each attribute's information on the valuation results can be scientifically evaluated, making the valuation results more accurate.

[0120] In summary, in this embodiment of the invention, the following steps are taken: 1) Obtain the attribute information of the vehicle to be valued; 2) Determine the attribute category corresponding to each attribute information, wherein the attribute category is determined based on the degree of influence of the attribute information on the vehicle valuation over time; 3) Input the attribute information and the attribute category into a target vehicle valuation model; 4) The valuation function in the target vehicle valuation model includes sub-valuation functions corresponding to each attribute information and weights corresponding to each attribute category; 5) The target vehicle valuation model is constructed based on vehicle valuation data within a historical time period; 6) Obtain the valuation result of the vehicle to be valued based on the attribute information and the valuation function. This solution, from the perspective of the vehicle lifecycle, classifies the vehicle's attribute information according to the degree of influence of time on vehicle valuation, thereby constructing valuation functions corresponding to each attribute information and weights corresponding to each attribute category. This comprehensively integrates the attribute information from various dimensions of the vehicle and standardizes the valuation algorithm, which helps to reflect the true value of the vehicle and improves the accuracy of the valuation.

[0121] Figure 2 A flowchart illustrating the steps of a training method for a target vehicle estimation model provided in an embodiment of the present invention. The method includes the following steps:

[0122] Step 201: Obtain sample vehicle valuation data within a preset historical time period. The sample vehicle valuation data includes: sample attribute information of the vehicle and the influence value of the sample attribute information on the vehicle valuation in each sub-time period. The preset historical time period includes multiple sub-time periods.

[0123] The preset historical time period can be a period of time before the current time, such as ten years or three months before the current time. The specific preset time can be determined according to the needs and data conditions.

[0124] Sub-time periods refer to time periods within a preset historical time period. For example, if the preset historical time period is ten years before the current time, then the sub-time periods can be the first year, the second year, the third year, etc., before the current time.

[0125] Vehicle valuation data for a preset historical time period can be retrieved from the database to obtain sample vehicle valuation data. This sample vehicle valuation data specifically includes vehicle sample attribute information and the impact of sample attribute information for each sub-time period on the vehicle valuation.

[0126] The impact value refers to the influence of sample attribute information on vehicle valuation. For example, for the first sample vehicle, whose registration information is "registered," the impact value is the influence of registration on vehicle valuation, which can be 0.9, meaning that registration reduces the vehicle's valuation to 0.9 times the original price. For the second sample vehicle, whose registration information is "not registered," the impact value is the influence of not being registered on vehicle valuation, which can be 1.1, meaning that not being registered increases the vehicle's valuation to 1.1 times the original price.

[0127] Step 202: From the sample vehicle valuation data, obtain the valuation data corresponding to each stable attribute and the valuation data corresponding to each volatility attribute, to obtain multiple stable attribute valuation data and multiple volatility attribute valuation data.

[0128] The valuation data for stable attributes includes stable attribute information and the corresponding impact value, while the valuation data for volatility attributes includes volatility attribute information and the corresponding impact value.

[0129] Step 203: Based on the influence values ​​in the stable attribute valuation data, determine the first initial valuation function corresponding to each of the stable attributes; and based on the influence values ​​in the volatility attribute valuation data, determine the second initial valuation function corresponding to each of the volatility attributes.

[0130] Using stable attribute information as function variables and influence values ​​as function values, a suitable algorithm is selected based on the data characteristics of the stable attribute to construct the estimation function corresponding to the stable attribute, thus obtaining the first initial estimation function.

[0131] Similar to volatility attributes, volatility attribute information is used as function variables, and influence value is used as function value. Based on the data characteristics of volatility attributes, an appropriate algorithm is selected to construct the valuation function corresponding to the volatility attribute, thus obtaining the second initial valuation function.

[0132] Since the data characteristics of each attribute information are different, different algorithm models can be selected to construct the initial estimation function. For example, for optional configuration information, which includes multiple categories of configuration information, a classification algorithm, such as the SVM (Support Vector Machine) algorithm or the decision tree algorithm, can be used. The feature vector is the optional configuration data, and the target result is the estimation, thus constructing the initial estimation function.

