KR102282058B1 - System for estimating collateral value of apartment house - Google Patents

Abstract

The present invention relates to a system for estimating collateral value of apartment houses to quickly provide a collateral value of the apartment house. According to the present invention, the system comprises: an estimation server acquiring a real transaction price data group of apartment houses, a housing status data group, and an auction history data group and mapping data sets included in each of the actual transaction price data group, the housing status data group, and the auction history data group on the basis of address information of the apartment houses to generate an apartment house master data group of the apartment houses; and a user terminal acquiring and transmitting the status data of an appraised house of an appraised apartment to be estimated to the estimation server. The estimation server generates an apartment house segment data group from the apartment houses master data group on the basis of the status data of the house to be evaluated; generates a collateral value estimation model on the basis of the apartment house segment data group; uses a plurality of optimization algorithms to generate a plurality of optimization models from the collateral value estimation model; inputs the status data of the house to be evaluated as input values for each of the plurality of optimization models to generate a plurality of candidate housing collateral value estimation data; compares an actual transaction price-based percentile with the plurality of candidate housing collateral value estimation data to select any one of the plurality of candidate housing collateral value estimation data as housing collateral value estimation data; and determines a housing collateral value of the apartment house to be evaluated on the basis of the selected collateral value estimation data.

Description

System for estimating collateral value of apartment house

The present invention relates to a system for estimating the value of collateral for an apartment building, and more particularly, by generating the mortgage value estimation model from the actual transaction price data group of the apartment house, the housing status data group and the auction history data group, the house of the apartment to be evaluated It relates to a system for estimating the value of collateral for an apartment house that can estimate the value of collateral.

The real estate valuation field plays a very important role from the general public who consumes real estate to investors and banks. The real estate market is an area that occupies a very large part in the modern financial market, and is an area in which many people directly and indirectly participate.

In particular, the housing market constitutes an important axis in which all citizens inevitably participate in any form, and in the reality that a significant amount of Korean household assets is concentrated on real estate, correct information on real estate values, especially information on prices, is of little importance. would be very large.

In the price according to the existing approach, there are mainly asking price, published price, actual transaction price, market price, appraised price, and successful bid price.

division Subject Characteristic Limit published price government Individual information for each apartment number is calculated for taxation purposes and can be adjusted through objection (for long-term holding purposes, it is underestimated to reduce property tax, but for short-term holding purposes, we try to increase it) Individual level information is visible but does not reflect real values actual transaction price government It informs the fact of the actual transaction, but provides only limited information (information on the low-floor-middle-floor-high-floor information up to the apartment building) The actual value is reflected, but the amount announced after the transfer of ownership after the balance payment, not at the time of the contract (time lag) quote proprietor In areas with high investment demand, such as areas with high prices per pyeong, the asking price is higher than the market price. There is no clear basis and it fluctuates inconsistently according to government policies and market conditions. Collect quotes and create market price information at an agency quote real estate agent Collects the information produced by the agency and provides market price information at an appropriate level connoisseur appraiser It is the most reliable price information that reflects the characteristics of individual real estate properties and market conditions at the time of appraisal. Based on this, the value of collateral of financial institutions is evaluated (used when calculating loan amount) It is possible to estimate the exact value of an individual object, but incurs appraisal costs in each case. winning bid auction result It is a price that was successfully bid to the general public through a court auction or a public sale, and is generally lower than the general transaction price. In the worst case, the amount the creditor can recover is often insufficient to reflect the value of the property.

Since the general housing type in Korea is a multi-family house (eg, an apartment), it has the advantage of quantifying house prices. In the current real estate market, information on the market price of individual real estate mainly reflects the asking price of the homeowners collected by real estate brokers, so price information based on the actual transaction amount is required. Although the real transaction price database (DB) of the Ministry of Land, Transport and Maritime Affairs uses actual transaction cases, it is difficult to accurately estimate prices when direct comparison examples are lacking.

On the other hand, in order to obtain a loan from a financial institution, there are secured loans that require collateral and credit loans based on credit. A credit loan is a loan provided by a financial company based on a customer's creditworthiness. The loan amount and interest rate are determined in consideration of the customer's economic situation, occupation, transaction matters, family matters, etc.

Credit loans are classified into pure non-guaranteed credit loans and credit loans that establish a guarantee by a credit guarantee company. Non-guaranteed credit loans are true credit loans and are determined according to one's creditworthiness. On the other hand, collateral loans are generally longer than credit loans, have a larger amount, and have a lower interest rate. A lender is a person who holds real estate or assets of economic value from the borrower in case the money is not returned.

In other words, a financial transaction in which a financial company sets up physical collateral and lends money is called a collateralized loan. Here, a financial transaction is a loan with real estate as collateral or a loan with real estate-secured bonds as collateral, and it is one of collateral finance that carries out economic and legal acts according to the financing. When ordinary people purchase a house, they run out of funds, so the shortfall is borrowed from a financial company, and the financial company establishes a mortgage as collateral.

At this time, the financial company approves the loan in response to the collateral value for the real estate. As described above, the market price used in the market is used to evaluate the collateral value or a separate appraisal is performed.

