KR20220166449A - Vehicle For Sale Verification Method Using Neural Network - Google Patents

Abstract

The present invention relates to a method for verifying a vehicle for sale using a neural network, capable of accurately verifying whether a vehicle to be checked corresponds to a false sale. According to embodiments of the present invention, the method for verifying a vehicle for sale using a neural network is performed by a sale vehicle inquiring and searching server. The method for verifying a vehicle for sale using a neural network includes: a data collecting step of receiving input of vehicle data and sale data of the vehicle to be checked from a user; a false sale verifying step of comparing a registration number included in the vehicle data to a false sale database stored in the sale vehicle inquiring server and verifying whether the vehicle to be checked corresponds to the false sale or not; a sale price verifying step of verifying whether a sale price included in the sale data corresponds to a reasonable price range calculated by the neural network; and a verification result providing step of providing the result of the false sale verifying step and the result of the sale price verifying step.

Description

Vehicle sale verification method using neural network {Vehicle For Sale Verification Method Using Neural Network}

The present invention relates to a method for verifying the sale of a vehicle using a neural network, and more particularly, to a method for verifying whether a vehicle for sale corresponds to a false sale and verifying whether a sale price falls within a reasonable price range.

Recently, with the development of artificial intelligence technology, artificial intelligence technologies are being applied to various fields, and methods for generating additional information by extracting features inherent in data through neural network models have been developed and used instead of existing data processing methods.

A neural network model used for artificial intelligence can quickly and accurately detect and recognize features within input data through learning compared to general data processing. In recent years, artificial intelligence technology has been applied not only to tracking and detecting objects, but also to learning past history and deriving current features that reflect future predictions or time-series change information.

Along with the vitalization of used transactions, a used car transaction method is being made in which non-existent products are posted as sales products to attract consumers' attention. Accordingly, it is verified whether the posted products for sale are fake or real. There is a growing demand from consumers about how to do it. In addition, in the case of used products whose value is not fixed according to depreciation or product condition, consumers' needs for appropriate value calculation of used products are also arising, and neural networks are also being applied to these value calculation fields.

In particular, machine learning technologies are being applied to market price calculation in the used car market, which is relatively expensive and has various variables.

For example, Prior Art 1 (Korean Registered Patent Publication No. 10-2153650 (registration date 2020.09.02)) discloses a method of providing a closed-type used car transaction brokerage service for eradicating fake products and prompt transactions using server-based computing or desktop virtualization, there is. However, the method presented in Prior Art 1 is a closed used car transaction brokerage service, which is inconvenient that many unspecified consumers cannot use it, and is forced to upload listing registration data to only one brokerage platform. The downside is that it brings no advantages.

As another example, Prior Art 2 (Korean Registered Patent Publication No. 10-2218287 (registered date 2021.02.16)) proposes a method for predicting used car prices through machine learning. However, the method presented in Prior Art 2 requires various learning data as a separate learning model is used according to conditions such as vehicle grade, model, and manufacturer. It is difficult to continuously acquire a large amount of training data.

Therefore, in order to more accurately verify false listings, it is necessary to review not only the information of the vehicle itself but also which media is being used online. There is a need to develop a neural network that can reflect the change in nature.

One of the various tasks of the present invention is to provide a method for verifying whether a vehicle to be confirmed corresponds to a false offer by comparing vehicle data with a false offer database.

In addition, one of the various tasks of the present invention is to provide a method of verifying whether the website where the vehicle to be confirmed is sold corresponds to a false sale transaction website by comparing the domain name of the website URL.

In addition, one of the various tasks of the present invention is to provide a method of verifying whether the sales price of the vehicle to be confirmed is appropriate after calculating an appropriate price range based on the inspection result of the vehicle using artificial intelligence.

In the vehicle listing verification method using a neural network performed in a vehicle listing inquiry server according to exemplary embodiments of the present invention for solving the above technical problem, the vehicle listing verification method using the neural network is a vehicle of a vehicle to be checked from a user. Data collection step of receiving input data and sales data; a false sale verification step of verifying whether the vehicle to be checked corresponds to a false sale by comparing the license plate number included in the vehicle data with a false sale database stored in the vehicle sale inquiry server; a sales price verifying step of verifying whether the sales price included in the sales data corresponds to an appropriate price range calculated through the neural network; and a verification result providing step of providing a result of the fake product verification step and a result of the sale price verification step.

