JP2001117973A - Device and method for evaluating real estate, recording medium storing real estate evaluation program and real estate information providing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily evaluate the price of real estate property with high accuracy in the case of selling the real estate property or moving. SOLUTION: This device is provided with an evaluation model generating part 200 for generating a price evaluation model for calculating the predictive price of real estate property from inputted property data for each segment composed of one or plural areas having mutually similar area characteristics, an evaluation model storage part 120d for storing the price evaluation model, a segment discriminating part 130 for discriminating the segment, to which the real estate property belongs, on the basis of the property data, an evaluation model selecting part 150 for selecting the evaluation model corresponding to the discriminated segment from the evaluation model storage part, and an evaluating part 160 for calculating the predictive price of the real estate property on the basis of the evaluation model selected by the evaluation model selecting part.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to a real estate evaluation device,
The present invention relates to a real estate evaluation method, a recording medium for storing a real estate evaluation program, and a real estate information providing system. In particular, when selling or replacing real estate such as condominiums or land, price evaluation is performed using an evaluation model built according to regional characteristics, thereby realizing highly accurate real estate property price evaluation and selling or replacing. By efficiently presenting relevant information required for
It relates to technology for supporting the distribution of real estate.

[0002]

2. Description of the Related Art Generally, when a user who owns a real estate such as a condominium, a detached house or land intends to sell or replace this real estate, the user executes the sale according to the following procedure.

FIG. 24 shows a procedure for selling real estate. First, as information necessary for selling the property owned by the user, information on the market price of the property owned by one or more real estate agents and price trends for knowing when to sell is entrusted to one or more real estate agents. (Step S22)
1). Based on the collected information, a real estate agent that brokers the sale is selected (step S223). A brokerage contract is concluded with the selected real estate agent, and an actual assessment of the property to be sold is performed (step S225).

[0004] Next, the intermediary real estate agent sells the property to be sold based on the assessed price (step S22).
7), a sales contract is concluded through price negotiation between the purchase requesting user and the user (step S229). The user purchases and purchases new real estate as necessary using the transfer price obtained in this sales contract.

[0005]

However, the above-mentioned conventional method of selling or replacing real estate has the following problems.

The above information collecting step (step S221)
, When a user obtains a market price or price trend of a property to be sold from a real estate agent, it is not always possible to obtain an objective and valid market price or price trend.
In this information collection, even if different price information was obtained from a plurality of real estate agents, the user himself could not judge which price information should be relied on.

[0007] A so-called "transaction case comparison method" is known as a simple method of assessing real estate.
In this assessment method, in principle, the same type of individual factor is selected from those related to real estate located in a nearby area or a similar area within the same supply and demand area, and this is compared with the condominium to be sold. , Get a valuation price.

However, according to this method, the accuracy of the evaluation decreases in an area where the number of properties owned is small. In particular, when assessing properties for sale belonging to an area with no transaction history, selecting a nearby area or the same supply and demand area, and selecting similar types of individual factors are difficult tasks and require arbitrary Accuracy is reduced due to the tendency to work. In other words, there is a problem that if a standard apartment similar to the apartment to be sold is not set in the same area, a reasonable evaluation price cannot be obtained.

As described above, the present invention has been made to solve the above problems.

An object of the present invention is to provide an evaluation model constructed for each area clustered according to the similarity of regional characteristics at the stage of collecting information for selling or replacing real estate such as condominiums and land. Provide a real estate evaluation device, a real estate evaluation method, a recording medium storing a real estate evaluation program, and a real estate information providing system capable of easily realizing a highly accurate real estate property price evaluation by performing price evaluation using Is to do.

Another object is to realize a more accurate price evaluation by optimizing an evaluation model according to a transaction history.

Another object is to present information useful for sale to a user by presenting an evaluation price in a certain price range and presenting attributes that have contributed much to the calculation of the evaluation price.

Another object is to support the distribution of real estate by efficiently presenting relevant information required for sale or replacement to users and real estate agents.

[0014]

A feature of the present invention for solving the above-mentioned problems is that the areas are classified into clusters based on the characteristics of the areas (areas), and each of the clustered areas is segmented. The point is that an evaluation model for price evaluation is generated for each segment and the target property is evaluated. That is, areas having high homogeneity of regional characteristics are defined as segments, and a price evaluation model is generated for each segment.

In this area clustering, area attribute information composed of indices indicating characteristics of the area;
Line attribute information indicating the rating of the line is used as a factor.

On the other hand, property attribute information indicating properties of the property is used as a parameter for statistical analysis in the price evaluation model.

A first feature of the present invention for realizing such a function is a real estate evaluation device which calculates a predicted price of a given real estate, and calculates a predicted price of the real estate based on input property data. The price evaluation model to be calculated is
An evaluation model generation unit that generates for each segment composed of one or more areas having characteristics of mutually similar areas, an evaluation model storage unit that stores the price evaluation model, and an evaluation target based on the property data. A segment determination unit that determines a segment to which the real estate property belongs; an evaluation model selection unit that selects an evaluation model corresponding to the segment determined by the segment determination unit from the evaluation model storage unit; and the evaluation model selection unit. An evaluator for calculating a predicted price of the real estate based on the selected evaluation model is provided.

Another feature of the present invention is that the evaluation model generation unit calculates the similarity between the respective areas based on area attribute information indicating the attribute of the area to which the real estate belongs, and determines the given area group. An area classification unit that obtains a segment group by clustering according to the calculated similarity; and, for each segment obtained by the area classification unit, statistics based on property attribute information indicating an attribute of a real estate property belonging to the segment. A model generation unit that generates the evaluation model by analysis.

Another feature of the present invention is that the statistical analysis is based on a stratified multiple regression analysis.

According to another feature of the present invention, the area classifying section further clusters the clustered segment group based on railroad attribute information indicating the rating of the railroad to obtain a second segment group. , The model generation unit generates the evaluation model corresponding to the second segment group.

Another feature of the present invention is that the area attribute information includes at least a population index relating to a population of a unit area,
It includes at least one of a productivity index indicating productivity per population, a consumption index indicating consumption index per unit, and a cultural index indicating culture degree per unit.

Another feature of the present invention resides in that the railroad attribute information includes a railroad ranking index indicating a unit price per tsubo.

According to another feature of the present invention, the property attribute information includes at least convenience information indicating convenience of transportation of the real estate, maintenance information indicating a degree of maintenance of the real estate, equipment of the real estate. Equipment information indicating specifications, city infrastructure information indicating the gas and water laying status of the real estate, unit price information indicating the unit price of the land parcel to which the real estate belongs, constraint information indicating the legal constraints of the real estate, housing Alternatively, it includes at least one of structural information indicating the area or structure of the land and seller condition information given by the selling entity of the real estate.

According to another feature of the present invention, the real estate evaluation apparatus further comprises a sales history storage unit for storing a history of past real estate sales information, and an optimization model based on the sales information history. And an evaluation model optimizing unit.

Another feature of the present invention is that the real estate evaluation device further presents the predicted price calculated by the evaluation unit in a price range of a predetermined width and includes the predicted price in the property attribute information. An evaluation information presentation unit that presents some property attribute information that relatively contributed to the calculation of the predicted price is provided.

Another feature of the present invention is that the evaluation information presenting unit further presents the predicted price as a change over time.

Another feature of the present invention is a real estate valuation method for calculating a predicted price of a given real estate, wherein a price evaluation model for calculating a predicted price of a real estate based on input property data is mutually exchanged. An evaluation model generating step of generating for each segment comprising one or more areas having similar area characteristics; a segment determining step of determining a segment to which a real estate property to be evaluated belongs based on the property data; An evaluation model corresponding to the segment determined by the segment determination step; an evaluation model selecting step of selecting from the evaluation models; and a predicted price of the real estate based on the evaluation model selected by the evaluation model selecting process. And a valuation step of calculating a real estate.

