JP7455254B1 - Prediction device and program - Google Patents

Abstract

[Problem] To make it possible to conduct real estate purchase or brokerage business, etc. to owners at a stage before the property for sale becomes apparent. [Solution] A lifeline information acquisition unit 13 that acquires information regarding a customer's lifeline, a general information acquisition unit 14 that acquires general information regarding condominiums, information regarding the customer's lifeline, general information, and information regarding the customer's condominium. The prediction device 10 includes a sales prediction unit 16 that predicts the possibility of selling a property where a customer resides by performing machine learning using training data and information indicating whether or not the condominium has been sold. . [Selection diagram] Figure 3

Description

The present invention relates to a prediction device and a program.

As awareness of environmental issues and the effective use of resources increases, the trend of overemphasizing new construction in real estate is being reconsidered, and the market for used housing such as used condominiums is being revitalized.
For example, Patent Document 1 describes a real estate purchase system that is connected to a network and supports the purchase of real estate. The network communication means includes a property registration means for registering property information, and a property providing means for preparing property provision information to be provided on the network from the property information registered by the property registration means, and the network communication means is provided by the property provision means. This document describes a technology for disclosing property information provided via a network.

Japanese Patent Application Publication No. 2002-024596

In recent years, in addition to transactions between individuals in which used properties are bought and sold through intermediaries, purchase transactions in which intermediaries purchase used properties, increase the asset value through renovation and renovation, and then resell them at a cheaper price than new construction. There is a resale business. In this purchase and resale business, it is important from the point of view of business continuity to purchase a large amount of used properties, just as in the traditional brokerage business.
Owners of second-hand properties who are considering selling generally contact a real estate agent (brokerage agent, purchase reseller The owner himself/herself contacts the owner (regardless of the matter) and conveys his/her intention to sell the property, but as mentioned above, there are only a few ways in which a real estate agent can approach the owner, and the owner is left passively waiting. In other words, it is difficult for real estate agents to identify owners who are willing to sell, and if they cannot identify them, it is difficult to stably purchase used properties in large quantities. Furthermore, it is even more difficult to identify owners who have no intention of selling, even if they are no longer living in the property or are likely to stop living in the near future.

The purpose of the present invention is to solve such problems, and specifically, it is possible to conduct real estate purchase or brokerage business for owners at a stage before the property for sale becomes apparent. It is to do so.

Based on such an objective, the present invention includes a lifeline information acquisition unit that acquires information regarding a customer's lifeline, a general information acquisition unit that acquires general information regarding condominiums, and a general information acquisition unit that acquires information regarding the customer's lifeline and the general information. a sales prediction unit that predicts the possibility of selling the property in which the customer resides by performing machine learning using the information and information indicating whether the customer has sold the condominium as training data; In case the residence belongs to a condominium, the general information includes at least information regarding the number of sales and rental transactions , and the sales prediction unit calculates the number of sales and rental transactions based on the information regarding the number of sales and rental transactions. This is a prediction device that determines the possibility that a property is an investment property .
Here, the information regarding the lifeline may include information regarding at least one of the type, name, model number, installation location, payment method, and service contract of equipment used for provision by the lifeline. can.
Further, the general information includes information indicating at least one of the property's size, location, zoning, whether or not it is rented, residence name, condominium company, construction company, and management company. I can do it.
Furthermore, the general information may include information indicating whether or not the property is rented.
Furthermore, the information indicating whether or not the property is rented may include information indicating at least one of the size of the property, the name of the residence, a sales company, a construction company, and a management company.
Furthermore, if the residence belongs to a condominium, the general information may further include information regarding whether the seller is a real estate agent or an individual.
Furthermore, the information regarding the lifeline may include information indicating at least one of the reason for starting use, the receiving medium, the payment method, and the contract name/type .
Another aspect of the present invention that achieves the above object is to provide an information processing device with a lifeline information acquisition function that acquires information about a customer's lifeline, a general information acquisition function that acquires general information about a condominium, and a general information acquisition function that acquires general information about a condominium. Predict the possibility of selling the property in which the customer lives by performing machine learning using information about lifelines, the general information, and information indicating whether the customer has sold the condominium as training data. If the residence belongs to a condominium, the general information includes at least information regarding the number of sales and rental transactions; This is a program that determines the possibility that the property is an investment property based on information regarding the number of properties .

According to the present invention, it becomes possible to conduct real estate purchase or brokerage business for owners at a stage before the property for sale becomes apparent.