If the market price is used to evaluate the value of collateral, there may be a problem in that a difference between the assessed value of collateral and the actual value of collateral occurs. There is a problem that not only the time until loan approval is delayed, but also the cost of appraisal is added.

To this end, the conventional method of estimating the value of collateral for real estate utilizes the characteristic that the mobility of real estate itself, a unit of goods, is extremely low compared to the general commodity market, and refers to a certain range of clusters including the real estate (hereinafter, “neighboring clusters”). Therefore, it is common to predict the collateral value of the desired real estate.

The conventional real estate collateral value estimation method generally clusters the real estate market in an arbitrary radius or simply clusters the real estate market based on administrative districts. For example, in the case of Seoul, the real estate market in Seoul was clustered based on 25 districts. Or, the real estate areas were clustered simply centered on defense, such as the downtown area, the east area, the west area, the south area, and the north area.

The conventional method of estimating the value of real estate collateral has a limitation in that it does not reflect the properties of individual real estate, such as type, area, year of construction, and the like for each real estate. To solve this problem, there are some cases of clustering the real estate market by reflecting individual properties of real estate through a clustering algorithm, but there is a limitation in that it is impossible to provide consistent reference information because initial points are set randomly every time. .

Korean Patent No. 10-2113083

The present invention provides a system for estimating the value of collateral for an apartment to be evaluated by generating the mortgage value estimation model from the actual transaction price data group, the housing status data group, and the auction history data group of the apartment house to estimate the mortgage value of the apartment to be evaluated. would like to provide

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention not mentioned may be understood by the following description, and will be more clearly understood by the embodiments of the present invention. Moreover, it will be readily apparent that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the claims.

The apartment house collateral value estimation system according to the present invention for solving the above technical problem acquires the real transaction price data group, the housing status data group, and the auction history data group of the apartment house, and the actual transaction based on the address information of the apartment house an estimation server that maps data sets included in each of the tentative data group, the housing status data group, and the auction history data group to generate the apartment house master data group of the apartment house; and a user terminal that acquires and transmits to the estimation server the status data of the apartment to be evaluated of the apartment to be evaluated to be estimated, wherein the estimation server is based on the status data of the apartment to be evaluated from the group of apartment housing master data. Generating a multi-family housing segment data group, generating a mortgage value estimation model based on the apartment house segment data group, and using a plurality of optimization algorithms to generate a plurality of optimization models from the mortgage value estimation model, respectively, , by inputting the housing status data to be evaluated as input values for each of the plurality of optimization models to generate a plurality of housing collateral value estimation candidate data, and comparing the actual transaction price reference percentile and the plurality of housing collateral value estimation candidate data Any one of a plurality of candidate data for estimating the value of a mortgage may be selected as the data for estimating the value of a mortgage, and the mortgage value of the apartment to be evaluated may be determined based on the selected data for estimating the value of a mortgage.

Preferably, the estimation server generates apartment house identification data for each of the data sets based on address information of a data set included in each of the actual transaction price data group, the housing status data group, and the auction history data group, The apartment house master data group may be generated by mapping the data sets of the actual transaction price data group, the housing status data group, and the auction history data group having the same apartment house identification data to each other.

Preferably, the estimation server sets a first extraction condition based on the status data of the house to be evaluated, and extracts the data sets satisfying the first extraction condition from the apartment house master data group to extract the apartment house segment data You can create groups.

Preferably, the estimation server may generate the mortgage value estimation model from the apartment house segment data group using an estimation model generation algorithm.

Preferably, the estimation model generation algorithm is a quantile regression algorithm, a Lasso algorithm, an adaptive Lasso algorithm, a Lasso percentile regression algorithm, and an adaptive Lasso percentile regression algorithm. Quantile Regression) may be any one of the algorithms.

Preferably, the plurality of optimization algorithms may be an Interior Point Method algorithm, a Majorization Minimization algorithm, and an Alternating Direction Method of Multipliers (ADMM) algorithm.

Preferably, the estimation server sets a second extraction condition based on the status data of the house to be evaluated, extracts actual transaction price data that satisfies the second extraction condition from the actual transaction price data group, and collects the actual transaction price data. The actual transaction price percentile data may be generated by sorting and analyzing, and the actual transaction price percentile corresponding to the reference percentile among the actual transaction price percentile data may be set as the actual transaction price reference percentile.

Preferably, the estimation server identifies each of a plurality of value estimation candidate percentiles corresponding to a reference percentile from each of the plurality of mortgage value estimation candidate data, the reference percentile and each of the plurality of value estimation candidate percentiles Based on a first percentile error between the two, any one of the plurality of housing collateral value estimation candidate data may be selected as the housing collateral value estimation data.

Preferably, the estimation server may select the mortgage value estimation candidate data having the minimum first percentile error as the mortgage value estimation data.

Preferably, the estimation server calculates a second percentile error for each percentile between the actual transaction price percentile of the actual transaction price percentile data and the value estimation percentile of the mortgage value estimation data, and based on the second percentile error to determine the value of the home equity.

Preferably, the estimation server may determine a value estimation percentile having the smallest second percentile error as the home equity value.