Another feature of the present invention is that, in the evaluation model generating step, the similarity between each area is calculated based on area attribute information indicating the attribute of the area to which the real estate belongs, and the given area group is calculated. An area classification step of obtaining a segment group by clustering according to the calculated similarity; and, for each segment obtained by the area classification step, statistics based on property attribute information indicating an attribute of a real estate property belonging to the segment. And a model generating step of generating the evaluation model by analysis.

Another feature of the present invention is that the statistical analysis is based on a stratified multiple regression analysis.

Another feature of the present invention is that the area classification step further includes the step of:
The second segment group is obtained by further clustering based on the alongside attribute information indicating the rating of the wayside, and the model generating step is to generate the evaluation model corresponding to the second segment group.

Another feature of the present invention is that the area attribute information includes at least a population index relating to a population of a unit area,
It includes at least one of a productivity index indicating productivity per population, a consumption index indicating consumption index per unit, and a cultural index indicating culture degree per unit.

Another feature of the present invention resides in that the railroad attribute information includes a railroad ranking index indicating a unit price per tsubo.

According to another feature of the present invention, the property attribute information includes at least convenience information indicating the convenience of transportation of the real estate property, maintenance information indicating a degree of maintenance of the real estate property, and equipment of the real estate property. Equipment information indicating specifications, city infrastructure information indicating the gas and water laying status of the real estate, unit price information indicating the unit price of the land parcel to which the real estate belongs, constraint information indicating the legal constraints of the real estate, housing Alternatively, it includes at least one of structural information indicating the area or structure of the land and seller condition information given by the selling entity of the real estate.

According to another feature of the present invention, the real estate evaluation method further comprises: a sales history storage step of accumulating a history of past real estate sales information; and optimizing the evaluation model based on the sales information history. And an optimization model optimization step.

Another feature of the present invention is that the real estate evaluation method further comprises presenting the predicted price calculated in the evaluation step in a price range of a predetermined width,
The present invention is characterized by including an evaluation information presenting step of presenting a part of the property attribute information which relatively contributed to the calculation of the predicted price in the property attribute information.

Another feature of the present invention is that the evaluation information presenting step further presents the predicted price as a change over time.

Another feature of the present invention is a computer-readable recording medium for storing a program for causing a computer to execute a real estate evaluation process for calculating a predicted price of a given real estate, wherein A price evaluation model that calculates the predicted price of a real estate property,
An evaluation model generation process that generates for each segment composed of one or more areas having characteristics of mutually similar areas; and a segment determination process that determines a segment to which a real estate property to be evaluated belongs based on the property data. An evaluation model corresponding to the segment determined by the segment determination process, an evaluation model selection process of selecting from the evaluation models; and an evaluation model selected by the evaluation model selection process, based on the evaluation model selected by the evaluation model selection process. It is another object of the present invention to provide a computer-readable recording medium including an evaluation process for calculating a predicted price.

Another feature of the present invention is that, in the evaluation model generation processing, similarity between each area is calculated based on area attribute information indicating an attribute of an area to which the real estate belongs, and a given area group is calculated. Area classification processing for obtaining a segment group by clustering according to the calculated similarity,
For each segment obtained by the area classification process,
And a model generation process of generating the evaluation model by statistical analysis based on property attribute information indicating the property of the real estate property belonging to the segment.

Another feature of the present invention is that the statistical analysis is based on stratified multiple regression analysis.

Another feature of the present invention is that, in the area classification processing, the clustered segment group is further clustered on the basis of railroad attribute information indicating the rating of the railroad to obtain a second segment group. , The model generation processing is to generate the evaluation model corresponding to the second segment group.

Another feature of the present invention is that the area attribute information includes at least a population index relating to a population of a unit area,
It includes at least one of a productivity index indicating productivity per population, a consumption index indicating consumption index per unit, and a cultural index indicating culture degree per unit.

Another feature of the present invention resides in that the railroad attribute information includes a railroad rating index indicating a unit price per tsubo.

According to another feature of the present invention, the property attribute information includes at least convenience information indicating convenience of transportation of the real estate, maintenance information indicating a maintenance level of the real estate, equipment of the real estate. Equipment information indicating specifications, city infrastructure information indicating the gas and water laying status of the real estate, unit price information indicating the unit price of the land parcel to which the real estate belongs, constraint information indicating the legal constraints of the real estate, housing Alternatively, it includes at least one of structural information indicating the area or structure of the land and seller condition information given by the selling entity of the real estate.

Another feature of the present invention is that the computer-readable recording medium further includes a sale history storage process for accumulating a history of past real estate sale information, and the evaluation based on the sale information history. And an evaluation model optimization process for optimizing the model.

Another feature of the present invention is that the computer-readable recording medium further presents the predicted price calculated by the evaluation processing in a price range having a predetermined width, and further displays the property attribute information of the property attribute information. The present invention is characterized in that it includes an evaluation information presenting process of presenting a part of the property attribute information that relatively contributed to the calculation of the predicted price.

Another feature of the present invention is that the evaluation information presenting process further presents the predicted price as a change over time.

Another feature of the present invention is a real estate information providing system for calculating a predicted price of a given real estate article and providing real estate information to a user and a real estate agent through the predicted price. An evaluation model generation unit that generates a price evaluation model for calculating a predicted price of a real estate property by property data for each segment including one or a plurality of areas having characteristics of mutually similar areas, A segment determining unit that determines a segment to which the real estate to be evaluated belongs; an evaluation model selecting unit that selects an evaluation model corresponding to the segment determined by the segment determining unit from the evaluation models; Calculate the predicted price of the real estate based on the evaluation model selected by the model selection unit A real estate evaluation device comprising: a real estate evaluation unit having a real estate evaluation unit, a real performance evaluation unit for determining whether or not there is a real performance in an area to which the real property data to be evaluated belongs or in an adjacent area; A real estate agent information storage unit that stores real estate agent information relating to the agent, and a real estate agent that has been determined to have the handling record is searched from the real estate agent information storage unit, and the real estate agent information that is a list of the searched real estate agents And a real estate agent information presenting section for presenting information, and introduction information distribution for giving the evaluation price information output by the evaluation section of the real estate evaluation device to the introduction request information of the specified real estate agent and distributing it to the specified real estate agent. In providing a real estate information providing system characterized by comprising a real estate agent requesting device comprising That.

[0048]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of the present invention will be described with reference to FIG.
This will be described in detail with reference to FIGS.

First Embodiment Hereinafter, a first embodiment of the present invention will be described in detail with reference to FIGS.

In the first embodiment, a price evaluation model is generated for each of the clustered areas (hereinafter, referred to as segments) according to the similarity of the characteristics of the area (area), and the target model is generated based on the price evaluation model. Provides a function to calculate the estimated price (evaluation price) of a property.

FIG. 1 is a block diagram showing the overall functional configuration of the real estate information providing system 1 according to the embodiment of the present invention. As shown in FIG. 1, a real estate information providing system 1 according to an embodiment of the present invention includes a real estate evaluation device 10 that calculates an evaluation price of the real estate based on data of the real estate input by the relocation request user 40. , Based on the property evaluation information calculated by the real estate evaluation device 10,
It is provided with a real estate agent request device 30 that provides real estate agent information to the user 40 who wants to move and also provides real estate agent 50 with sale requester information. Data is transmitted and received between the real-estate information providing system 1 and the user 40 or the real-estate agent 50 who wants to relocate via the Internet by various communication means such as Web, facsimile or mail. The real estate information providing system 1 includes a property information database 70 that holds a history of information on a property requested to be evaluated or a property actually traded, user identification information and a real estate agent 5
0, using the user / contractor identification information having information on the renovation contractor 60a, the moving contractor 60b, etc., to provide the required evaluation information of the property to be sold, and mediate between the user 40 and each contractor. To provide real estate agent information or client information.