The sales data may include a URL (Uniform Resource Locator) of a website where the vehicle to be confirmed is sold (hereinafter referred to as a website to be confirmed), and the false sale database is a website where the vehicle for sale is sold (hereinafter referred to as a website to be confirmed). Hereinafter, referred to as a false product transaction website) may include a URL, and the false product verification step compares the URL of the website to be confirmed with the domain name of the URL of the fake product transaction website. It is possible to determine whether the vehicle corresponds to a false sale.

In the data collection step, when the URL of the website to be confirmed is not input from the user, the false product verification step compares the license plate number with the actual product database stored in the vehicle listing search server, so that the vehicle to be confirmed is It can be verified whether or not it corresponds to a fake sale.

The sales price verification step may include: collecting a new car price, vehicle type, and year model included in the vehicle data of the vehicle to be confirmed, and then calculating a first sales price by applying model-based remaining price data to the new car price; calculating a second sales price by applying mileage-based depreciation rate data to the first sales price after collecting the mileage included in the vehicle data of the vehicle to be confirmed; and calculating a third sale price by applying history-based depreciation rate data to the second sale price after searching for the history data of the vehicle to be checked through a used car history information inquiry website stored in the vehicle sale inquiry server. The sales price verification step may be performed by calculating an appropriate price range based on the third sales price and then verifying whether the sales price of the vehicle to be confirmed falls within the appropriate price range.

The lower limit of the fair price range may be 90% of the third sale price, the upper limit of the fair price range may be 110% of the third sale price, and in the sale price verification step, the sale price of the vehicle to be confirmed When it is larger than the lower limit value and smaller than the upper limit value, it can be determined that it corresponds to the appropriate price range.

The first sales price may be calculated by extracting the residual value rate corresponding to the vehicle type and model year of the vehicle to be confirmed from the model year-based residual value rate data, and then multiplying the new vehicle price by the residual value rate.

The second sales price may be calculated by applying a first depreciation rate calculated in correspondence with the annual average mileage calculated based on the year of the vehicle to be confirmed to the first sales price.

The history data may include a special use history of the vehicle to be confirmed, and the third sales price may be calculated by applying a second depreciation rate calculated in response to the existence and type of the special use history to the second sales price. can

The history data may further include an accident history of the vehicle to be confirmed, and the third sales price is calculated by further applying a third depreciation rate calculated in response to the existence and type of the accident history to the second sales price. can

In the sale price verification step, after further collecting inspection information for each vehicle evaluation item of the vehicle to be confirmed, a fourth depreciation rate calculated in response to the inspection information for each vehicle evaluation item is further applied to the third sale price to obtain a fourth The method may further include calculating a sales price, and the sales price verification step may include calculating whether the sales price of the vehicle to be confirmed falls within the appropriate price range after calculating a proper price range based on the fourth sales price. This can be done by verifying whether

The neural network may be learned using learning data in which predicted market prices are labeled in continuous state information according to the positional relevance of the structural features of the vehicle, and the fourth depreciation rate is arranged by arranging the inspection information according to the structural features of the vehicle. It can be calculated by inputting the vectorized state information to a pre-learned neural network.

If it is determined that the vehicle to be confirmed corresponds to a false sale in the false sale verification step, the sales price verification step may be controlled not to be performed, and in the verification result providing step, the confirmation target vehicle is a false sale. If it is determined to be applicable, only the result of the false sale verification step may be provided, and the result of the sale price verification step may not be provided.

According to a vehicle listing verification method using a neural network according to exemplary embodiments of the present invention, vehicle data input from a user is compared with a false listing database stored in a vehicle listing verification server to determine whether a vehicle to be checked corresponds to a false listing. can be verified more accurately.

In addition, since the present invention performs false product verification based on an open online network, it has the advantage that a large number of unspecified consumers can freely use it. Since the verification of the false sale can be performed by checking whether or not, the vehicle to be confirmed can bring an effective advantage to activating the used car transaction in the fake sale.