1 is a diagram showing an example of the overall configuration of a real estate transaction guidance system according to the present embodiment. It is a diagram showing an example of the hardware configuration of a prediction device and a terminal. It is a figure showing an example of functional composition of a prediction device. It is a diagram explaining a machine learning model. FIG. 7 is a diagram illustrating another example of teacher data. FIG. 4 is a diagram showing an example of a screen output and displayed on a terminal of a business operator in the case of a telephone reception, in which (a) shows an example of a reception screen for a telephone operator closure request, and (b) shows another example of the reception screen. An example is partially shown, and (c) partially shows another example of a reception screen. FIG. 3 is a diagram showing an example of a screen output and displayed on a user's terminal in the case of web reception, where (a) shows a web reception screen (1/n) for a stopper application, and (b) shows a web reception screen for a stopper application. An example of the reception screen (2/n) is shown. FIG. 4 is a diagram showing an example of a screen output and displayed on a user's terminal in the case of Web reception, in which (a) shows another example of the reception screen (2/n), and (b) shows another example of the reception screen (2/n). /n) is shown below.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
FIG. 1 is a diagram showing an example of the overall configuration of a real estate transaction guidance system 1 according to the present embodiment.
A real estate transaction guidance system 1 according to the present embodiment is a system that estimates vacant houses where no one will live, using information regarding contracts or usage of infrastructure projects, and guides candidates for real estate transaction partners that will improve the business efficiency of real estate transactions. It is.
Note that properties that are presumed to be vacant include cases where the current residents plan to move, as well as cases where the current residents have already moved.

Infrastructure business here refers to business that serves as the foundation for supporting industry and daily life, and includes energy infrastructure businesses such as electricity and gas. Changes in energy infrastructure project contracts and usage status are useful information for estimating vacant housing. Furthermore, the infrastructure project referred to here can include other infrastructure projects that may be useful for estimating vacant houses. There are various classifications of infrastructure projects, but examples include daily life infrastructure projects such as water supply and sewage systems, broadcast reception, and communication infrastructure for telephone networks and communication networks.

The real estate transaction guidance system 1 includes a prediction device 10, a terminal 20, a terminal 30, and a terminal 40. The prediction device 10 and the terminal 20, the prediction device 10 and the terminal 30, and the prediction device 10 and the terminal 40 are connected via a network 90.

The terminal 20 is a device of a consumer, user, or user (hereinafter referred to as a user) related to infrastructure business, and is located in a residential building together with the telephone T20.
The terminal 30 is a device of an infrastructure business supplier, provider, or business operator (hereinafter referred to as a business operator). The operator has a telephone T30 for talking to the user's telephone T20.
The terminal 40 is a device of a transaction agent that conducts real estate transactions, such as real estate purchase transactions and real estate brokerage transactions. The business operator has a printer P40 for performing various types of printing from the terminal 40.

The prediction device 10 of the real estate transaction guidance system 1 acquires information on used houses such as used condominiums, and also acquires information on contracts and usage status of gas, electricity, water, telephone, communications, etc. in the used houses. More specifically, the above-mentioned used housing information may be obtained from a database such as housing information, or may be automatically or manually obtained from information published on the Internet or the like.
The used houses referred to here are residential buildings that currently have occupants, and do not include residential buildings that have not been occupied since they were newly built. To explain further, used homes include not only cases where the current resident has lived in the home since it was newly built, but also cases where there was a previous resident before the current resident. In addition, newly built condominiums for which a sales contract has not been concluded even after a predetermined number of years have passed since construction are also referred to as new and old condominiums, but such new and old condominiums are not included in used housing.
In addition, used properties include condominiums and other residential buildings, as well as detached houses for sale. In the case of a condominium, the unit is a room.

The prediction device 10 is realized by, for example, an information processing device such as a computer. The prediction device 10 may be configured by a single computer, or may be realized by distributed processing by multiple computers.

As described above, the terminal 20 is a device of a user of an infrastructure business, and is realized by, for example, a PC (Personal Computer) or the like.
As described above, the terminal 30 is a device of an infrastructure business operator, and is realized by, for example, a PC or the like. Examples of the business operator include a gas business that supplies gas and an electric power business that supplies electricity. Such gas utilities and electric power utilities can be called service providers.
As described above, the terminal 40 is a device of a real estate transaction agent, and is realized by, for example, a PC or the like. Transaction agents include real estate purchasers and real estate brokers.