FIG. 2 is a block diagram showing a detailed functional configuration of the real estate evaluation device 10 according to the first embodiment. The real estate evaluation device 10 includes an input unit 110 and a preprocessing unit 120
And an area segment determination unit 130, a alongside line segment determination unit 140, a price evaluation model selection unit 150, a price calculation unit 160, a property evaluation information display unit 170, and a price evaluation model generation unit 200. Is done.

The input unit 110 is a user 4 who wants to move.
Input and analyze the property evaluation request from 0. The preprocessing unit 120 loads each database such as the sale case database 120a, the published price base land price database 120b, and the nearby rental database 120c used for the evaluation of the analyzed property and the generation of the evaluation-related information into the work storage area. Initialize the required work area. The area segment determination unit 130 determines an area segment (described later) of a region to which the input property to be evaluated belongs, and outputs the area segment to the alongside line segment determination unit 140. The alongside line segment determination unit 140
It determines the alongside segment to which the input evaluation object belongs and outputs the determined alongside segment (described later) to price evaluation model selecting section 150. The price evaluation model selection unit 150 selects a price evaluation model corresponding to the input roadside segment of the evaluation target property by searching the price evaluation model database 120d. The price calculation unit 160 calculates an evaluation price of the input evaluation target property based on the searched price evaluation model.

The price evaluation model generation unit 200 clusters regions (areas) having similar characteristics to each other,
By further classifying the clustered areas according to the ratings along the railroad, a price evaluation model that models a property for each area segment, which is a group of areas to which areas having high similarity belong, is generated.

As shown in FIG. 2, the real estate evaluation apparatus 10 further includes a property located in the vicinity of the property input as required from the sale case database 120a, the published price base land price database 120b, and the neighborhood rental database 120c. May be provided with a neighborhood sale case search unit 180a, a neighborhood published price search unit 180b, and a neighborhood rental case search unit 180c for respectively searching for information on the above. Alternatively, these search units 180a, 180b and 180
c may be configured inside the property evaluation information display section 170.

Next, a procedure for generating a price evaluation model used for evaluating the input property will be described.

FIG. 3 shows the price evaluation model generator 20 of FIG.
FIG. 2 is a block diagram showing a detailed functional configuration of the first embodiment. The price evaluation model generation unit 200 includes an area clustering unit 210
, A railroad cluster unit 220, a price evaluation model creation unit 230, and a price evaluation model optimization unit 240. First, the area clustering unit 210 determines, for example, the similarity of each administrative city and ward group from area statistical data (not shown) and the like, and area attribute information (described later) of each area.
Is calculated by calculating the distance between the respective areas based on. Various generally known methods can be used for the similarity calculation, but in the first embodiment,
For example, the k-means method (SUGI-J98 “Clustering method for large-scale category data”, Sanpei Nakabayashi) can be used. This k-means method is:
This is a clustering method using an attribute variable with an interval scale, and is suitable for clustering large-scale data because of its high computational efficiency. Areas having a calculated distance between the areas are determined to be similar areas, and the administrative city area group is classified into a plurality of areas smaller than the total number of the administrative city area groups using a predetermined threshold value. . The unit of the area is
The unit is not limited to the administrative city block, but may be a unit larger or smaller than this.

Along the wayside clustering section 220, the areas classified by the area clustering section 210 are further classified based on the wayside attribute information indicating the rating of the wayside. This classification can be performed, for example, by a generally known classification tree method. The area classified by the cluster classification unit 220 is called an area segment. The area segment is a group composed of one or more areas having higher similarity to each other, that is, higher homogeneity than the areas classified by the area clustering unit 210.

The price evaluation model creation section 230 generates a price evaluation model based on property attribute information (described later) for each of the segments classified by the railroad cluster section 220, that is, for each area having a high similarity to each other. To build. For construction of this price evaluation model, for example, a generally known stratified multiple regression method (for example, disclosed in Sannomiya University Press, “Multivariate Analysis for Easy Marketing”) can be used. The price evaluation model created by the price evaluation model creation unit 230 is registered in the price evaluation model database 120d, and is referred to by the price calculation unit 160.

The price evaluation model optimizing unit 240 stores in the price evaluation model database 120d based on past property transaction information in the property information database 70 of FIG. 1 and temporal elements obtained from the corresponding property information. Monitor the price valuation model that is used and optimize it to a more accurate price valuation model that is more suited to the current market conditions.

Next, the hardware configuration of the real estate information providing system according to the first embodiment will be described. The real estate information providing system according to the present invention includes a workstation,
It is mounted on a single computer such as a general-purpose computer, a PC, and various portable information terminals, or on a network system interconnecting the computers.

Each computer has a CPU, a data memory, a program memory, a communication interface,
An input unit and an output unit are provided. Further, it may be connected to an external storage device. The CPU realizes the first embodiment by controlling software for realizing the real estate evaluation device, the real estate evaluation method, and the real estate information providing system.

A program for implementing various processes of communication address management and communication address control according to the present invention can be stored in various recording media. The present invention is carried out by reading out such a recording medium by a CPU of a computer having the above hardware and executing the program. Here, the recording medium is
For example, semiconductor memory, magnetic disk (floppy disk, hard disk), optical disk (CD-ROM,
DVD, etc.) and all devices capable of recording programs. Further, the program may be distributed through various communication means such as a network.

The first embodiment of the present invention is configured as described above, and the flow of the processing will be described below in order with reference to FIGS.

First, an outline of the price valuation model generation processing in the first embodiment will be described.

FIG. 4 is a diagram for explaining the concept of a method of constructing a price evaluation model. For example, based on regional statistical information and the like, the entire administrative municipalities are classified into groups (area segments) of areas having high homogeneity in the characteristics of the regions. In FIG. 4, all administrative city groups are classified into, for example, seven areas. Next, the dwelling units, such as condominiums, land, and detached houses, in the classified areas are further classified into area segments, which are geographically highly similar groups along the line, based on line attribute information indicating the grade of the line. I do. In FIG. 4, 7 × N segments are generated. For each of these generated segments (area segments), a group of dwelling units belonging to each segment is analyzed by the stratified multiple regression analysis to create a price evaluation model.

FIG. 5 is a flowchart showing a procedure of the area clustering process performed by the area clustering section 210. First, data of a group of administrative wards is input (step S50). Based on the area attribute information, the input data of the administrative city / ward group is converted into the K-means.
(Step S5)
1). This regional attribute information can be derived using regional statistical data such as a commonly known "Private Power CD-ROM" (Asahi Shimbun).

The area cluster setting in the area clustering unit 210 is performed for the purpose of reflecting the characteristics of the area such as the living environment of the area where the property is located in the price evaluation model. Here, the characteristic of the area refers to a qualitative characteristic expressed as a “rating” of the area, for example, a luxury residential area. The elements expressing such characteristics include, for example, the following items.

A.) Regional “status” b.) Overall living environment, cityscape c.) Neighborhood residents' lifestyle or characteristics These factors are generally qualitative and impaired when compared with attribute items that express the property itself. Since it is a clear element, it is not easy to directly incorporate it into the price evaluation model as a parameter such as an explanatory variable of a regression equation. For this reason,
In the first embodiment, the characteristics of these areas (here,
Area attribute information and alongside line attribute information) are not incorporated as parameters in the price evaluation model, but by building a price evaluation model in regional units that are positioned in the same case, the qualitative element of the region's case is converted to the price of the real estate. To reflect. The area defined by the cluster analysis of “areas of the same rank” is the area segment. The input of the area clustering process is not limited to the data of the administrative city and ward group unit. If there is a means for obtaining more detailed regional characteristics of the regional unit, the data of the regional unit may be input.