[0133] It is important to note that stable or volatile properties include a variety of attribute information, and each attribute information corresponds to an initial valuation function.

[0134] Step 204: Using the sample vehicle estimation data, train the first initial estimation function and the second initial estimation function respectively to obtain the stable attribute initial model composed of the first initial estimation functions of each stable attribute and the fluctuation attribute initial model composed of the second initial estimation functions of each fluctuation attribute.

[0135] First, using the first and second initial estimation functions, the first and second algorithm models can be obtained. For each stable attribute, the stable attribute estimation data is input into the first algorithm model to obtain the predicted value. The predicted value is compared with the influence value corresponding to the stable attribute to obtain the residual value. Based on the residual value, the parameters of the first algorithm model are modified, and training continues to obtain the first initial model corresponding to that stable attribute.

[0136] The initial models corresponding to each stable attribute are combined to obtain the initial model of the stable attribute.

[0137] Similar to the fluctuation attribute, after training is completed, the second initial models corresponding to each fluctuation attribute are combined to obtain the initial model of the fluctuation attribute.

[0138] Step 205: Obtain the target vehicle valuation model based on the pre-set weights of the initial model of the stable attribute and the initial model of the fluctuation attribute.

[0139] Furthermore, the first weight of the pre-set stable attribute and the second weight of the volatility attribute are assigned to the two initial models to obtain the target vehicle valuation model.

[0140] In one possible implementation, determining the first initial estimation function corresponding to each of the stable attributes based on the influence values ​​in the stable attribute estimation data includes steps 2031-2033:

[0141] Step 2031: Set the weights corresponding to each of the sub-time periods;

[0142] Step 2032: For each of the stable attributes, calculate the weighted average of the influence value and the weight corresponding to each of the sub-time periods to obtain the weighted influence value;

[0143] Step 2033: Determine the functional relationship between the vehicle age information and the weighted influence value under various values ​​of the stable attribute, and obtain the initial valuation function corresponding to the stable attribute.

[0144] In steps 2031-2033, since data from more recent historical periods is relatively more reliable than data from more distant historical periods, more recent sub-time periods can be assigned higher weights, while more distant sub-time periods can be assigned lower weights. For example, if the preset historical time period is ten years prior to the current time, and the sub-time periods are each year within those ten years, then each year within those ten years can be assigned weights of 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, and 1.0 respectively, from most distant to most recent.

[0145] For each type of sample attribute information, the impact value corresponding to each sub-time period is obtained. For example, the impact value for each year over ten years. Then, the impact value for each year is weighted and averaged with the current weight, and a functional relationship between vehicle age and weighted impact value is constructed to further obtain the initial estimation function corresponding to the sample attribute information.

[0146] For each stable factor, the valuation is influenced by two factors: vehicle age and stable attributes. To obtain the functional relationship, a value needs to be determined first to obtain the functional relationship between the other two factors. Therefore, the value of the stable attribute is fixed first to determine the functional relationship between vehicle age and the influence value. Thus, here we determine the functional relationship between the vehicle age and the weighted influence value under various values ​​of the stable attribute, obtaining the initial valuation function corresponding to the stable attribute.

[0147] For example, if the sample attribute information is vehicle registration information, the method for constructing the initial valuation function corresponding to the sample attribute information is as follows:

[0148] The impact value and weighted impact value corresponding to the license plate issuance in each sub-time period are calculated, as shown in Table 1:

[0149] Table 1. Impact Values ​​and Weights of Vehicle Registration

[0150] 2021 0.875 0.1 0.0875 2020 0.825 0.2 0.165 2019 0.7875 0.3 0.23625 2018 0.7875 0.4 0.315 2017 0.8625 0.5 0.43125 2016 0.75 0.6 0.45 2015 0.7875 0.7 0.55125 2014 0.825 0.8 0.66 2013 0.7875 0.9 0.70875 2012 0.8375 1.0 0.8375

[0151] In Table 1, the weighted impact value is: sum(mean*weight) / sum(weight) = 0.807727.

[0152] Similarly, determine the impact value corresponding to the absence of license plates in each sub-time period, and use the weights in Table 1 to calculate the impact value corresponding to the absence of license plates in each sub-time period and the weighted impact value of the weights. For example, the result obtained is 1.030.