According to the present invention, by generating the mortgage value estimation model from the actual transaction price data group of the apartment house, the housing status data group, and the auction history data group, and estimating the mortgage value of the apartment house to be evaluated, the apartment to be evaluated is secured It can provide prompt and accurate home equity value during loan review of loans received by

1 is a diagram illustrating a system for estimating the value of a home equity collateral according to various embodiments of the present disclosure.
2 is a block diagram illustrating components of an estimation server of a system for estimating the value of a home equity collateral according to an embodiment of the present invention.
3 is a diagram illustrating an example of an actual transaction price data group used in a system for estimating the value of a home equity collateral according to an embodiment of the present invention.
4 is a diagram illustrating an example of a housing status data group used in a system for estimating the value of a housing collateral according to an embodiment of the present invention.
5 is a diagram illustrating an example of an auction history data group used in a system for estimating the value of a home equity collateral according to an embodiment of the present invention.
6 is a block diagram illustrating components of a user terminal of a system for estimating the value of a home equity collateral according to an embodiment of the present invention.
7 is a diagram illustrating an example of information displayed by a user terminal of a system for estimating the value of a common home mortgage according to an embodiment of the present invention.

Hereinafter, various embodiments of the present invention will be described with reference to the accompanying drawings. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that various modifications, equivalents, and/or alternatives of the embodiments of the present invention are included. In connection with the description of the drawings, like reference numerals may be used for like elements.

In this document, expressions such as "have", "may have", "includes", or "may include" refer to the presence of a corresponding characteristic (eg, a numerical value, function, operation, or component such as a part). and does not exclude the presence of additional features.

In this document, expressions such as “A or B”, “at least one of A or/and B”, or “one or more of A or/and B” may include all possible combinations of the items listed together. . For example, "A or B", "at least one of A and B", or "at least one of A or B" means (1) includes at least one A, (2) includes at least one B; Or (3) it may refer to all cases including both at least one A and at least one B.

Expressions such as "first", "second", "first", or "second" used in this document may modify various elements, regardless of order and/or importance, and may modify one element. It is used only to distinguish it from other components, and does not limit the components. For example, the first user equipment and the second user equipment may represent different user equipment regardless of order or importance. For example, without departing from the scope of the rights described in this document, a first component may be referred to as a second component, and similarly, the second component may also be renamed as a first component.

A component (eg, a first component) is "coupled with/to (operatively or communicatively)" to another component (eg, a second component) When referring to "connected to", it will be understood that the certain element may be directly connected to the other element or may be connected through another element (eg, a third element). On the other hand, when it is said that a component (eg, a first component) is "directly connected" or "directly connected" to another component (eg, a second component), the component and the It may be understood that other components (eg, a third component) do not exist between other components.

The expression “configured to (or configured to)” as used in this document, depending on the context, for example, “suitable for”, “having the capacity to )", "designed to", "adapted to", "made to", or "capable of" . The term “configured (or set up to)” may not necessarily mean only “specifically designed to” hardware. Instead, in some contexts, the expression “a device configured to” may mean that the device is “capable of” with other devices or parts. For example, the phrase "a processor configured (or configured to perform) A, B, and C" refers to a dedicated processor (eg, an embedded processor) for performing the operations, or by executing one or more software programs stored in memory. , may mean a generic-purpose processor (eg, a CPU or an application processor) capable of performing corresponding operations.

Terms used in this document are only used to describe specific embodiments, and may not be intended to limit the scope of other embodiments. The singular expression may include the plural expression unless the context clearly dictates otherwise. Terms used herein, including technical or scientific terms, may have the same meanings as commonly understood by one of ordinary skill in the art described in this document. Among the terms used in this document, terms defined in general dictionary may be interpreted with the same or similar meaning to the meaning in the context of the related art, and unless explicitly defined in this document, have an ideal or excessively formal meaning. not interpreted In some cases, even terms defined in this document cannot be construed to exclude embodiments of this document.

1 is a diagram illustrating a system for estimating the value of a home equity collateral according to various embodiments of the present disclosure.

Referring to FIG. 1 , a system for estimating the value of a home equity collateral 10 may include an estimation server 100 and a user terminal 200 .

The estimation server 100 obtains the actual transaction price data group, the housing status data group, and the auction history data group of the apartment house, and each of the actual transaction price data group, the housing status data group and the auction history data group based on the address information of the apartment house By mapping the data set included in , it is possible to generate a multi-family house master data set of multi-family houses.

In this case, the estimation server 100 may acquire the actual transaction price data group from the Ministry of Land, Infrastructure and Transport server, the housing status data group from the Ministry of Land, Infrastructure and Transport server, and the auction history data group from the data transaction server.

For example, the actual transaction price data group may be information on the actual transaction price of the Ministry of Land, Infrastructure and Transport provided by the Ministry of Land, Infrastructure and Transport, the housing status data group may be basic information on apartment houses provided by the Ministry of Land, Infrastructure and Transport, and the auction history data group may be provided by the data trading company. It may be successful bid history information provided.