As the data used for the clustering process, basically, factors that are likely to affect real estate prices such as land prices and rental rents are selected. However, due to the nature of the above area clustering process, qualitative data expressing the “grade” of the region, specifically, regional statistical data, and the like are more desirable than data that directly affects the real estate price. The regional statistical data includes, for example, population growth, old-age / young-age population ratio, degree of public facilities, income level index, and the like.
In addition, from the viewpoint of maintenance of the price evaluation model, it is desirable that the regional statistical data used for the analysis be as easily available as possible.

As the clustering technique in the first embodiment, any technique for calculating some similarity, usually expressed as a distance, between objects can be used. In this clustering method, the number of generated clusters is determined by the number of clusters including a clustering target. As typical cluster types, there are a non-hierarchical cluster in which an object is classified into only one cluster, and a hierarchical cluster in which one cluster can be a subset of another cluster. In the first embodiment, the above-described k-means method will be described as an example. The k-means method is a clustering method using an attribute variable having an interval scale, and is a method suitable for clustering large-scale data because of high calculation efficiency. The distance used as the reference for division is defined by the Euclidean distance, and the generated cluster is a non-hierarchical cluster.

By performing clustering using the k-means method, classified administrative city blocks are obtained (FIG. 5).
Step S52). In clustering,
Appropriate division can be performed without increasing the number of clusters by excluding in advance the municipalities in which the number of properties generated is small and it is not meaningful to set the clusters in the first place. 1 according to the first embodiment
According to one experimental example, when the analysis target is 291 municipalities, first, it is divided into 30 clusters. Of the divided clusters, 109 municipalities belonging to the cluster which is considered to generate a small amount of properties are excluded. Clustering is performed again on the remaining 182 municipalities, and the final number of clusters is set to, for example, seven. However,
The number of clusters is not limited to this example, and the final number of clusters needs to be appropriately determined in consideration of the number of samples (that is, the volume of generated properties) in constructing the price evaluation model. For the final adjustment of the number of clusters, for example, municipalities that form a single cluster can be merged into the nearest cluster (nearest neighbor cluster), or a cluster with a large property volume can be subdivided. .

Next, an F value is determined for the clustered administrative city blocks (step S53). This F
The value is an index for evaluating the likelihood of clustering, is called a pseudo F statistic, and is expressed by the following equation (1).

Pseudo F Statistic = R2 / (c-1) / (1-R2 / (nc)) (Equation 1) where R2 is a coefficient of determination calculated by adjusting the number of clusters and the number of observations. Then, the obtained value for the whole, c is the number of clusters, and n is the number of observation targets.

The F value is preferably as large as possible. The coefficient of determination R2 is an index for evaluating the suitability of the model,
For example, when the 294 municipalities are divided into 294 clusters, the determination coefficient R2 becomes 1. The distance between a certain cluster and the closest cluster is represented by the distance between the center of gravity of the cluster and the center of gravity of the closest cluster. Further, the average value of the area attribute information (explanatory variable) used for clustering for each cluster serves as an index expressing the characteristics of the generated cluster. For example, when the elderly ratio is used as one of the explanatory variables (area attribute information), a cluster having a high average elderly ratio can be evaluated as “a cluster having a characteristic of a high elderly ratio”. This average value can be used as a reference when assigning new data to any cluster.

The processes of steps S51 and S52 are repeated until the above-mentioned F value exceeds a predetermined threshold value corresponding to the number of clusters (step S53N). On the other hand, if the F value exceeds the predetermined threshold (step S53Y), it can be evaluated that an appropriately classified administrative city group has been obtained (step S54). If there is a cluster that has only the number of properties less than the minimum number of properties among the generated clusters (step S55Y), the cluster is integrated with the closest cluster described above (step S5).
6). On the other hand, when there is no cluster that holds only the number of properties less than the minimum number of properties (step S55N), the cluster at this stage is set as a finally classified cluster group (step S57).

Next, the procedure of the along-line cluster processing will be described.

FIG. 6 is a flow chart showing a procedure of the classification along the railway performed by the railway cluster 220 of FIG.

First, a property group belonging to the finally classified administrative city / ward group (area segment) obtained in step S57 of FIG. 5 is input (step S60). The property group is classified by the classification tree method based on the "rating" along the line to which each property belongs (step S61). The ranking information along the railroad, that is, the attribute information along the railroad, is evaluated not as a factor expressing the characteristic of each property but as a qualitative and unclear element surrounding the property like the above-mentioned area cluster. Specifically, it is expressed by creating each cluster based on the average square meter unit price of each route.
A price valuation model is created for each of the alongside grades.

The classified property group is obtained by the classification based on the alongside rating (step S62). If the number of properties in the group of properties after this classification is less than the minimum number of properties (step S63N), steps S61 and S62 are repeated. On the other hand, when the number of properties is equal to or greater than the minimum number of properties (step S63Y), the property is evaluated as a properly classified property group (step S64).

Next, the procedure of the price evaluation model creation process will be described.

FIG. 7 is a flowchart showing a procedure of a price evaluation model generation process corresponding to each of the property groups clustered by the roadside clustering process.

First, the property group classified as the alongside cluster in step S64 of FIG. 6 is input (step S7).
0). The input property group is further classified by, for example, the stratified factor method (step S71). The cluster classification based on the stratification factors may be omitted as appropriate. In this case, a price evaluation model is generated for each area segment obtained by clustering by the wayside clustering unit 220, and significant items of information extracted from the property attribute information are internal parameters in the evaluation model, which will be described later. Used as an explanatory variable for regression analysis.

According to the stratification factor method, a group of articles classified is obtained (step S72). If the number of property groups held in each of these clusters is less than the minimum number of properties (step S73N), steps S71 and S72 are repeated. On the other hand, when the number of the property groups is equal to or more than the minimum number of properties (step S73Y), the property group can be evaluated as a properly classified property group (step S7).
4). A price evaluation model is created using multiple regression analysis for each property group classified in step S74 (step S7).
5) The created price evaluation model is output to the price evaluation model database 120d (step S76).

Next, a procedure of an automatic updating process according to a change in market conditions after the price evaluation model is created will be described.

FIG. 8 is a flowchart showing the procedure of price evaluation model monitoring and optimization processing performed by price evaluation model optimization section 240.

First, when a new property is input (step S80), the class level of the new property along the railway is compared. If it is inappropriate (step S81N), the class along the railway used for the clustering process is determined. The level is adjusted (step S83) to obtain a properly classified property group (step S84). A price evaluation model is created again based on the newly classified property group (step S8).
5). The created price evaluation model becomes an optimized price evaluation model, and is used in subsequent price evaluation processing (step S86). On the other hand, if the class level along the railway is appropriate (step S81Y), the appropriateness of the error distribution is determined, and if the error distribution is within the range of the appropriate error distribution (step S81).
82Y), the corresponding price evaluation model becomes the optimized price evaluation model. On the other hand, if it is out of the range of the appropriate error distribution (step S82N), the price evaluation model is created again (step S85), and the created model becomes the optimized price evaluation model (step S86).

Next, the area attribute information, the wayside attribute information, and the property attribute information used in the price evaluation model generation processing will be described in detail.

FIG. 9 shows an example of data items of the area attribute information 210b, the roadside attribute information 220b and the property attribute information 230b used in the first embodiment.

The area attribute information 210b and the wayside attribute information 220b are clustering factors used in the area clustering process. On the other hand, property attribute information 230
b may be further used as a stratification factor in the clustering process, or may be a characteristic factor included as a parameter in the price evaluation model.