[0153] Since vehicle registration information is a stable attribute, its impact on vehicle valuation does not change with vehicle age. Therefore, the functional relationship between vehicle age and the weighted impact value is that of two constant functions.

[0154] Figure 3 This is a schematic diagram illustrating the functional relationship between vehicle age and weighted influence value corresponding to the registration information provided in this embodiment of the invention.

[0155] like Figure 3 As shown, the function corresponding to vehicle registration is y = 0.807x, and the function corresponding to vehicle not registered is y = 1.030x. Here, x represents the vehicle's age, and y represents the impact value. This functional relationship is the initial valuation function.

[0156] Figure 4 This is a schematic diagram illustrating the functional relationship between vehicle age and weighted influence value corresponding to license plate attribute information provided in an embodiment of the present invention.

[0157] like Figure 4 As shown, when the license plate attribute is enterprise, the corresponding function is y = 0.970x, and when the license plate attribute is individual, the corresponding function is y = 1.000x.

[0158] Figure 5 This is a schematic diagram illustrating the functional relationship between vehicle age and weighted influence value corresponding to the usage information provided in this embodiment of the invention.

[0159] like Figure 5 As shown, when the usage is in operation, the corresponding function is y = 0.71x, and when the usage is in non-operation, the corresponding function is y = 1.00x.

[0160] Figure 6 This is a schematic diagram illustrating the functional relationship between vehicle age and weighted influence value corresponding to the number of ownership transfers provided in this embodiment of the invention.

[0161] like Figure 6 As shown, when the number of transfers is one, the corresponding function is y = -0.4x; when the number of transfers is two, the corresponding function is y = -0.8x; when the number of transfers is three, the corresponding function is y = -1.2x; and when the number of transfers is four, the corresponding function is y = -1.6x.

[0162] In one possible implementation, the sample attribute information includes optional configuration information, and before setting the weights corresponding to each of the sub-time periods, it further includes:

[0163] Cluster the optional configuration information of each of the sample vehicle valuation data to obtain multiple sets of similar optional configuration information;

[0164] Determining the functional relationship between the vehicle age information and the weighted influence value includes:

[0165] For each group of similar optional configuration information, the functional relationship between the vehicle age information and the weighted influence value is determined.

[0166] In this embodiment of the invention, the optional configuration information includes various types such as adding navigation, adding leather seats, adding electric sunroof, and adding vehicle stability system. Clustering methods can be used to classify the various types of optional configuration information to obtain multiple sets of similar optional configuration information.

[0167] In this way, when constructing functional relationships, corresponding functional relationships are constructed for each group of similar optional configuration information.

[0168] In one possible implementation, determining the second initial valuation function corresponding to each volatility attribute based on the influence value in the volatility attribute valuation data includes the following steps 2034-2035:

[0169] Step 2034: For each of the fluctuation attributes, the value of the fluctuation attribute is used as a feature vector, and the influence value corresponding to the fluctuation attribute is used as a target value. The preset algorithm model is trained to obtain the functional relationship between the fluctuation attribute and the influence value.

[0170] Step 2035: Determine the initial valuation function corresponding to the volatility attribute based on the functional relationship.

[0171] In steps 2034-2035, the impact value refers to the ratio between the average used car transaction price and the average new car transaction price for each car model. For example, for model 001, the average used car transaction price is 90,000 and the average new car transaction price is 100,000, so the impact value is 9 / 10.

[0172] Fluctuation attributes include vehicle age, vehicle condition, mileage, claims, and the month of vehicle transaction. Taking vehicle age as an example, Table 2 is a vehicle age dataset provided in the embodiments of the present invention.

[0173] Table 2 Vehicle Age Dataset

[0174] 001 2 0.92 002 4 0.88 003 1 0.91 004 5 0.86 …… …… ……

[0175] For each vehicle model, inputting the vehicle age dataset from Table 2 into the algorithm model allows for the training of an initial estimation function corresponding to the vehicle age.

[0176] Figure 7 This is a schematic diagram of the initial estimation function corresponding to the vehicle age provided in an embodiment of the present invention.