The estimation server 100 obtains the actual transaction price data group, the housing status data group, and the auction history data group of the apartment house, and each of the actual transaction price data group, the housing status data group and the auction history data group based on the address information of the apartment house By mapping the data set included in , it is possible to generate a multi-family house master data set of multi-family houses.

Specifically, the estimation server 100 may generate apartment house identification data of each data set based on address information of the data sets included in each of the actual transaction price data group, the housing status data group, and the auction history data group.

Thereafter, the estimation server 100 may generate the apartment house master data group by mapping each data set of the actual transaction price data group, the housing status data group, and the auction history data group having the same apartment house identification data.

The user terminal 200 may be a terminal controlled by a user performing a loan review of a loan applied for a loan to be evaluated using the multi-family house to be evaluated as collateral, and the server 100 for estimating the mortgage value of the multi-family house to be assessed used for the loan review ) can be requested.

To this end, the user terminal 200 may acquire the status data of the house to be assessed of the apartment house to be assessed to be estimated and transmit it to the estimation server 100 .

The user terminal 200 is an electronic device operated by a user, and may receive a control command from the user, output information in response to the control command, and measure various information.

For example, the user terminal 200 may be any one of a tablet, a smart phone, a notebook computer, a wearable device, a PDA, and a personal computer.

The estimation server 100 may generate the apartment house segment data group from the apartment house master data group based on the status data of the house to be evaluated.

To this end, the estimation server 100 sets a first extraction condition based on the status data of the house to be evaluated, and extracts data sets that satisfy the first extraction condition from among the apartment house master data group to generate the apartment house segment data group. can do.

The estimation server 100 may generate a mortgage value estimation model based on the apartment house segment data group.

Specifically, the estimation server 100 may generate a mortgage value estimation model from the apartment house segment data group by using the estimation model generation algorithm.

Here, the estimation model generation algorithm is a percentile regression algorithm, a Lasso algorithm, an adaptive Lasso algorithm, a Lasso percentile regression algorithm, and an adaptive Lasso percentile regression algorithm. ) may be any one of the algorithms.

Thereafter, the estimation server 100 may generate a plurality of optimization models from the mortgage value estimation model by using a plurality of optimization algorithms, respectively.

The estimation server 100 may generate a plurality of housing collateral value estimation candidate data by inputting the evaluation target housing status data as input values of each of the plurality of optimization models.

Here, the plurality of optimization algorithms may be an Interior Point Method algorithm, a Majorization Minimization algorithm, and an Alternating Direction Method of Multipliers (ADMM) algorithm.

Thereafter, the estimation server 100 may compare the actual transaction price reference percentile with a plurality of housing collateral value estimation candidate data and select any one of the plurality of housing collateral value estimation candidate data as the housing collateral value estimation data.

To this end, the estimation server 100 may set a second extraction condition based on the status data of the house to be evaluated, and extract the actual transaction price data satisfying the second extraction condition from the actual transaction price data group.

The estimation server 100 may align and analyze the actual transaction price data to generate actual transaction price percentile data, and set the actual transaction price percentile corresponding to the reference percentile among the actual transaction price percentile data as the actual transaction price reference percentile.

Subsequently, the estimation server 100 may identify each of the plurality of value estimation candidate percentiles corresponding to the reference percentile from each of the plurality of home mortgage value estimation candidate data.

The estimation server 100 estimates any one of the plurality of mortgage value estimation candidate data based on the first percentile error between the reference percentile and the plurality of value estimation candidate percentiles, respectively, the mortgage value estimation candidate data data can be selected.

Specifically, the estimation server 100 may select the mortgage value estimation candidate data having the minimum first percentile error as the mortgage value estimation data.

Through this, the estimation server 100 may determine the mortgage value of the apartment to be evaluated based on the selected mortgage value estimation data.

Specifically, the estimation server 100 may calculate a second percentile error for each percentile between the actual transaction price percentile of the actual transaction price percentile data and the value estimation percentile of the home mortgage value estimation data.

Thereafter, the estimation server 100 may determine the value of the home equity based on the second percentile error.

The estimation server 100 may determine the value estimation percentile having the smallest second percentile error as the mortgage value.

The estimation server 100 may transmit the determined home equity value of the multi-family house to be evaluated to the user terminal 200 .

2 is a block diagram illustrating components of the estimation server 100 of the multi-family mortgage value estimation system 10 according to an embodiment of the present invention, and FIG. 3 is a multi-family mortgage according to an embodiment of the present invention. It is a diagram illustrating an example of air quality information used in the value estimation system 10, and FIG. 4 is an example of infectious disease information used in the apartment mortgage value estimation system 10 according to an embodiment of the present invention. It is the drawing shown.

2 to 4 , the estimation server 100 may include a communication unit 110 , a database 120 , and a processor 130 .

The communication unit 110 may be connected to a communication network to communicate with the Ministry of Land, Infrastructure and Transport server, the data transaction server, and the user terminal 200 using general-purpose communication. To this end, the communication unit 110 may include a general-purpose communication module for performing general-purpose communication.