The area attribute information 210b includes at least
It includes an area basic index related to the population of a unit area, an industrial activity index indicating productivity per population, a consumption index indicating a consumption index per unit, and a cultural index indicating a culture degree per unit. The area basic index is based on population composition by age, population growth rate, population density, population concentration, It is indicated by data such as taxable income, in-migration population rate, out-migration population rate, daytime population density, and household growth rate. The area industrial activity index is represented by data such as the amount of agricultural gross production per population and the annual shipment value of industrial products per population. The area consumption index is based on the per capita retail sales ratio, newly built housing units, total office density, employee density, total retail stores per area, number of supermarkets per area, number of restaurants per area, Total expenditure per capita, savings balance per household,
Shown by data such as the number of passenger cars owned per household. The area culture index is indicated by data such as the number of TV contracts per household.

[0092] The railroad attribute information 220b includes at least a railroad index. The alongside rating index is indicated by data such as the unit price of apartments along the railway and the unit price of land along the railway.

The property attribute information 230b includes at least convenience, maintenance status / management, equipment specifications, city infrastructure,
Includes parameters such as system restrictions, housing conditions, land conditions, seller conditions, and so on. The convenience factor is indicated by items (data) such as traffic division, walking (minute), pass (minute), bus stop (minute), car (minute), and time required to reach the terminal. The maintenance status / management factor is indicated by data such as management form, management cost, interior renovation, exterior renovation, and the like. The equipment specification factor is indicated by data of auto lock, flooring, two-family house, Grenier method, balcony area, storage room, roof balcony, private garden, basement, and the like. Urban infrastructure factors are indicated by data such as gas, sewage, and water supply. The system constraint factors are indicated by data such as used public corporations, setbacks, land features, restricted areas, building ratios, floor area ratios, and the like. Housing condition factors include the number of rooms, types, Japanese-style rooms, Western-style rooms,
LDK, opening direction, number of balconies, corner room, exterior wall type, occupied area, age, structure, building floor, room floor, total number of units, building area, etc. The land condition factor is indicated by data such as land area, private road area, floor area, road width, corner lot, subdivision, number of roads, parking lot, garage, and the like. The seller condition factor is indicated by data such as the current situation and delivery time.

The above attribute information includes various statistical data,
It is appropriately extracted from the data in the property information database 70 and the user identification information accumulated according to the input from the user.

The second area attribute information 210b
As an example of the grouping, an index related to the surrounding space of a real estate resident or a way of spending time may be used for the area clustering process.

The index relating to the user space is at least a life index which is a spatial awareness index relating to occupation and business,
Life index which is a spatial consciousness index related to consumption access, daily index which is a spatial consciousness index related to physical maintenance activities, extraordinary index which is a spatial consciousness index related to mental maintenance activities, de-life index which is a spatial consciousness index related to travel and travel access Includes the anti-secular index, which is a spatial awareness index related to ceremonial occasions, ceremonies, and customs. The life index is represented by data such as the amount of taxable income, the type of total business establishment density, and the daytime population density. The life index is calculated from the types of supermarkets, mass retailers and department stores per area. The daily index is indicated by data on parks, convenience stores, restaurants, other restaurants, sports facilities, and the like. The extraordinary index is indicated by data of a rental video store, a bookstore, a movie theater, and the like. The de-life index is indicated by data that also correlates with the convenience of airports, high-speed interchanges, hotels, and the like. The rebellious world index is indicated by data on religious facilities, temples and shrines, cemeteries, etc. By using these numbers per unit area, their types, and indices with appropriate weights for each type in area clustering, it is possible to create a cluster that reflects the user's awareness of the space (environment) surrounding the living environment. Obtainable. Also in this second grouping,
Similarly to FIG. 9, statistical data such as “private power”, geographical data, and the like can be used. In this second grouping, how to capture the user's life time, the size of the memory grain,
The clustering process can be performed by reflecting a unit (hereinafter referred to as an experience time) in weighting or the like. That is,
In the area clustering process, it is desirable to add a difference in the level of the particle size of the experience time (for example, year to day, week to minute, and time) as a guide for the clustering process. Other factors that affect the particle order of the experience time in these areas include network providers, access points calculated from the backbone, etc., communication speed, communication quality, communication costs, etc. calculated from the coverage of radio wave power, etc. Can also be used.

FIGS. 10 and 11 show an example of a real estate evaluation application form input by the user when the property to be evaluated is an apartment. FIG. 10 shows user identification information, and FIG. 11 shows property information of an apartment owned by the user and desired to be sold. On the other hand, FIG. 12 shows an example of property information in a real estate evaluation application form input by the user when the property to be evaluated is a detached house. The user identification information is the same as the user identification information in FIG. These input forms may be provided by any communication means via the Internet, including a Web page or an e-mail, and may be input to the input unit 110 of FIG. 2 as evaluation request information, or provided by a print form. The information may be input to the input unit 110 via mail or facsimile. The input user identification information and property information are used as area attribute information, railroad and property attribute information.

Next, an example of an experiment for generating a price evaluation model performed by the real estate evaluation device 10 according to the first embodiment will be described.

(1) Area Clustering Process FIG. 13 shows an example of data items of area attribute information used by the area clustering section 210 in the area clustering process. Using this data, an area segment including areas having high similarity to each other is derived.

(2) Railroad Clustering Process FIG. 14 shows an example of a railroad cluster set by the above-described classification tree method in the railroad clustering unit 220 in the railroad clustering process. In FIG. 14, N is the corresponding property number, P is shows the average m ² unit price and the standard deviation. For example, node N00001 (N = 2883, P = 58.
208, 16.667) [whole group tree name area2b2
a] and is, node data to be analyzed overall number of N00001 is 2883 cases, the average ^{m 2} unit price is 582,000 yen,
It shows that the standard deviation is 166,000 yen. [] Indicates the identification of the tree. The entire group indicated by the node N00001 is divided into the following two segments (N01001, N01002) depending on the difference in the wayside name.

A.) N01001 (N = 1305, P = 6
4.793, 17.272) Line name = Odakyu Line, Tokyu Toyoko Line, ... b.) N01002 (N = 1578, P = 52.763)
13.979) Railroad name = Chuo Line, Keio Inokashira Line, ... Similarly, N01002 is also divided into two segments by railroad name. For example, segment T03006 is N
Since the vehicle travels through the divided routes of 01002, N02003, and T03006, finally, the class along the western Ikebukuro Line, the Tobu Tojo Line, the Toei Line 12, and the Nanbu Line is defined. The unit price is 450,000 yen and the standard deviation is 118,000 yen.

For the setting of the railroad class, an analysis method generally known as AID (Automatic Interaction Detector) can be used. This AID is a method for dividing the fluctuation of the price of all data (sum of squares in a group) into a plurality of groups that have as small a variation as possible, that is, as uniform as possible within a group, due to a difference in an attribute called a railroad name. It is. When segmentation is performed based on such a criterion, price fluctuations, which are all original fluctuations, are aggregated as a difference in price average value between groups (that is, sum of squares between groups). This AID
Can be executed by giving the minimum number of segments and the maximum number of combinations of attributes as parameters. In the first embodiment, for example, the minimum number of samples of a segment is 100,
It can be executed with the maximum number of attribute combinations being four. Note that the objective variable used as the basis for segmentation is
^Two unit prices can be used.

(3) Price Evaluation Model Creation Processing FIG. 15 shows a list of stratification factor candidate items used in the experimental example of the first embodiment. The stratification factors include items extracted from the property attribute information based on input from the user or obtained by processing the property attribute information as needed.