[0177] Reference Figure 7 As the age of a vehicle increases, the impact of vehicle age on vehicle valuation gradually decreases.

[0178] In one possible implementation, the fluctuation attribute includes vehicle condition information, which includes accident information and maintenance information of the vehicle. Before training the preset algorithm model by using the value of the fluctuation attribute as a feature vector and the influence value corresponding to the fluctuation attribute as a target value, the following steps 301-304 are also included:

[0179] Step 301: Obtain vehicles with the same vehicle age information from the sample vehicle valuation data to obtain a data set of multiple vehicles with the same vehicle age information;

[0180] Step 302: Cluster the accident information of vehicles in the data set of vehicles with the same vehicle age information to obtain multiple sets of similar accident information;

[0181] Step 303: Cluster the vehicle repair information in the accident information set to obtain multiple sets of similar repair information;

[0182] Step 304: Obtain the impact value of the vehicle in each group's maintenance information set.

[0183] In steps 301-304, the vehicle condition has a step-like impact on the vehicle's valuation. Throughout the vehicle's lifecycle, accidents and repairs may occur at different points in time, with the impact weight gradually decreasing over time. Transaction data and repair records within a preset historical time period are acquired, and data on vehicle model, age, parts, accident information, repair information, and impact value are extracted. Parts include hundreds of core automotive components such as the hood, front bumper, windshield, and front fenders; accident types include dozens such as damage, claims, malfunctions, rust, leaks, and oxidation; repair types include dozens such as painting, replacement, sheet metal work, and cutting. Each vehicle's accident and repair information represents one or more of the aforementioned accident and repair types. For each vehicle model, vehicles with the same age, accident information, and repair information are selected, and their impact on the valuation is calculated.

[0184] In one possible implementation, training a preset algorithm model by using the value of the fluctuation attribute as a feature vector and the influence value corresponding to the fluctuation attribute as a target value includes:

[0185] For each set of maintenance information, the vehicle age information, the accident information, and the maintenance information are used as feature vectors, and the impact value of the vehicle in the maintenance information set is used as the target value. A preset algorithm model is trained to obtain the functional relationship between the vehicle age information and the impact value corresponding to the maintenance information set. The functional relationship is a piecewise function with the vehicle age information as the independent variable and the impact value as the dependent variable. Each sub-segment of the piecewise function is a constant function, and the value of the constant function decreases as the vehicle age information increases.

[0186] In this embodiment of the invention, for vehicles with the same age, accident information and maintenance information, the age, accident information and maintenance information are used as feature vectors, and the influence value of the vehicles in the maintenance information set is used as the target value. The preset algorithm model is trained to obtain the functional relationship between the age information and the influence value corresponding to each maintenance information set.

[0187] Figure 8 This is a schematic diagram illustrating the functional relationship between vehicle condition and influence value provided in an embodiment of the present invention.

[0188] like Figure 8 As shown, since each set of maintenance information includes different accident information and maintenance information, the function curves for vehicle age and impact value are different for different sets of maintenance information.

[0189] Similar to stable attributes, for each volatility attribute, the valuation is influenced by two factors: vehicle age and the volatility attribute itself. To obtain the functional relationship, one factor needs to be defined first to determine the functional relationship between the other two. Therefore, we first fix the value of the volatility attribute to determine the functional relationship between vehicle age and valuation (valuation is the influence value in the diagram). Therefore, in Figure 8 In this context, the independent variable is the vehicle's age, and the dependent variable is the influence value.

[0190] Depend on Figure 8 It can be seen that the functional relationship is a piecewise function with vehicle age information as the independent variable and the influence value as the dependent variable. Each sub-segment of the piecewise function is a constant function, and the value of the constant function decreases as the vehicle age information increases.

[0191] In one possible implementation, the fluctuation attribute includes mileage information. For each value of the mileage information, the functional relationship between the vehicle age information and the influence value corresponding to the mileage information value is: a linear function with the vehicle age information as the independent variable and the influence value as the dependent variable, and the slope of the linear function is negative.

[0192] Specifically, as mileage increases, the vehicle's estimated value gradually decreases. Table 3 shows the mileage dataset provided in this embodiment of the invention.