Here, the general-purpose communication is communication using the Internet network, or as a cellular communication protocol, for example, LTE (Long-Term Evolution), LTE-A (LTE Advanced), CDMA (Code Division Multiple Access), WCDMA (Wideband CDMA), At least one of Universal Mobile Telecommunications System (UMTS), Wireless Broadband (WiBro), and Global System for Mobile Communications (GSM) may be used.

The communication unit 110 may periodically receive the real transaction price data group and the housing status data group of the apartment house from the server of the Ministry of Land, Infrastructure and Transport. Also, the communication unit 110 may periodically receive the auction history data group from the data transaction server.

Here, the actual transaction price data group is a data set associated with the actual transaction price and the actual transaction price of all apartment houses, and may include a plurality of data sets divided by the actual transaction price (transaction amount) information, as shown in FIG. , each of the plurality of data sets includes actual transaction price information, construction year information, transaction year information, transaction month information, transaction date information, apartment complex name information, exclusive area information of the transaction apartment, serial number information of the actual transaction price, and address of the transaction apartment Information (statutory code) may be included.

Here, as shown in FIG. 4 , the housing status data group is a set of data on the status of all apartment houses, and may include a plurality of data sets divided by apartment house code information, and each of the plurality of data sets includes It may include apartment house code information, apartment complex name information, address information (statutory dong address) of the apartment house, heating method information, the same number of information and the number of households information.

Here, the auction history data group is a set of data on the auction history of all apartment houses, and as shown in FIG. 5 , may include a plurality of data sets divided by auction case number information, and each of the plurality of data sets Auction case number, auction classification information, billing amount information, start decision date information, auction item number information, total appraised value information, minimum sale price information, final sale date information, initial sale date information, number of sale information, bid number information , information on the successful bid amount, and address information (statutory dong address) of the apartment for auction.

On the other hand, the communication unit 110 may receive an estimation request signal for requesting estimation of the mortgage value of the multi-family house to be evaluated to evaluate and estimate the value of the mortgage from the user terminal 200 .

In this case, the communication unit 110 may receive from the user terminal 200 the status data of the house to be evaluated of the apartment to be evaluated together with the estimation request signal.

Here, the status data of the house to be evaluated is data on the status of the apartment to be evaluated, and the address information (the lot number address) of the apartment to be evaluated, the apartment name information, the number of households information, the same number information, the heating method information, the corridor type information, the building It may include year information, exclusive area information, and number of stories information.

Meanwhile, the communication unit 110 may transmit the determined home equity value to the user terminal 200 when the mortgage value of the multi-family house to be evaluated is determined by the processor 130 .

As shown in FIGS. 3 to 5 , the database 120 may classify and store an actual transaction price data group, a housing status data group, and an auction history data group, respectively.

Accordingly, when the latest real transaction price data group, housing status data group, and auction history data group are received by the communication unit 110 , the database 120 stores the actual transaction price data group, the housing status data group, and the auction history data group in advance. It may be accumulated or updated and stored.

In addition, the database 120 may store the apartment house master data group generated by the processor 130 .

The database 120 may include a non-volatile memory and a volatile memory, and the database 120 may be implemented in the form of storing information through a cloud service.

The processor 130 may generate the apartment house master data group of the apartment house by mapping the data sets included in each of the actual transaction price data group, the housing status data group, and the auction history data group based on the address information of the apartment house.

At this time, the processor 130 maps the data sets included in each of two or more data groups among the actual transaction price data group, the housing status data group, and the auction history data group based on the address information of the apartment house to map the apartment house master data of the apartment house You can create groups.

In an embodiment, the processor 130 may generate the apartment house master data group of the apartment house by mapping the data sets included in each of the actual transaction price data group and the housing status data group based on the address information of the apartment house.

Hereinafter, when the processor 130 maps the data sets included in each of the actual transaction price data group, the housing status data group, and the auction history data group based on the address information of the apartment house to generate the apartment house master data group of the apartment house to explain about it.

The processor 130 may generate apartment house identification data for each of the data sets based on address information of the data sets included in each of the actual transaction price data group, the housing status data group, and the auction history data group.

That is, the processor 130 generates apartment house identification data of each of the plurality of data sets from address information of the plurality of data sets included in the actual transaction price data group, and address information of the plurality of data sets included in the housing status data group. may generate apartment house identification data of each of the plurality of data sets, and may generate apartment house identification data of each of the plurality of data sets from address information of the plurality of data sets included in the auction history data group.

At this time, when the address information is a statutory dong code, the processor 130 generates the statutory dong code as apartment house identification data, and when the address information is a lot address or street name address formed in text, the address information is converted into a statutory dong code. It can be created with apartment house identification data.

Thereafter, the processor 130 may generate the apartment house master data group by mapping each data set of the actual transaction price data group, the housing status data group, and the auction history data group having the same apartment house identification data.

Meanwhile, the processor 130 may start estimating the mortgage value of the multi-family house to be assessed when the status data of the house to be evaluated is received together with the estimation request signal to the communication unit 110 .

The processor 130 may generate a group of apartment house segment data from the group of apartment house master data based on the status data of the house to be evaluated.