FIG. 16 shows a list of candidate items for explanatory variables used in the multiple regression analysis in step S75 of FIG. It should be noted that the stratified factor candidate items in FIG. 15 that are not extracted as stratified factors, or that are employed but whose categories are not subdivided, that is, items 6 and 6 in FIG.
The items related to the use areas and road divisions shown in lines 7, 8, 13, 14, 15, 16, and 19 can be created as dummy variables and used as regression variables for multiple regression analysis. In the multiple regression analysis, a logarithmic transformation is performed on the regression explanatory variable and the explained variable (4, 5 in FIG.
(The items shown in the ninth, tenth, seventeenth, and eighteenth rows are applicable), thereby improving the prediction accuracy.

The price evaluation model is based on the above-mentioned line-side rating
That is, the land price is built for each class
Objective factors (X1, X
2,..., Xp) as stratification factors or explanatory variables,
“Several multiple regression model with stratified price as explained variable (Y)”
". Use this stratified (clustered) model
By this, for example, land area 100m²Properties and more
Land area 1m with property below ²Price per unit is different, etc.
Model the complex relationship between price and explanatory variables.
You.

The structure of the price evaluation model is expressed by the following equation (2). The predicted value of the model is Σβs
iXi.

U = Tree (V) S = Tree (U, X1, X2,..., Xp) = Tree (V, X1, X2,..., Xp) Y = ΣβsiXi + e (Equation 2) Is an area segment (roadside class) determined by the classification tree using the railroad as a factor, Tree (V) is a classification tree having the railroad name V as a stratification factor, S is a segment number determined by the classification tree for each railroad class, Tree (X
1, X2,..., Xp) are X1, X2,.
Classification tree to stratification factor, peripheral price Y is the objective variable (peripheral m ² bid), Betasi applies to the segment S i
The regression parameter of the i-th explanatory variable, Xi is the i-th explanatory variable, and e is the error.

In step S151, stratification factors are extracted, and a classification tree based on the AID is created.

FIG. 17 shows an example of a classification tree of stratified factors.
Here, the objective variable that is the basis for segmentation is deleted.
Fig. 1 shows ^two unit prices and attribute information expressing segments.
Six stratification factor candidate items were used. The extraction of the stratification factor is performed for each of the above-mentioned line-side clusters (for each area segment).

FIG. 18 shows the estimation results of the regression parameters of each node obtained by executing the multiple regression variable selection model for each segment. The criterion for selecting the variable is the maximum criterion of the contribution ratio adjusted with the degree of freedom (R2), and a combination of variables that satisfies the maximum criterion can be obtained by a variable brute force method. In FIG. 18, the value at the intersection of the variable name and the node ID (segment number) is the regression coefficient βsi of each explanatory variable corresponding to the segment. Regression coefficient β
When si is a missing value (indicated by a blank column in FIG. 18),
0 means that the variable is not relevant for price valuation
Can be considered.

FIG. 19 shows the property evaluation information display section 17 of FIG.
0 shows an output example of the property evaluation information output by 0. The evaluation price range of the property for which the evaluation was requested is between 4,600 and 5,100,000 yen, and the contribution of presence / absence of “Flooring” in the property attribute information is particularly high in calculating the evaluation value. It is suggested that a room or facing southwest affects the evaluation price calculation. In the property evaluation information, in addition to the information on the evaluation price, as shown in FIG.
Can be presented by searching for. In addition, by presenting the time-lapse market data of the municipalities based on the history information of the property transaction, it is possible to provide the user with information for determining the appropriate time to sell the property.

FIG. 21 shows that the property evaluation information display section 170
13 shows another output example in which property evaluation information is displayed on a screen on the Web or the like.

According to the first embodiment, the following effects can be obtained.

The real estate evaluation device 10 according to the first embodiment
Generates a price evaluation model for each clustered area (area segment) classified according to the similarity of the area characteristics, and calculates the evaluation price of the target property based on the price evaluation model. In addition, the price evaluation model is optimized according to the actual situation of the transaction in the market.

[0115] For this reason, it is possible to easily provide the user with a high-accuracy rating for a real estate property that he / she wants to sell.

Second Embodiment Hereinafter, a second embodiment of the present invention will be described in detail only with respect to differences from the first embodiment with reference to FIGS.

The second embodiment provides a function of mediating between the user 40 and the real estate agent 50 by using the property evaluation information obtained by the real estate evaluation device 10 according to the first embodiment.

FIG. 22 is a block diagram showing a functional configuration of a real estate agent requesting apparatus 30 according to the second embodiment of the present invention. In FIG. 22, the same portions as those in FIG. 2 are denoted by the same reference numerals, and only different portions will be described.

The real estate agent requesting apparatus 30 according to the second embodiment includes an input unit 310, a neighborhood area handling result determination unit 320, a real estate agent display unit 330, and a property evaluation information distribution unit 340. Is done.

The input unit 310 inputs a property evaluation request similar to that of the real estate evaluation device 10 according to the first embodiment, and inputs real estate agent introduction request information from the user 40.

[0121] The neighborhood area handling performance determination unit 320
Real estate agent handling results database 32 that holds history information of real estate handling results including real estate agent information
0b is searched using the area in the evaluation request information as a key to search for a real estate agent that has been handled in that area or a nearby area, and sends it to the real estate agent display section 330.

The real estate agent display section 330 is a real estate agent database 320b holding information for each real estate agent.
With the identifier of the found real estate agent as a key,
Outputs real estate agent information.

[0123] The property evaluation information distribution unit 340 includes the input unit 31
Based on the instruction from 0, the property evaluation information is added to the real estate agent introduction request information describing the name of the real estate agent selected by the user 40, and output to the real estate agent 50 as seller introduction information.

FIG. 23 shows an example of a real estate agent information providing screen output by the real estate agent display section 330. As real estate agent information, the company name, company profile, provided service, user score held in the real estate agent database 320b,
The presence or absence of user evaluation and the like are presented. Further, in addition to the provision of the real estate agent information, it is possible to provide the user with introduction information and arrangement information of the remodeling agent 60a and the moving agent 60b in FIG. Thereby, the user 40 can enjoy a so-called one-stop service that can be realized by accessing all of the information on the property to be sold necessary for relocation, selecting a real estate company, and remodeling / moving arrangements, etc. at one place. In addition, for example, in the arrangement information of the renovation company 60a,
Scheduled renovation, property type (apartment or single-family house), desired renovation site (water, living room, exterior,
Etc.).

According to the second embodiment, the following effects can be further obtained in addition to the effects of the first embodiment.

The real estate agent introduction device 30 according to the second embodiment constitutes the real estate information providing system 1 shown in FIG. 1 together with the real estate evaluation device 10 according to the first embodiment, and provides real estate agent information to users. At the same time, the property evaluation information is added to the introduction request information of the real estate agent selected by the user from the provided real estate agent information, and the information is distributed to the real estate agent designated as the sale requester information. For this reason,
Quick mediation between the user and the real estate agent becomes possible.

It is needless to say that each of the above embodiments may be implemented independently, or may be implemented in combination with other embodiments as appropriate.

Further, the present invention is not limited to the above-described embodiment, and does not depart from the gist of the present invention.
It goes without saying that various changes and modifications can be made. All of these changes and modifications are included in the scope of the present invention.

[0129]

As described above, according to the present invention,
The following effects are obtained. That is, the present invention has a function of generating a price evaluation model for each clustered area (area segment) classified according to the similarity of the area characteristics and calculating the evaluation price of the target property based on the price evaluation model. provide. In addition, a function is provided for correcting the price evaluation model in accordance with the actual situation of transactions in the market. Furthermore, a function to mediate between the user and the real estate agent based on the obtained evaluation price information is provided.