Note that although this embodiment employs a configuration example in which the terminal 30 and the terminal 40 are different devices, the present invention is not limited to this, and a configuration example in which the terminal 30 and the terminal 40 are the same device may be adopted. . Further, in this embodiment, a configuration example in which the prediction device 10 and the terminal 40 are different devices is adopted, but the present invention is not limited to this, and a configuration example in which the prediction device 10 and the terminal 40 are the same device is adopted. Good too.
Furthermore, in this embodiment, a configuration example in which the prediction device 10, the terminal 30, and the terminal 40 are different devices is adopted, but the present invention is not limited to this, and the prediction device 10, the terminal 30, and the terminal 40 are the same device. A configuration example may also be adopted.

In addition, the network 90 used for the connection between the prediction device 10 and the terminal 20, the connection between the prediction device 10 and the terminal 30, and the connection between the prediction device 10 and the terminal 40 may be of any type if it is possible to send and receive data. is not particularly limited, and may be, for example, the Internet, LAN (Local Area Network), WAN (Wide Area Network), or the like. The communication line used for data communication may be wired or wireless. Further, a configuration may be adopted in which each device is connected via a plurality of networks 90 or communication lines.

<Hardware configuration example>
FIG. 2 is a diagram showing an example of the hardware configuration of the prediction device 10 and the terminals 20 to 40.
As shown in FIG. 2, each of the prediction device 10, the user's terminal 20, the business operator's terminal 30, and the transaction agent's terminal 40 has a CPU (Central Processing Unit) 100a as a calculation means and a main storage means. and a memory 100c. Each device also includes, as external devices, a magnetic disk drive (HDD: Hard Disk Drive) 100g, a network interface 100f, a display mechanism 100d, an audio mechanism 100h, an input device 100i such as a keyboard and a mouse, and the like.

The memory 100c and display mechanism 100d are connected to the CPU 100a via the system controller 100b. Further, the network interface 100f, magnetic disk device 100g, audio mechanism 100h, and input device 100i are connected to the system controller 100b via the I/O controller 100e. Each component is connected by various buses such as a system bus and an input/output bus.

The magnetic disk device 100g stores programs for realizing each function. Then, this program is loaded into the memory 100c, and processes based on this program are executed by the CPU 100a, thereby realizing various functions.

<Functional configuration of prediction device>
Next, the functional configuration of the prediction device 10 will be explained.
FIG. 3 is a diagram showing an example of the functional configuration of the prediction device 10.
The prediction device 10 includes a transmitting/receiving section 11, an attribute information acquisition section 12, a lifeline information acquisition section 13, a general information acquisition section 14, a sold information acquisition section 15, a sale prediction section 16, and an output section 17. , is provided.

The transmitting/receiving unit 11 is a network interface for connecting to a user's terminal 20, a business operator's terminal 30, and a trading company's terminal 40 via a network 90 (see FIG. 1). Thereby, the prediction device 10 exchanges data with the terminals 20 to 40.

The attribute information acquisition unit 12 acquires customer attribute information, which is information (customer information) about contracted customers stored in the terminal 30 (see FIG. 1) of the operator, via the network 90 (see FIG. 1). and obtain it. The customer attribute information includes information such as the subscriber's name, address (residence indication), telephone number, and subscriber number.

The lifeline information acquisition unit 13 acquires lifeline information that is information regarding a customer's lifeline. The lifeline of such customers is intended to include water, gas, electricity, etc., and is synonymous with infrastructure business.
The lifeline information referred to here refers to information indicating the amount of water, gas, electricity, etc. used or utilized, the consumption amount, and the charges corresponding to the usage amount, etc., during a predetermined period.

The general information acquisition unit 14 acquires general information regarding condominiums. More specifically, the general information acquisition unit 14 obtains condominium information in advance from the Internet, etc., including condominium name or building name, location, year of construction, number of floors, etc. , obtained via the network 90. A list of condominiums can be obtained from such information on condominiums. Note that the general information acquisition unit 14 may acquire data on a list of condominiums currently on sale.
The general information referred to here includes information specifying that the building is a condominium building, and also includes information indicating a room that is a unit of sale in the condominium building.
Further, the information indicating a room may include information indicating whether the room is a corner room in a building. For example, depending on the property, if a non-corner room is less likely to be sold than a corner room, it may be more likely that a property that is not planned to be lived in will be rented rather than sold. It can be used to predict possibilities.

The sold information acquisition unit 15 acquires information indicating whether the customer has sold the condominium in association with the attribute information provided by the attribute information acquisition unit 12. Such information can be obtained from a data vendor via the Internet or the like. Note that such information collection may be performed by the business operator by calling the customer.