Specifically, the processor 130 sets a first extraction condition based on the status data of the house to be evaluated, and extracts data sets that satisfy the first extraction condition from the apartment house master data group to generate the apartment house segment data group. can

Here, the first extraction condition is whether the statutory dong code corresponding to the address information included in the housing status data to be evaluated and the statutory dong code of the data set of the apartment master data group are the same, and information on the price range per pyeong corresponding to the apartment to be evaluated It may be whether the flat price information of the data set of the apartment house master data group is included.

That is, the processor 130 determines that the statutory dong code of the data set of the apartment house master data group is the same as the statutory dong code corresponding to the address information included in the assessed housing status data, and the price information per pyeong of the data set of the apartment house master data group If 'is included in the price per pyeong range information corresponding to the apartment house to be evaluated, it is determined that the data set satisfies the first extraction condition, and the data set can be extracted to generate the apartment house segment data group.

In this case, the processor 130 may generate the apartment house segment data group by using the K-means clustering algorithm.

The processor 130 may generate a mortgage value estimation model based on the apartment house segment data group.

To this end, the processor 130 may derive and transform information included in each of a plurality of data sets of the apartment house segment data group to be used for generating a mortgage value estimation model.

For example, the processor 130 divides and derives information in which year information, month information, and date information are combined into year information, month information, and date information, respectively, or uses actual transaction price information and supply area information to generate a price per pyeong information can be derived.

In addition, the processor 130 may convert information that is a text type or another type of code type into a numeric variable type, a dummy variable type, and a categorical variable type so that information can be used for generating a mortgage value estimation model.

The processor 130 may generate a mortgage value estimation model from the apartment house segment data group by using the estimation model generation algorithm.

In an embodiment, the estimation model generation algorithm includes a Quantile Regression algorithm, a Lasso algorithm, an Adaptive Lasso algorithm, a Lasso Quantile Regression algorithm, and an Adaptive Lasso percentile regression (Adaptive) algorithm. It may be any one of Lasso Quantile Regression) algorithms.

On the other hand, a percentile regression algorithm, a Lasso algorithm, an adaptive Lasso algorithm, a Lasso percentile regression algorithm, and an adaptive Lasso percentile regression algorithm are the present invention. As an algorithm widely known in the art of , descriptions related to basic equations, basic parameters, and the like will be omitted.

In an embodiment, the processor 130 may generate a mortgage value estimation model from the apartment house segment data group by using an adaptive Lasso quantile regression algorithm as an estimation model generation algorithm.

Through this, the processor 130 calculates a regression coefficient through an adaptive lasso quantile regression algorithm, and information (explanatory variables) indicating that it is significant in the estimation of the value of a home equity among the information of the apartment house segment data group. can be selected.

Thereafter, the processor 130 may generate a plurality of optimization models from the mortgage value estimation model by using a plurality of optimization algorithms, respectively.

The processor 130 may generate a plurality of candidate data for estimating the value of a home equity by inputting the status data of the house to be evaluated as input values of each of the plurality of optimization models.

In an embodiment, the plurality of optimization algorithms may be an Interior Point Method algorithm, a Majorization Minimization algorithm, and an Alternating Direction Method of Multipliers (ADMM) algorithm.

That is, the processor 130 performs each of a plurality of optimization algorithms, an Interior Point Method algorithm, a Majorization Minimization algorithm, and an Alternating Direction Method of Multipliers (ADMM) algorithm from one generated mortgage value estimation model. A plurality of optimization models can be generated respectively by using.

On the other hand, the interior point method (Interior Point Method) algorithm, the maximum minimization (Majorization Minimization) algorithm, and the ADMM (Alternating Direction Method of Multipliers) algorithm is a well-known algorithm in the technical field of the present invention, basic formulas, descriptions related to basic parameters, etc. are omitted let it do

Thereafter, the processor 130 generates the mortgage value estimation candidate data through the optimization model generated using the interior point method algorithm, and through the optimization model generated using the maximum minimization (Majorization Minimization) algorithm. Candidate data for estimating the value of a mortgage may be generated, and the candidate data for estimating the value of a mortgage may be generated through an optimization model generated using an Alternating Direction Method of Multipliers (ADMM) algorithm.

Here, the mortgage value estimation candidate data may be percentile data in which data that can be estimated as the mortgage value of the apartment to be evaluated are sorted based on the percentile. For example, the candidate data for estimating the value of the home equity may be data including a percentile of a percentile having a percentile range of 5% to 95% and a percentile unit of 5%.

Subsequently, the processor 130 may compare the actual transaction price reference percentile with the plurality of housing collateral value estimation candidate data and select any one of the plurality of housing collateral value estimation candidate data as the housing collateral value estimation data.

To this end, the processor 130 may set a second extraction condition based on the status data of the house to be evaluated, and extract the actual transaction price data satisfying the second extraction condition from the actual transaction price data group.

Here, the second extraction condition is whether the statutory dong code corresponding to the address information included in the housing status data to be evaluated and the statutory dong code of the data set of the actual transaction data group are the same, and whether the data set of the actual transaction price data group in the transaction base date range It may be whether the transaction year information, transaction month information, and transaction date information of the transaction data group are included. For example, the transaction reference date range may be two years from the estimated time point.