In addition, since the present invention calculates an appropriate sales price range by using sequential inspection results according to the convenience of mechanics as learning data for checking the condition of the vehicle to be confirmed through artificial intelligence, the market price based on the inspection result of the vehicle can be more accurately can be provided to consumers. Therefore, consumers can further review whether or not the selling price of the vehicle to be confirmed is appropriate according to the sales price verification result in addition to the false sale verification result, thereby increasing the false sale verification effect and determining the proper selling price. can promote the vitalization of used car vehicle sales transactions.

1 is an exemplary diagram illustrating a vehicle listing verification system using a neural network according to exemplary embodiments of the present invention.
2 is a flowchart illustrating a vehicle listing verification method using a neural network according to exemplary embodiments of the present invention.
3 is a flow chart illustrating a false sale verification step according to exemplary embodiments of the present invention.
4 is a flowchart illustrating a sale price verification step according to exemplary embodiments of the present invention.
5 is a flowchart illustrating a vehicle price prediction method using a neural network according to exemplary embodiments of the present invention.

Hereinafter, specific embodiments of the present invention will be described. The detailed descriptions that follow are provided to provide a comprehensive understanding of the methods, devices and/or systems described herein. However, this is only an example and the present invention is not limited thereto.

In describing the embodiments of the present invention, if it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted. In addition, terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of a user or operator. Therefore, the definition should be made based on the contents throughout this specification. Terminology used in the detailed description is only for describing the embodiments of the present invention and should in no way be limiting. Unless expressly used otherwise, singular forms of expression include plural forms. In this description, expressions such as “comprising” or “comprising of” are intended to indicate certain characteristics, numbers, steps, operations, elements, some or combinations thereof, and one or more other than those described. It should not be construed to exclude the existence or possibility of any other feature, number, step, operation, element, part or combination thereof.

Also, terms such as first, second, A, B, (a), and (b) may be used to describe components of an embodiment of the present invention. These terms are only used to distinguish the component from other components, and the nature, order, or order of the corresponding component is not limited by the term.

1 is an exemplary diagram illustrating a vehicle listing verification system using a neural network according to exemplary embodiments of the present invention.

Referring to FIG. 1 , a vehicle listing verification system 1 may include a vehicle listing verification server 100 and a user terminal 200 .

A user, preferably a person requesting confirmation of a vehicle for sale may input vehicle data 210 and sales data 220 using the user terminal 200 .

The vehicle data 210 and sales data 220 input from the user are collected by the data collection unit 110 of the vehicle listing verification server 100 and stored in the data storage unit 120 of the vehicle listing verification server 100. It can be stored, and can be used in the false sale verification step (S2) and the sales price verification step (S3) described later through the false sale verification unit 130 and the sales price verification unit 140.

The vehicle data 210 may include information related to the vehicle itself, such as the vehicle number, year, and mileage of the vehicle to be confirmed, and the sales data 220 is a website where the vehicle to be confirmed is sold (hereinafter, the target website for confirmation). It may include information related to vehicle sales, such as a URL of a site) and a sales price of the vehicle to be checked.

In the data storage unit 120 of the vehicle listing verification server 100, a false product database storing information on a false product transaction website (hereinafter referred to as a false product website) and a real product transaction website (hereinafter referred to as a real product website) site) may include a database of actual products in which information about the product is stored. In one embodiment, the fake product database may include a URL (Uniform Resource Locator) of the fake product website, and the real product database may include the URL of the real product website.

The result of the false sale verification step (S2) and the sale price verification step (S3) performed by the false sale verification unit 130 and the sale price verification unit 140, respectively, are the verification results of the vehicle sale verification server 100 It may be provided to the user through the providing unit 150 .

Hereinafter, a vehicle listing verification method using a neural network according to exemplary embodiments of the present invention will be described in detail with reference to FIGS. 2 to 4 .

2 is a flowchart illustrating a vehicle listing verification method using a neural network according to exemplary embodiments of the present invention.