The sale prediction unit 16 receives as input the lifeline information acquired by the lifeline information acquisition unit 13, the general information acquired by the general information acquisition unit 14, and information indicating whether the customer has sold the condominium. A generated machine learning model 50, which will be described later, is used. The machine learning model 50 defines the relationship between lifeline information and general information and information indicating whether the item has been sold or not, and is updated by verifying using prediction results.
The sale prediction unit 16 uses the generated machine learning model 50 to predict the possibility of selling the property where the customer resides. Such prediction is performed at a stage before making a closure request. Note that there are two ways to request closure: by telephone and by web. Such closure refers to one mode of suspension of use, and may be other modes such as suspension of use of electric power or water supply.

The output unit 17 outputs the possibility of selling the property in which the customer resides, as predicted by the sale prediction unit 16. The output destination is the transaction agent's terminal 40, and is performed via the network 90 (see FIG. 1).
Therefore, the information output by the output unit 17 makes it possible to conduct real estate purchase sales and the like to the owner at a stage before the property for sale becomes apparent.
Note that the real estate purchase business mentioned here includes real estate purchase business and real estate brokerage business, and includes real estate purchase where a real estate agent purchases real estate that the owner intends to sell, and This includes cases of real estate brokerage where real estate agents find.

FIG. 4 is a diagram illustrating the machine learning model 50. The machine learning model 50 is generated by the sale prediction unit 16 (see FIG. 3).
The machine learning model 50 shown in FIG. 4 is created by performing machine learning on information about the customer's lifeline held by the business operator and condominium data collected from outside by the business operator as training data for the property where the customer resides. Build a model to predict the likelihood of selling. Data owned by such businesses and data obtained from outside are included as feature quantities.
This condominium data includes general information regarding the condominium and information indicating whether the customer has sold the condominium.

In the machine learning model 50, the input data includes lifeline information 5b, general information 5d, and sold information 5f, which are input to the input layer 51. The customer's lifeline information and information indicating whether the customer has sold the condominium are used as training data.
The lifeline information 5b is information acquired by the lifeline information acquisition unit 13, the general information 5d is information acquired by the general information acquisition unit 14, and the sold information 5f is information acquired by the lifeline information acquisition unit 13. This is information acquired by the unit 15.

In the intermediate layer 52, the lifeline information 5b, general information 5d, and sold information 5f are combined and correlated with each other through arithmetic processing.
The intermediate layer 52 performs machine learning using part of the lifeline information 5b, general information 5d, and sold information 5f input to the input layer 51, performs verification using the remaining information, and performs weighting adjustment.
Note that whether the item has been sold or not is indicated as 0/1 as correct data. It is possible to call a customer and ask them whether they own a home or not, and use that as correct data.

The output layer 53 outputs an output solution that predicts the possibility of selling the property in which the customer resides. In this way, the possibility of selling the property in which the customer resides is predicted for the customer who has not closed the valve. Note that predicting the possibility of selling a customer's residential property can be said to be a prediction of whether the customer owns the property.

Here, a description will be given of machine learning using, as training data, information regarding the lifeline of a customer from which information indicating whether the customer has sold a condominium has been acquired. The examples described below are examples of data that have been confirmed to be significant out of a large number of data, and are not limited to these.

For example, the information regarding Lifeline includes information regarding at least one of the type, name, model number, installation location, payment method, and service contract of equipment used for provision by Lifeline.

The equipment referred to here is, for example, equipment installed in a residence, such as a water heater, and the type, name, and model number of the equipment are information for identifying such equipment. The installation location of the equipment is, for example, information indicating whether it is outdoors or indoors, the payment method for the equipment is information indicating whether it is a lump-sum payment or installment payment, etc., and the equipment service contract is information that indicates whether the equipment is installed outdoors or indoors. This is information indicating such things. For example, a maintenance contract for a particular water heater tends to have a high probability that the customer is a homeowner.

For example, the general information includes information indicating at least one of the property's size, location, zoning, whether or not it is rented, name of residence, condominium company, construction company, and management company.

The size of the property referred to here refers to the exclusive area in the case of an apartment, and the number of tsubos or total floor area in the case of a detached house. If the exclusive area is small, there is a high probability that the property will be rented; in other words, the probability that the property will be sold is low.
Location includes factors such as being close to the nearest station, being within walking distance from the nearest station, and having commercial facilities nearby. Zoning is one of the regional districts under the City Planning Act, and there are several types, such as Class 1 low-rise residential zones.

Whether the property is rented or not is information indicating whether or not it is a rental property. More specifically, in addition to information on whether the building was built for rental purposes, in the case of condominiums, some of the rooms may be rented, so information is provided on a per-room basis.