The processor 130 may align and analyze the extracted actual transaction price data to generate actual transaction price percentile data, and set the actual transaction price percentile corresponding to the reference percentile among the actual transaction price percentile data as the actual transaction price reference percentile.

In an embodiment, the reference percentile may be “50%”, but may be arbitrarily changed and set by the processor 130 .

On the other hand, the actual transaction price percentile data may be percentile data in which the actual transaction price corresponding to the apartment to be evaluated is sorted based on the percentile. For example, the actual transaction price percentile data may be data including a percentile of a percentile having a percentile range of 5% to 95% and a percentile unit of 5%.

Subsequently, the processor 130 may identify each of the plurality of value estimation candidate percentiles corresponding to the reference percentile from each of the plurality of home equity value estimation candidate data.

The processor 130 determines any one of the plurality of mortgage value estimation candidate data based on the first percentile error between the reference percentile and each of the plurality of value estimation candidate percentiles identified from each of the plurality of mortgage value estimation candidate data The candidate data for estimating the value of the mortgage may be selected as the data for estimating the value of the mortgage.

In detail, the processor 130 may select the mortgage value estimation candidate data having the smallest first percentile error as the mortgage value estimation data.

Through this, the processor 130 may determine the mortgage value of the multi-family house to be evaluated based on the selected mortgage value estimation data.

Specifically, the processor 130 may calculate a second percentile error for each percentile between the actual transaction price percentile of the actual transaction price percentile data and the value estimation percentile of the home mortgage value estimation data.

For example, the processor 130 may calculate a second percentile error between the actual transaction price percentile and the value estimation percentile corresponding to 5% of the percentile, and calculate a second percentile error for each percentile for all percentiles. .

Thereafter, the processor 130 may determine the value of the home equity based on the second percentile error.

The processor 130 may determine the value estimation percentile having the smallest second percentile error as the home equity value.

On the other hand, the processor 130 according to another embodiment may determine the upper limit value of the mortgage and the lower limit on the basis of the reference upper and lower percentiles.

Specifically, the processor 130 according to another embodiment calculates the upper limit percentile by summing the reference upper and lower limit percentiles to the percentile corresponding to the value estimation percentile having the minimum second percentile error, and the upper limit in the housing collateral value estimation data The value estimate percentile corresponding to the percentile may be determined as the upper-bounded mortgage value.

In addition, the processor 130 according to another embodiment calculates the lower limit percentile by subtracting the reference upper and lower limit percentiles from the percentile corresponding to the value estimation percentile having the smallest second percentile error, and the lower limit percentile from the home equity value estimation data. A value estimation percentile corresponding to can be determined as the lower bound value of the mortgage.

On the other hand, the processor 130 according to another embodiment collects and generates actual transaction price trend information, actual transaction price history information, and auction successful bid information corresponding to the apartment to be evaluated, and actual transaction price trend information, actual transaction price history information, and auction The communication unit 110 may be controlled to transmit successful bid information to the user terminal 200 .

Meanwhile, the processor 130 according to an embodiment may include one or more of a central processing unit (CPU), an application processor (AP), and a communication processor (CP). there is. The processor 130 may execute an operation or data processing related to control and/or communication of at least one other component of the estimation server 100 .

6 is a block diagram illustrating the components of the user terminal 200 of the multi-family mortgage value estimation system 10 according to an embodiment of the present invention, and FIG. 7 is a multi-family mortgage according to an embodiment of the present invention. It is a diagram illustrating an example of information displayed by a user terminal of a value estimation system.

6 and 7 , the user terminal 200 may include a communication unit 210 , a storage unit 220 , a processor 230 , and a display unit 240 .

The communication unit 210 may be connected to a communication network to communicate with the estimation server 100 using general-purpose communication. To this end, the communication unit 210 may include a general-purpose communication module for performing general-purpose communication. Here, the general-purpose communication is communication using the Internet network, or as a cellular communication protocol, for example, at least one of LTE, LTE-A, CDMA, WCDMA, UMTS, WiBro, and GSM may be used.

The communication unit 210 may transmit to the estimation server 100 the status data of the house to be evaluated together with the estimation request signal.

Also, the communication unit 210 may receive the mortgage value of the multi-family house to be evaluated in response to the estimation request signal from the estimation server 100 .

Also, the communication unit 210 may receive, from the estimation server 100 , actual transaction price trend information, actual transaction price history information, and auction successful bid information corresponding to the apartment to be evaluated.

Various information received by the communication unit 210 may be stored in the storage 220 .

In addition, the storage 220 may store an operation program (eg, an application) programmed so that the user terminal 200 is controlled by the processor 230 .

The storage unit 220 may include a non-volatile memory and a volatile memory, and an operation program may be stored in the non-volatile memory and loaded into the volatile memory to operate.

The processor 230 according to an embodiment may include one or more of a central processing unit (CPU), an application processor (AP), and a communication processor (CP). The processor 230 may execute an operation or data processing related to control and/or communication of at least one other component of the user terminal 200 .

Meanwhile, the display unit 240 may include a display module to display various information.