[0193] Table 3 Mileage Dataset

[0194] 001 800 0.93 002 4000 0.89 003 2000 0.91 004 6000 0.87 …… …… ……

[0195] Figure 9 This is a schematic diagram illustrating the functional relationship between mileage and impact value provided in an embodiment of the present invention.

[0196] and Figure 8 Similarly, first fix the value of the fluctuation attribute (i.e., mileage) to determine the functional relationship between vehicle age and the impact value. Figure 9 It can be seen that the functional relationship between the mileage value and the vehicle age and the impact value is a linear function, and the slope of the linear function is negative. That is, the larger the mileage value, the smaller the impact of vehicle age on the vehicle valuation.

[0197] In one possible implementation, the fluctuation attribute is the monthly information of the vehicle transaction, and the functional relationship between the monthly information and the influence value is a broken line function with the monthly information as the independent variable and the influence value as the dependent variable.

[0198] Because sales fluctuate between peak and off-peak seasons depending on the month, the valuation will vary depending on the time of sale. We obtain transaction data from the past five years, extracting data on vehicle model, transaction month, and residual value impact. We divide the year into twelve months and align vehicle transaction data according to the month of the transaction. For a given vehicle model, we calculate the weighted average of its impact on residual value in the same month. The calculation method involves weighting the data from the past five years for the same month, allocating weights according to a range of [0.1, 1.0], to obtain the magnitude of the model's impact on valuation each month.

[0199] Figure 10 This is a schematic diagram illustrating the functional relationship between the month of a vehicle transaction and its impact value, provided in an embodiment of the present invention.

[0200] As can be seen, the functional relationship is a broken line function with month as the independent variable and influence value as the dependent variable.

[0201] Furthermore, throughout the vehicle's lifecycle, claims may arise due to product quality issues with the automaker. This involves acquiring used vehicle transaction data and claim data within a preset historical timeframe, extracting vehicle models, claim items, and impact values. Since the impact of claims on valuation gradually decreases with vehicle age, a predictive algorithm, such as Lasso regression, is used for each vehicle model to train a claim regression model based on claim item features and impact values.

[0202] In summary, the target vehicle valuation model training method provided in this embodiment of the invention introduces multiple valuation influencing factors such as vehicle age, mileage, month, discounts, and vehicle condition, which is beneficial for calculating vehicle valuation from multiple dimensions. Furthermore, by introducing more sample vehicle valuation data, it can better reflect the true valuation of the vehicle and improve the accuracy of the valuation.

[0203] Figure 11 This is a structural block diagram of a vehicle valuation device provided in an embodiment of the present invention.

[0204] like Figure 11 As shown, the vehicle valuation device 400 includes:

[0205] The vehicle model attribute information acquisition module 401 is used to acquire the vehicle model information of the vehicle to be valued and the attribute information related to the valuation of the vehicle to be valued.

[0206] The attribute information determination module 402 is used to determine the attribute category corresponding to each attribute information and to label the attribute information with the attribute category to obtain labeled attribute information; the attribute category includes stable attributes and fluctuating attributes, the stable attributes are attributes whose impact on the valuation of the vehicle remains unchanged over time, and the fluctuating attributes are attributes whose impact on the valuation changes over time.

[0207] Input module 403 is used to input the labeled attribute information and the vehicle model information into the target vehicle valuation model; the target vehicle valuation model includes a valuation function corresponding to each of the vehicle model information, and the valuation function includes a sub-valuation function corresponding to each of the attribute information and a weight corresponding to each of the attribute categories; the target vehicle valuation model is constructed based on vehicle valuation data within a historical time period.

[0208] The valuation result determination module 404 is used to obtain the valuation result of the vehicle to be valued based on the labeled attribute information, vehicle model information, the valuation function and the weight.

[0209] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0210] In another embodiment of the present invention, a device is also provided, the device including a processor and a memory, the memory storing at least one instruction, at least one program, code set or instruction set, the at least one instruction, the at least one program, the code set or instruction set being loaded and executed by the processor to implement the vehicle valuation method described in the embodiment of the present invention.

[0211] In another embodiment of the present invention, a computer-readable storage medium is also provided, wherein at least one instruction, at least one program, code set or instruction set is stored in the storage medium, wherein the at least one instruction, the at least one program, the code set or instruction set is loaded and executed by a processor to implement the vehicle valuation method described in the embodiment of the present invention.