[0130] Thus, the real estate property that the user wants to sell, buy or refinance, etc.
It is possible to easily provide a highly accurate evaluation value.
For this reason, the user can sell the target property at a reasonable bargain price at an early stage. In addition, by using this evaluation price information, it is possible to introduce a real estate agent that brokers the sale to the user, and to provide information of prospective customers to the real estate agent, the remodeling agent, the moving company, and the like. Alternatively, information of these prospective customers can be provided to a financial institution or the like and used for collateral evaluation.

As described above, according to the present invention, quick and efficient promotion of real estate is realized.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an overall functional configuration of a real estate information providing system according to the present invention.

FIG. 2 is a block diagram illustrating a functional configuration of the real estate evaluation device according to the first embodiment.

FIG. 3 is a block diagram illustrating a detailed functional configuration of a price evaluation model generation unit 200 in FIG. 2;

FIG. 4 is a diagram for explaining the relationship between clustered segments along the roadside feature and a price evaluation model.

FIG. 5 is a flowchart illustrating a procedure of an area clustering process performed by an area clustering unit 210;

FIG. 6 is a flowchart illustrating a processing procedure of a wayside clustering process performed by a wayside clustering unit 220;

FIG. 7 is a flowchart illustrating a procedure of a price evaluation model creation process performed by a price evaluation model creation unit 230.

FIG. 8 is a flowchart showing a processing procedure of price evaluation model optimization processing performed by a price evaluation model optimization unit 240.

FIG. 9 is a diagram illustrating an example of data items of area attribute information, line attribute information, and property attribute information.

FIG. 10 is a diagram illustrating an example of an input form of a personal information part input as property evaluation request information of an apartment.

FIG. 11 is a diagram illustrating an example of an input form of a property information portion input as property evaluation request information of an apartment.

FIG. 12 is a diagram illustrating an example of an input form for a property information portion input as property evaluation request information for a single-family house.

FIG. 13 is a diagram illustrating an example of items of area attribute information used in the area clustering process.

FIG. 14 is a diagram illustrating an example of an area (segment) classified by the classification tree method in the wayside clustering processing.

FIG. 15 is a diagram illustrating an example of a candidate item of a stratification factor used in a price evaluation model creation process.

FIG. 16 is a diagram illustrating an example of candidate items for regression analysis required explanatory variables used in the price evaluation model creation process.

FIG. 17 is a diagram illustrating an example of stratified factors classified by the classification tree method in the price evaluation model creation process.

FIG. 18 is a diagram showing a list of parameter values in stratified multiple regression analysis.

FIG. 19 is a diagram illustrating an example of an output form of property evaluation information output by the property evaluation information display unit 170 of the real estate evaluation device 10 according to the first embodiment.

20 is a diagram illustrating an example of incidental information added to the property evaluation information of FIG. 190 and output to an output form.

FIG. 21 is a diagram illustrating an example of an output screen of property evaluation information output by a property evaluation information display unit 170.

FIG. 22 is a block diagram showing a functional configuration of a real estate agent introduction device 30 according to a second embodiment of the present invention.

FIG. 23 is a diagram showing an example of an output screen of real estate agent information output by a real estate agent display section 330.

FIG. 24 is a flowchart illustrating a procedure up to sale of a user who sells owned real estate.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Real-estate information provision system 10 Real-estate evaluation device 30 Real-estate agent request device 40 User who wants to move 50 Real-estate agent 60a Remodeling agent 60b Moving company 70 Property information database 80 Personal information database 110, 310 Input part 120 Pre-processing part 120a Sale example DB 120b Public announcement Price-based land price DB 120c Neighborhood rental DB 120d Price evaluation model DB 130 Area segment judgment unit 140 Lineside segment judgment unit 150 Price evaluation model selection unit 160 Price calculation unit 170 Property evaluation information display unit 180a Neighborhood sale case search unit 180b Neighborhood posted price search Unit 180c Neighborhood rental case search unit 200 Price evaluation model generation unit 210 Area clustering unit 220 Lineside clustering unit 230 Price evaluation model generation unit 240 Price evaluation model optimization unit 320 Neighboring area handling performance determination unit 320b Real estate agent handling performance DB 330 Real estate agent display unit 330b Real estate agent DB 340 Property evaluation information distribution unit

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Shinya Hayakawa 7-3-5 Ginza, Chuo-ku, Tokyo Inside Recruit Co., Ltd. (72) Inventor Hideo Tanioka 1-2-2 Akasaka, Minato-ku, Tokyo Financial Engineering Co., Ltd. -Within the group (72) Inventor Shigeo Sano 1-2-2 Akasaka, Minato-ku, Tokyo F-term within the Financial Engineering Group, Inc. (reference) 5B049 BB11 CC00 CC44 CC45 EE00 EE31

Claims (31)