Referring to FIG. 2 , the vehicle listing verification method using a neural network performed in the vehicle listing inquiry server 100 includes a data collection step of receiving vehicle data 210 and sales data 220 of a vehicle to be checked from a user (S1 ), a false sale verification step (S2) of verifying whether the vehicle to be confirmed corresponds to a false sale by comparing the vehicle number included in the vehicle data 210 with the false sale database stored in the vehicle sale inquiry server 100 ), the sales price verification step (S3) of verifying whether the sales price included in the sales data 220 corresponds to the appropriate price range calculated through the neural network, and the result of the false sale verification step (S2) and the sales price It may include a verification result providing step (S4) of providing a result of the verification step (S4).

The data collection step (S1) may be performed by receiving the vehicle number, year and mileage of the vehicle to be confirmed, the URL of the website to be confirmed, and the sales price of the vehicle to be confirmed from the user.

In one embodiment, when the URL of the website to be confirmed is collected in the data collection step (S1), the false product verification step (S2) is performed by filtering the URL of the website to be checked and the false product database. A process of determining whether the vehicle to be confirmed corresponds to a false sale by comparing each domain name of the URL of the false sale website stored in the website (hereinafter referred to as a URL filtering process) may be included. The URL filtering process will be described later in detail with reference to FIG. 2 .

Meanwhile, the data collection step (S1) may not receive the URL of the confirmation target website from the user, and in this case, the fake product verification step (S2) may not include the URL filtering process.

3 is a flow chart illustrating a false sale verification step according to exemplary embodiments of the present invention.

Referring to FIG. 3, in the false product verification step (S2), the URL of the website to be confirmed is input (S21), the URL filtering process is performed based on the false product database (S22), and the confirmation target It is determined whether the domain name of the website URL is the same as the domain name of the fake website URL (S23), and after performing a property registration inquiry based on the license plate number of the vehicle to be confirmed (S24), the real property It can be performed by filtering registered vehicles for sale based on the database (S25) and determining whether the vehicle number of the vehicle to be confirmed is the same as the vehicle number of the vehicle for sale registered on the website for sale (S26).

In the process of determining whether the URL domain names are the same (S23), if the domain name of the URL of the website to be confirmed is the same as the domain name of the URL of the website for fake products, in the step of verifying false products (S2), the vehicle to be confirmed is It can be determined that it corresponds to a false sale (S28). Unlike this, in the process of determining whether the URL domain names are the same (S23), if the domain name of the URL of the website to be confirmed is different from the domain name of the URL of the fake product website, the false product verification step (S2) Property registration inquiry (S24) may be performed based on the license plate number of the target vehicle.

On the other hand, in the false product verification step (S2), the process of receiving the URL of the website to be confirmed (S21) may correspond to the user's choice, and accordingly, the process of inputting the URL of the website to be confirmed ( S21) may not be performed. In this case, in the fake product verification step (S2), the URL filtering process (S22) and the URL domain name identification process (S23) may not be performed, respectively, and the fake product verification step (S2) is a data collection step. After performing a listing registration inquiry based on the vehicle number of the vehicle to be confirmed collected in (S1) (S24), and filtering the vehicle listing registered based on the actual listing database (S25), It may be performed by determining whether the vehicle number and the vehicle number of the vehicle for sale registered on the website for sale are the same (S26).

The listing registration inquiry process (S24) may be performed by listing the vehicle registration number of the vehicle data 210 collected in the data collection step (S1) on the actual listing website stored in the actual listing database, and vehicle listing filtering. Process S25 may be performed by filtering vehicles for sale registered in the database of actual products based on the license plate number of the vehicle data 210 .

In the process of determining whether the vehicle numbers are the same (S26), if it is confirmed that the vehicle for sale identical to the vehicle number of the vehicle to be confirmed is registered on the actual product website, it is determined that the vehicle to be confirmed corresponds to the actual product (S27) On the other hand, if it is confirmed that the vehicle for sale identical to the license plate number of the vehicle to be confirmed is not registered on the real sale website, it is determined that the vehicle to be confirmed does not correspond to the real sale (S28), that is, it corresponds to a false sale. It can be performed by determining that it is (S28).