[0212] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0213] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0214] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0215] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A vehicle valuation method, characterized in that, The method includes: Obtain the attribute information of the vehicle to be valued; The attribute category corresponding to each attribute information is determined. The attribute category is determined based on the degree of influence of the attribute information on the vehicle valuation over time. The attribute category includes stable attributes and fluctuating attributes. The stable attribute is the attribute whose influence on the vehicle valuation over time is less than a first threshold. The fluctuating attribute is the attribute whose influence on the vehicle valuation over time is greater than or equal to the first threshold. The attribute information and the attribute category are input into the target vehicle valuation model; the valuation function in the target vehicle valuation model includes sub-valuation functions corresponding to each attribute information and weights corresponding to each attribute category; the target vehicle valuation model is constructed based on vehicle valuation data within a historical time period. The valuation result of the vehicle to be valued is obtained based on the attribute information and the valuation function. The target vehicle estimation model was trained using the following method: Obtain sample vehicle valuation data within a preset historical time period. The sample vehicle valuation data includes: sample attribute information of the vehicle and the influence value of the sample attribute information on the vehicle valuation in each sub-time period. The preset historical time period includes multiple sub-time periods. From the sample vehicle valuation data, the valuation data corresponding to each stable attribute and the valuation data corresponding to each volatility attribute are obtained respectively, resulting in multiple stable attribute valuation data and multiple volatility attribute valuation data. Based on the influence values ​​in the stable attribute valuation data, a first initial valuation function is determined for each of the stable attributes; and based on the influence values ​​in the volatility attribute valuation data, a second initial valuation function is determined for each of the volatility attributes. Using the sample vehicle estimation data, the first initial estimation function and the second initial estimation function are trained respectively to obtain the stable attribute initial model composed of the first initial estimation function of each stable attribute and the fluctuation attribute initial model composed of the second initial estimation function of each fluctuation attribute. The target vehicle valuation model is obtained based on the pre-set weights of the initial model of the stable attribute and the initial model of the volatility attribute.

2. The method according to claim 1, characterized in that, The step of determining the first initial estimation function corresponding to each of the stable attributes based on the influence values ​​in the stable attribute estimation data includes: Set the weight corresponding to each of the sub-time periods; For each of the stable attributes, the weighted average of the influence value and the weight corresponding to each of the sub-time periods is calculated to obtain the weighted influence value; The functional relationship between vehicle age information and weighted influence value is determined for each value of the stable attribute to obtain the initial estimation function corresponding to the stable attribute.

3. The method according to claim 2, characterized in that, The sample attribute information includes optional configuration information, and before setting the weights corresponding to each of the sub-time periods, it also includes: Cluster the optional configuration information of each of the sample vehicle valuation data to obtain multiple sets of similar optional configuration information; The functional relationship between determining the vehicle age information and the weighted influence value includes: For each group of similar optional configuration information, the functional relationship between vehicle age information and the weighted influence value is determined.

4. The method according to claim 2, characterized in that, The step of determining the second initial valuation function corresponding to each volatility attribute based on the influence value in the volatility attribute valuation data includes: For each of the aforementioned fluctuation attributes, the value of the fluctuation attribute is used as a feature vector, and the influence value corresponding to the fluctuation attribute is used as a target value. The preset algorithm model is trained to obtain the functional relationship between the fluctuation attribute and the influence value. The initial valuation function corresponding to the volatility attribute is determined based on the aforementioned functional relationship.

5. The method according to claim 4, characterized in that, The fluctuation attribute includes vehicle condition information, which includes accident information and maintenance information of the vehicle. Before training the preset algorithm model using the value of the fluctuation attribute as a feature vector and the influence value corresponding to the fluctuation attribute as a target value, the following steps are also included: Obtain vehicles with the same vehicle age information from the sample vehicle valuation data to obtain a data set of multiple vehicles with the same vehicle age information; Cluster the accident information of vehicles in the data set with the same vehicle age information to obtain multiple sets of similar accident information; The vehicle repair information in the accident information set is clustered to obtain multiple sets of similar repair information; Obtain the impact value of the vehicle in each group's maintenance information set.