[Claims] 1. A real estate evaluation apparatus for calculating a predicted price of a given real estate, comprising: a price evaluation model for calculating a predicted price of the real estate based on input property data; An evaluation model generation unit that generates for each segment including one or a plurality of areas, an evaluation model storage unit that stores the price evaluation model, and determines a segment to which a real estate to be evaluated belongs based on the property data. A segment determination unit to perform, a price evaluation model corresponding to the segment determined by the segment determination unit, an evaluation model selection unit that selects from the evaluation model storage unit, and a price evaluation model selected by the evaluation model selection unit. An evaluation unit for calculating a predicted price of the real estate based on the Evaluation device. 2. The evaluation model generation unit calculates a similarity between each area based on area attribute information indicating an attribute of an area to which a real estate property belongs, and calculates a given area group based on the calculated similarity. An area classification unit that obtains a segment group by clustering according to the degree; and for each segment obtained by the area classification unit, the price evaluation model by statistical analysis based on property attribute information indicating an attribute of a real estate property belonging to the segment. The real estate evaluation device according to claim 1, further comprising: a model generation unit that generates the real estate evaluation. 3. The real estate evaluation device according to claim 2, wherein the statistical analysis is based on a stratified multiple regression analysis. 4. The area classification unit further obtains a second segment group by further clustering the clustered segment group on the basis of railroad attribute information indicating a rating of a railroad. 4. The real estate evaluation device according to claim 2, wherein the price evaluation model corresponding to the second segment group is generated. 5. The area attribute information includes at least a population index related to a population of a unit area, a productivity index indicating a productivity per population, a consumption index indicating a consumption index per unit, and a culture indicating a culture degree per unit. One of the indices
The real estate evaluation device according to claim 2, wherein the real estate evaluation device includes at least one. 6. The real estate evaluation device according to claim 4, wherein the railroad attribute information includes a railroad rating index indicating a unit price per tsubo. 7. The property attribute information includes at least convenience information indicating convenience of transportation of the real estate property, maintenance information indicating a maintenance level of the real estate property, facility information indicating equipment specifications of the real estate property, Urban infrastructure information indicating the gas and water laying status of the real estate, unit price information indicating the unit price of the land parcel to which the real estate belongs, constraint information indicating the legal constraints of the real estate, the area or structure of the house or land 7. The real estate evaluation device according to claim 2, wherein the real estate evaluation device includes at least one of structural information indicating the following and seller condition information provided by a selling entity of the real estate. 8. The real estate evaluation device further includes: a sales history storage unit that stores a history of past real estate sales information; and an evaluation model optimization unit that optimizes the price evaluation model based on the sales information history. The real estate evaluation device according to any one of claims 1, 2, 3, 4, 5, 6, and 7, further comprising a conversion unit. 9. The real estate evaluation device further includes: presenting the predicted price calculated by the evaluation unit in a price range of a predetermined width, and comparing the predicted price in the property attribute information with the predicted price calculation. The real estate evaluation device according to any one of claims 1, 2, 3, 4, 5, 6, 7, and 8, further comprising an evaluation information presenting unit that presents part of the property attribute information that has contributed to the project. . 10. The real estate evaluation device according to claim 9, wherein the evaluation information presenting unit further presents the predicted price as a change over time. 11. A real estate evaluation method for calculating a predicted price of a given real estate, comprising: a price evaluation model for calculating a predicted price of the real estate based on input property data; An evaluation model generating step of generating for each segment composed of one or more areas having the same; a segment determining step of determining a segment to which a real estate to be evaluated belongs based on the property data; and a determination by the segment determining step. An evaluation model selecting step of selecting a price evaluation model corresponding to the segment selected from the price evaluation models, and calculating a predicted price of the real estate based on the price evaluation model selected by the evaluation model selection process. And a valuation step. 12. The evaluation model generating step calculates a similarity between each area based on area attribute information indicating an attribute of an area to which the real estate belongs, and calculates a given area group based on the calculated similarity. An area classification step of clustering according to the degree to obtain a segment group; and for each segment obtained in the area classification step, the price evaluation model by statistical analysis based on property attribute information indicating an attribute of a real estate property belonging to the segment. And generating a model. 13. The real estate evaluation method according to claim 12, wherein the statistical analysis is based on a stratified multiple regression analysis. 14. The area classifying step further comprises: further clustering the clustered segment group on the basis of the wayside attribute information indicating the rating of the wayside to obtain a second segment group; 14. The real estate valuation method according to claim 12, wherein the price valuation model corresponding to the second segment group is generated. 15. The area attribute information includes at least:
Including one or more of a population index related to a population of a unit area, a productivity index indicating a productivity per population, a consumption index indicating a consumption index per unit, and a cultural index indicating a culture degree per unit. Claims 11, 12, 1
15. The real estate evaluation method according to 3 or 14. 16. The real estate evaluation method according to claim 14, wherein the railroad attribute information includes a railroad rating index indicating a unit price per tsubo. 17. The property attribute information includes at least convenience information indicating convenience of transportation of the real estate property, maintenance information indicating a degree of maintenance of the real estate property, facility information indicating equipment specifications of the real estate property, Urban infrastructure information indicating the gas and water laying status of the real estate, unit price information indicating the unit price of the land parcel to which the real estate belongs, constraint information indicating the legal constraints of the real estate, the area or structure of the house or land And at least one of seller condition information provided by a selling entity of the real estate property and the structural information indicating the real estate property.
The real estate evaluation method according to any one of 2, 13, 14, 15, and 16. 18. The real estate evaluation method, further comprising: a sales history storage step of accumulating a history of past real estate sales information; and an evaluation model optimization for optimizing the price evaluation model based on the sales information history. The real estate valuation method according to any one of claims 11, 12, 13, 14, 15, 16 and 17, further comprising: 19. The real estate evaluation method further includes presenting the predicted price calculated in the evaluation step in a price range having a predetermined width, and comparing the predicted price calculated in the property attribute information with the calculated predicted price. 11. An evaluation information presenting step of presenting a part of property attribute information that has contributed to the market.
The property evaluation method according to any one of 5, 16, 17 and 18. 20. The real estate evaluation method according to claim 19, wherein the evaluation information presenting step further presents the predicted price as a change over time. 21. A computer-readable recording medium for storing a program for causing a computer to execute a real estate evaluation process for calculating a predicted price of a given real estate article, wherein the predicted price of the real estate article is calculated based on the input property data. An evaluation model generation process of generating a price evaluation model to be calculated for each segment including one or more areas having characteristics of mutually similar areas; and a segment to which a real estate to be evaluated belongs based on the property data. , A price evaluation model corresponding to the segment determined by the segment determination process, and an evaluation model selection process of selecting from the price evaluation models; and a price selected by the evaluation model selection process. Based on the valuation model, calculate the predicted price of the real estate A computer-readable recording medium characterized in that it comprises an evaluation process for output. 22. The evaluation model generation processing includes calculating a similarity between each area based on area attribute information indicating an attribute of an area to which a real estate property belongs, and dividing a given area group into the calculated similarity. Area classification processing that obtains a segment group by clustering according to degree, and for each segment obtained by the area classification processing,
22. The computer-readable recording medium according to claim 21, further comprising: a model generation process of generating the price evaluation model by statistical analysis based on property attribute information indicating an attribute of a real estate property belonging to the segment. . 23. The computer-readable recording medium according to claim 22, wherein the statistical analysis is based on a stratified multiple regression analysis. 24. The area classification processing further comprises: further clustering the clustered segment group on the basis of the alongside attribute information indicating the rating of the alongside to obtain a second segment group; 24. The computer-readable recording medium according to claim 22, wherein the price evaluation model corresponding to the second segment group is generated. 25. The area attribute information includes at least:
Including one or more of a population index related to a population of a unit area, a productivity index indicating a productivity per population, a consumption index indicating a consumption index per unit, and a cultural index indicating a culture degree per unit. Claims 21, 22, 2,
25. A computer-readable recording medium according to any one of items 3 or 24. 26. The computer-readable recording medium according to claim 24, wherein said railroad attribute information includes a railroad rating index indicating a unit price per tsubo. 27. The property attribute information includes at least convenience information indicating convenience of transportation of the real estate property, maintenance information indicating a degree of maintenance of the real estate property, facility information indicating equipment specifications of the real estate property, Urban infrastructure information indicating the gas and water laying status of the real estate, unit price information indicating the unit price of the land parcel to which the real estate belongs, constraint information indicating the legal constraints of the real estate, the area or structure of the house or land And at least one of seller condition information provided by a selling entity of the real estate property and structural information indicating the property information.
27. The computer-readable recording medium according to any one of 2, 23, 24, 25, and 26. 28. The computer-readable recording medium further comprising: a sales history storage process for accumulating a history of past real estate sales information; and optimizing the price evaluation model based on the sales information history. The computer-readable recording medium according to any one of claims 21, 22, 23, 24, 25, 26, and 27, comprising an evaluation model optimization process. 29. The computer-readable recording medium further comprises: presenting the predicted price calculated by the evaluation processing in a price range having a predetermined width; and presenting the predicted price in the property attribute information. 27. An evaluation information presenting process for presenting a part of property attribute information that relatively contributed to the calculation.
29. The computer-readable recording medium according to any of 27 or 28. 30. The computer-readable recording medium according to claim 29, wherein the evaluation information presenting process further presents the predicted price as a change over time. 31. A real estate information providing system for calculating a predicted price of a given real estate property and providing real estate information to a user and a real estate agent via the predicted price, wherein the real estate property is calculated based on the input property data. An evaluation model generation unit that generates a price evaluation model for calculating a predicted price for each segment including one or more areas having characteristics of mutually similar areas; and a real estate to be evaluated based on the property data. A segment determination unit that determines a segment to which the property belongs; an evaluation model selection unit that selects an evaluation model corresponding to the segment determined by the segment determination unit from the evaluation models; and a selection unit that is selected by the evaluation model selection unit. An evaluation unit that calculates a predicted price of the real estate based on the evaluation model A real estate evaluation device to be provided, a handling performance determining unit for determining whether or not there is a handling performance in an area to which the input target real estate data is to be evaluated or in a nearby area, and real estate agent information on a real estate agent for each past handling property A real estate agent information storage unit for storing the real estate agent that has been determined to have a handling record from the real estate agent information storage unit, and presents real estate agent information that is a list of the searched real estate agents. An information presenting unit; and an introduction information distributing unit that adds the evaluation price information output by the evaluation unit of the real estate evaluation device to the referral request information of the designated real estate agent and distributes it to the designated real estate agent. A real estate information providing system comprising a real estate agent requesting device.