In one embodiment, when it is determined that the vehicle to be confirmed corresponds to a false sale in the false sale verification step (S2), the sale price verification step (S3) can be controlled not to be performed, and the verification result providing step ( S4) may be controlled to provide only the result of the false sale verification step (S2) and not to provide the result of the transaction price verification step (S3).

4 is a flowchart illustrating a sale price verification step according to exemplary embodiments of the present invention.

Referring to FIG. 4 , in the sales price verification step (S3), after collecting the new car price, vehicle type, and model year included in the vehicle data 210 of the vehicle to be confirmed, the data storage unit 120 converts the new car price to the new car price. Calculating the first sales price by applying the year-based residual price data stored in (S31), collecting the mileage included in the vehicle data 210, and then the mileage stored in the data storage unit 120 as the first sales price. Calculating the second sales price by applying the base depreciation rate data (S32), after searching the history data of the vehicle to be confirmed through the used car history information search website stored in the vehicle listing inquiry server 100, the second sales price Calculating a third sales price by applying the history-based depreciation rate data stored in the data storage unit 120 (S33), calculating a fourth sales price by applying depreciation rate data based on inspection information for each vehicle evaluation item of the vehicle to be confirmed It may include step (S34), and step (S36) of verifying whether the sales price of the vehicle to be confirmed falls within the appropriate price range after calculating a proper price range based on the fourth sales price (S35). there is.

In the proper price verification step (S36), when it is determined that the selling price of the vehicle to be confirmed is within the appropriate price range, in the selling price verification step (S3), it is determined that the selling price of the vehicle to be confirmed is the appropriate price (S37 )can do. In contrast, when it is determined that the selling price of the vehicle to be confirmed does not fall within the appropriate price range in the fair price verification step (S36), the selling price verification step (S3) determines whether the selling price of the vehicle to be confirmed is an improper price. It can be determined as (S38).

In one embodiment, the lower limit of the fair price range may be 90% of the fourth sale price, the upper limit of the fair price range may be 110% of the fourth sale price, and the sale price verification step (S3) When the selling price of the vehicle to be confirmed is greater than the lower limit value and less than the upper limit value, it may be determined that the vehicle falls within the appropriate price range.

However, the concept of the present invention is not necessarily limited thereto, and the fourth sale price calculation step (S34) may be omitted. In this case, the lower limit of the fair price range may be 90% of the third sale price, and the upper limit of the fair price range may be 110% of the third sale price.

In the first sales price calculation step (S31), the first sales price is calculated by extracting the residual value rate corresponding to the vehicle type and model year of the vehicle to be confirmed from the year-based residual value rate data, and then multiplying the new vehicle price by the residual value rate. can At this time, the year-based residual value rate data may be calculated as shown in Table 1 below (example of residual value rate of non-commercial vehicles).

division useful life 1 year
under 1 year 2 years 3 years 4 years 5 years 6 years 7 years riding domestic 15 years 0.826 0.725 0.614 0.518 0.437 0.368 0.311 0.262 Oesan 15 years 0.842 0.729 0.605 0.500 0.412 0.340 0.281 0.232 ride 15 years 0.810 0.726 0.609 0.510 0.426 0.357 0.298 0.250 freight 15 years 0.761 0.671 0.597 0.510 0.426 0.357 0.298 0.250 division useful life 8 years 9 years 10 years 11 years 12 years 13 years 14 years 15 years riding domestic 15 years 0.221 0.186 0.157 0.132 0.112 0.094 0.079 0.067 Oesan 15 years 0.172 0.142 0.117 0.097 0.080 0.066 0.054 0.050 ride 15 years 0.215 0.184 0.157 0.134 0.113 0.096 0.081 0.067 freight 15 years 0.229 0.200 0.172 0.149 0.128 0.110 0.098 0.086

In the second sales price calculation step (S32), the second sales price may be calculated by applying a first depreciation rate calculated in correspondence with the annual average mileage calculated based on the year of the vehicle to be confirmed to the first sales price. there is.

In one embodiment, when the annual average mileage is less than 15,000 km, the first depreciation rate may not be calculated. In contrast, when the annual average mileage exceeds 15,000 km, the first depreciation rate may be calculated as N%, where N may be a positive integer.