6. The method according to claim 5, characterized in that, The step of using the value of the fluctuation attribute as a feature vector and the influence value corresponding to the fluctuation attribute as a target value to train a preset algorithm model includes: For each set of maintenance information, the vehicle age information, the accident information, and the maintenance information are used as feature vectors, and the impact value of the vehicle in the set of maintenance information is used as the target value. The preset algorithm model is trained to obtain the functional relationship between the vehicle age information and the impact value corresponding to the set of maintenance information. The functional relationship is a piecewise function with the vehicle age information as the independent variable and the influence value as the dependent variable. Each sub-segment of the piecewise function is a constant function, and the value of the constant function decreases as the vehicle age information increases.

7. The method according to claim 4, characterized in that, The fluctuation attribute includes mileage information. For each value of the mileage information, the functional relationship between the vehicle age information and the influence value corresponding to the mileage information value is: a linear function with the vehicle age information as the independent variable and the influence value as the dependent variable, and the slope of the linear function is negative.

8. The method according to claim 4, characterized in that, The fluctuation attribute is the monthly information of the vehicle transaction, and the functional relationship between the monthly information and the influence value is a broken line function with the monthly information as the independent variable and the influence value as the dependent variable.

9. The method according to any one of claims 1-8, characterized in that, The attribute information includes at least the vehicle model information, optional configuration information, license plate information, usage information, and external environment information of the vehicle to be valued.

10. The method according to claim 9, characterized in that, The stable attributes include license plate information and license plate attribute information; the stable attributes include usage information and number of transfers; the stable attributes include national preferential policy information.

11. The method according to claim 9, characterized in that, The usage information in the fluctuation attribute includes vehicle age information, vehicle condition information, mileage information, and claim information. The external policy information in the fluctuation attribute includes the month information of the vehicle transaction.

12. A vehicle valuation device, characterized in that, The device includes: The attribute information acquisition module is used to acquire the attribute information of the vehicle to be valued. The attribute category determination module is used to determine the attribute category corresponding to each attribute information. The attribute category is determined based on the degree of influence of the attribute information on the vehicle valuation over time. The attribute category includes stable attributes and fluctuating attributes. The stable attribute is the attribute whose influence on the vehicle valuation over time is less than a first threshold. The fluctuating attribute is the attribute whose influence on the vehicle valuation over time is greater than or equal to the first threshold. The data input module is used to input the attribute information and the attribute category into the target vehicle valuation model; the valuation function in the target vehicle valuation model includes sub-valuation functions corresponding to each attribute information and weights corresponding to the attribute category; the target vehicle valuation model is constructed based on vehicle valuation data within a historical time period. The valuation module is used to obtain the valuation result of the vehicle to be valued based on the attribute information and the valuation function. The device further includes a training module, the training module being used for: Obtain sample vehicle valuation data within a preset historical time period. The sample vehicle valuation data includes: sample attribute information of the vehicle and the influence value of the sample attribute information on the vehicle valuation in each sub-time period. The preset historical time period includes multiple sub-time periods. From the sample vehicle valuation data, the valuation data corresponding to each stable attribute and the valuation data corresponding to each volatility attribute are obtained respectively, resulting in multiple stable attribute valuation data and multiple volatility attribute valuation data. Based on the influence values ​​in the stable attribute valuation data, a first initial valuation function is determined for each of the stable attributes; and based on the influence values ​​in the volatility attribute valuation data, a second initial valuation function is determined for each of the volatility attributes. Using the sample vehicle estimation data, the first initial estimation function and the second initial estimation function are trained respectively to obtain the stable attribute initial model composed of the first initial estimation function of each stable attribute and the fluctuation attribute initial model composed of the second initial estimation function of each fluctuation attribute. The target vehicle valuation model is obtained based on the pre-set weights of the initial model of the stable attribute and the initial model of the volatility attribute.

13. An electronic device, characterized in that, The electronic device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set, or an instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the vehicle valuation method as described in any one of claims 1-11.

14. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, the at least one program, the code set, or instruction set is loaded and executed by a processor to implement the vehicle valuation method as described in any one of claims 1-11.

Images