As shown in FIG. 7 , the display unit 240 may display the mortgage value of the multi-family house to be evaluated, which is received in response to the estimation request signal. In this case, the display unit 240 may display both the upper limit value of the mortgage and the lower limit of the value of the mortgage.

Also, the display unit 240 may display actual transaction price trend information, actual transaction price history information, and auction successful bid information corresponding to the apartment to be evaluated.

Through this, the user of the user terminal 200 can review the mortgage loan by reading the housing collateral value, actual transaction price trend information, actual transaction price history information, and auction successful bid information provided from the estimation server 100 .

So far, the present invention has been focused on preferred embodiments. Those of ordinary skill in the art to which the present invention pertains will understand that the present invention can be implemented in modified forms without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive point of view. The scope of the present invention is indicated in the claims rather than the foregoing description, and all differences within the scope equivalent thereto should be construed as being included in the present invention.

As described above, although the present invention has been described with reference to limited embodiments and drawings, the present invention is not limited thereto, and the technical idea of the present invention and the following by those of ordinary skill in the art to which the present invention pertains. Of course, various modifications and variations are possible within the scope of equivalents of the claims to be described.

10: Estimation system for the value of a home equity mortgage
100: estimation server
200: user terminal

Claims (10)

Acquire the real transaction price data group, the housing status data group, and the auction history data group of the apartment house, and include each of the actual transaction price data group, the housing status data group, and the auction history data group based on the address information of the apartment house an estimation server for generating a multi-family house master data group of the multi-family house by mapping the data set; and
Includes; a user terminal that acquires the status data of the house to be assessed of the apartment house to be estimated to be estimated and transmits it to the estimation server;
The estimation server is
Generate a multi-family house segment data group from the multi-unit house master data group based on the data to be evaluated based on the housing status data, generate a mortgage value estimation model based on the multi-family house segment data group, and use a plurality of optimization algorithms, respectively to generate a plurality of optimization models from the mortgage value estimation model, respectively, and input the assessed housing status data as input values for each of the plurality of optimization models to generate a plurality of mortgage value estimation candidate data, and the actual transaction price A reference percentile is compared with the plurality of candidate data for estimating the value of a home equity, and any one of a plurality of candidate data for estimating the value of a home is selected as the data for estimating the value of the home equity. determine the value of the mortgage for the apartment house;
The estimation server is
A second extraction condition is set based on the current status data of the house to be evaluated, the actual transaction price data satisfying the second extraction condition is extracted from the actual transaction price data group, and the actual transaction price data is sorted and analyzed to determine the actual transaction price percentile. generating data, and setting an actual transaction price percentile corresponding to a reference percentile among the actual transaction price percentile data as the actual transaction price reference percentile;
The estimation server is
Identify each of a plurality of value estimation candidate percentiles corresponding to a reference percentile among each of the plurality of mortgage value estimation candidate data, and a first percentile error between the reference percentile and each of the plurality of value estimation candidate percentiles Based on the plurality of housing collateral value estimation candidate data, any one of the housing collateral value estimation candidate data is characterized in that the selection of the housing collateral value estimation data
A system for estimating the value of a home equity mortgage.
According to claim 1,
The estimation server is
The apartment house identification data of each of the data sets is generated based on the address information of the data sets included in each of the actual transaction price data group, the housing status data group, and the auction history data group, and the apartment house identification data is the same characterized in that the data sets of each of the actual transaction price data group, the housing status data group, and the auction history data group are mapped to each other to generate the apartment house master data group
A system for estimating the value of a home equity mortgage.
◈Claim 3 was abandoned when paying the registration fee.◈ According to claim 1,
The estimation server is
A first extraction condition is set based on the status data of the house to be evaluated, and the data sets satisfying the first extraction condition are extracted from the apartment house master data group to generate the apartment house segment data group. doing
A system for estimating the value of a home equity mortgage.
According to claim 1,
The estimation server is
generating the mortgage value estimation model from the apartment house segment data group using an estimation model generation algorithm;
The estimation model generation algorithm is
Any one of a Quantile Regression algorithm, a Lasso algorithm, an Adaptive Lasso algorithm, a Lasso Quantile Regression algorithm, and an Adaptive Lasso Quantile Regression algorithm. characterized
A system for estimating the value of a home equity mortgage.
According to claim 1,
The plurality of optimization algorithms are
Characterized in that the interior point method (Interior Point Method) algorithm, the maximum minimization (Majorization Minimization) algorithm and ADMM (Alternating Direction Method of Multipliers) algorithm
A system for estimating the value of a home equity mortgage.
delete delete According to claim 1,
The estimation server is
The first percentile error is characterized in that the selection of the mortgage value estimation candidate data with the smallest as the mortgage value estimation data
A system for estimating the value of a home equity mortgage.
According to claim 1,
The estimation server is
Calculate a second percentile error for each percentile between the actual transaction price percentile of the actual transaction price percentile data and the value estimation percentile of the home equity value estimation data, and determine the value of the home equity based on the second percentile error characterized by
A system for estimating the value of a home equity mortgage.
According to claim 1,
The estimation server is
A value estimation percentile having a minimum second percentile error is determined as the value of the home equity
A system for estimating the value of a home equity mortgage.

Images