In the third sales price calculation step (S33), the history data may include the special use history and accident history of the vehicle to be confirmed, and the third sales price depends on the existence and type of the special use history for the second sales price. It may be calculated by applying the second depreciation rate calculated in response and the third depreciation rate calculated in correspondence with the existence and type of the accident history.

In one embodiment, when the special use history is not confirmed, the second depreciation rate may not be calculated. In contrast, when the special use history is confirmed for commercial use, the second depreciation rate may be calculated as 15%, and when the special use history is confirmed as a rental car, the second depreciation rate may be calculated as 13% .

In the fourth sales price calculation step (S34), inspection information for each vehicle evaluation item may be collected from an inspector of the vehicle to be confirmed, and the neural network of the vehicle listing inquiry server 100 determines the value of the vehicle to be confirmed based on the input inspection information. It may be controlled to calculate depreciation rate data based on inspection information for each vehicle evaluation item.

5 is a flowchart illustrating a method for calculating depreciation rate data based on inspection information for each vehicle evaluation item using a neural network according to exemplary embodiments of the present invention.

Referring to FIG. 5 , a method for calculating depreciation rate data based on inspection information for each vehicle evaluation item using a neural network includes receiving inspection information for each vehicle evaluation item input from the user (S41), and converting the inspection information to the vehicle to be checked. Calculating depreciation rate data based on inspection information for each vehicle evaluation item by inputting the vectorized state information arranged according to structural characteristics into a pre-learned neural network (S42); It may include a step (S43) of calculating a fourth sales price by applying it to the sales price.

Here, the inspection information can be structured into items classified according to the hierarchy, and the large item is the most comprehensive item and can be composed of a total of five items: exterior, interior and function, engine room and powertrain, wheels and tires, and lower part. .

The middle item may refer to a part or function that exists in a position that is distinguished from each other in the large item. A subitem means a part or component called by a direct name within a subitem, and may actually be a unit inspected by an inspector.

In other words, for each sub-item, the inspector divides the exterior, interior and function of the car, engine and powertrain, wheels and tires, and the lower part of the car excluding the frame into exterior / safety as performance points, and then goes through a series of inspection processes to score according to the condition. can be calculated.

As described above, according to the vehicle listing verification method using a neural network according to exemplary embodiments of the present invention, the vehicle data 210 input from the user is compared with the false listing database stored in the vehicle listing verification server to verify the target vehicle. It is possible to more accurately verify whether or not this corresponds to a false sale.

In addition, since the present invention performs false product verification (S2) based on an open online network, it has the advantage that a large number of unspecified consumers can freely use it. Since verification of a false sale can be performed by checking whether it is a stored brokerage platform, the vehicle to be confirmed can bring an effective advantage to activating used car transactions in a false sale.

In addition, the present invention calculates an appropriate sales price range by using sequential inspection results according to the convenience of mechanics as learning data for checking the condition of the vehicle to be confirmed through artificial intelligence (S34), so that the market price based on the inspection result of the vehicle can be more accurately provided to consumers. Therefore, consumers can further review whether the selling price of the vehicle to be confirmed is appropriate according to the sales price verification result (S3) in addition to the false sale verification result (S2), thereby increasing the false sale verification effect. and promote the vitalization of used car sales transactions at an appropriate selling price.

However, the concept of the present invention is not necessarily limited thereto, and the method for manufacturing rice bread according to exemplary embodiments of the present invention can be applied to various foods in addition to the above-mentioned products.

Although various embodiments of the present invention have been described in detail above, those skilled in the art will understand that various modifications are possible to the above-described embodiments without departing from the scope of the present invention. . Therefore, the scope of the present invention should not be limited to the described embodiments and should not be defined, and should be defined by not only the claims to be described later, but also those equivalent to these claims.

Claims (12)

In the vehicle listing verification method using a neural network performed in a vehicle listing inquiry server,
A data collection step of receiving vehicle data and sales data of a vehicle to be confirmed from a user;
a false sale verification step of verifying whether the vehicle to be checked corresponds to a false sale by comparing the license plate number included in the vehicle data with a false sale database stored in the vehicle sale inquiry server;
a sales price verifying step of verifying whether the sales price included in the sales data corresponds to an appropriate price range calculated through the neural network; and
A vehicle listing verification method using a neural network comprising: a verification result providing step of providing a result of the false sale verification step and a result of the sale price verification step. According to claim 1,
The sales data includes a URL (Uniform Resource Locator) of a website where the vehicle to be confirmed is sold (hereinafter referred to as a website to be confirmed),
The false sale database includes a URL of a website where a false sale vehicle is sold (hereinafter referred to as a false sale transaction website),
The false product verification step compares the domain name of the URL of the website to be confirmed and the URL of the fake product transaction website to determine whether the vehicle to be confirmed corresponds to a false product. Vehicle listing verification method. According to claim 2,
In the data collection step, when the URL of the website to be confirmed is not input from the user, the false product verification step compares the license plate number with the actual product database stored in the vehicle listing search server, so that the vehicle to be confirmed is A vehicle sale verification method using a neural network, characterized in that for verifying whether it corresponds to a false sale. According to claim 1,
The sale price verification step,
After collecting the new car price, vehicle type, and year model included in the vehicle data of the vehicle to be confirmed, calculating a first sales price by applying model-based residual rate data to the new car price;
calculating a second sales price by applying mileage-based depreciation rate data to the first sales price after collecting the mileage included in the vehicle data of the vehicle to be confirmed; and
After searching for the history data of the vehicle to be checked through a used car history information inquiry website stored in the vehicle listing inquiry server, calculating a third sales price by applying history-based depreciation rate data to the second sales price,
The sales price verification step is performed by calculating an appropriate price range based on the third sales price and then verifying whether the sales price of the vehicle to be confirmed falls within the appropriate price range. Vehicle listing verification method. According to claim 4,
The lower limit of the fair price range is 90% of the third sale price, and the upper limit of the fair price range is 110% of the third sale price;
In the sales price verification step, when the sales price of the vehicle to be confirmed is greater than the lower limit value and less than the upper limit value, it is determined that the vehicle falls within the appropriate price range. According to claim 4,
The first sales price is calculated by extracting the residual value rate corresponding to the vehicle type and model year of the vehicle to be confirmed from the model year-based residual value rate data, and then multiplying the new vehicle price by the residual value rate. verification method. According to claim 6,
The second sales price is calculated by applying a first depreciation rate calculated in correspondence with the annual average mileage calculated based on the year of the vehicle to be confirmed to the first sales price, vehicle listing verification using a neural network. method. According to claim 7,
The history data includes a special use history of the vehicle to be confirmed,
The third sales price is calculated by applying a second depreciation rate calculated in response to the existence and type of the special use history to the second sales price. According to claim 8,
The history data further includes an accident history of the vehicle to be confirmed,
The vehicle listing verification method using a neural network, characterized in that the third sales price is calculated by further applying a third depreciation rate calculated in response to the existence and type of the accident history to the second sales price. According to claim 4,
The sale price verification step,
After further collecting inspection information for each vehicle evaluation item of the vehicle to be confirmed, calculating a fourth sale price by further applying a fourth depreciation rate calculated in response to the inspection information for each vehicle evaluation item to the third sale price. include more,
The sales price verification step is performed by calculating a fair price range based on the fourth sale price and then verifying whether the sales price of the vehicle to be checked falls within the fair price range. Vehicle listing verification method. According to claim 10,
The neural network is trained using learning data in which predicted market prices are labeled in continuous state information according to the positional relevance of the structural features of the vehicle,
The fourth depreciation rate is calculated by arranging the inspection information according to the structural characteristics of the vehicle and inputting vectorized state information to a pre-trained neural network. According to claim 1,
When it is determined that the vehicle to be confirmed corresponds to a false sale in the false sale verification step, the transaction price verification step is controlled not to be performed,
In the step of providing verification results, when it is determined that the vehicle to be confirmed corresponds to a false sale, only the result of the verification of the false sale is provided and the result of the verification of the sale price is not provided. Using a neural network, characterized in that Vehicle listing verification method.

Images