The house name indicates whether the building is for rental, such as "Corpo" or "Maison" in the case of a condominium, and it is possible to determine whether the building is for rent or not on a building-by-building basis. The condominium sales company, construction company, and management company are businesses involved in buildings, and whether or not the property is rented by the sales company, the construction company that handles the construction work, and the management company that handles the management contract for the common areas of the completed building. It is possible to determine the Furthermore, sales companies may also be included.

For example, if the residence belongs to a condominium, the general information includes at least one of information regarding the number of sales or rental transactions and information regarding whether the seller is a real estate agent or an individual.
The number of transaction cases referred to here is the actual value of sales and rentals, and the higher the number of transaction cases, the higher the probability of owning a home, and therefore the higher the possibility of selling. Furthermore, even if the number of sales transactions is large, if there is a recent rental transaction, the probability that the property is an investment property for the purpose of rental income increases. Furthermore, if the transaction frequency is high and the property is bought and sold within a relatively short period of time, it is considered that the probability that the property is an investment property is high.

For example, the information regarding the lifeline includes information indicating at least one of the reason for starting use, the acceptance medium, the payment method, and the contract name/type.
The reason for starting use here is information indicating the reason for opening the bottle, and is information that can be obtained by the business operator confirming the reason for the application with the customer when applying for opening the bottle. It is. The receiving medium may be, for example, a fixed telephone, a mobile telephone, an email, an SMS (short message service), a website, or the like. The payment method is information indicating whether to pay for gas or water heater charges by bank account transfer, credit card payment, payment by invoice, etc.
The contract name/type is information indicating the contents of the contract between the customer and the business operator, and includes, for example, options for using hot water floor heating.

Next, another example of teacher data will be explained using FIG. 5.
FIG. 5 is a diagram illustrating another example of teacher data.
As such training data, for example, information such as the age of the building, whether the door is closed via the web or by phone, contract period, etc., or a combination thereof, is used. Includes considerations for real estate transaction operations that utilize predictions of lifeline suspension.

The building age referred to here is information about the building in which the lifeline is used, and is included in the lifeline information. The information about the building includes, for example, information indicating a room, which is a unit of sale in a used condominium, and is a candidate for a business partner of a real estate transaction.

The age of construction is narrowed down to properties that have been completed for a certain number of years, and properties that have recently been completed are excluded. For newly built properties that are only a few years old, the asset value has not fallen as much as in the case of newly built properties, and there are not many properties for sale, so competition among real estate agents who buy or broker such properties is fierce. On the other hand, for used properties that have been around for a certain number of years, the asset value can be further increased by renovating them to standard specifications. In addition, it is possible to determine whether a building complies with the new earthquake resistance standards or not based on the age of the building, and to reflect to some extent the characteristics of a building that does not use asbestos, thereby reducing the work burden during renovation. Furthermore, since the new seismic standards are applied to buildings whose building certification date is after June 1, 1981, an upper limit on the building age will be set from the perspective of whether or not the new seismic standards are applied. be able to.

The difference between Web closure and telephone closure is the method of requesting closure, and by distinguishing between Web closure and telephone closure, the added value of sales materials can be increased. To give an example, in the case of web closures, it is generally estimated that a relatively high proportion of young people are involved, and in the case of telephone closures, it is generally estimated that a relatively high proportion of elderly people are involved. .
In the case of web shutoffs, it is estimated that a high proportion of the generation is skilled at gathering information online, so for example, when a gas company advertises real estate transactions online, it has high sales efficiency. It is considered to be a group. In addition, it is considered to be highly compatible with business when conducting telephone sales for real estate transactions to customers whose telephones are closed.
Although we have explained the distinction between web shutoff and telephone shutoff for gas, there are both web and telephone shutoffs for electricity, and they can be distinguished in the same way as in the case of gas, and there is a meaning to the distinction.
To explain further, requests for turning on gas or starting to use electricity may be made online as well as over the phone. For this reason, there is a certain percentage of customers who request both the start and end of energy usage online, so by advertising real estate transactions online, you can create a group with high sales efficiency. can be found.

The reason for including the contract period as a condition is that in the case of rentals, there is a tendency for residents of rental apartments to move and cancel their energy contracts at the time of contract renewal or a little before the contract renewal. This is to increase the proportion of actual demand projects by excluding customers with short energy contract years, as there is a high possibility that the average energy consumption will be shorter than that for residents of condominiums.
The contract period referred to here may include not only a gas contract but also a contract for other than gas, such as electricity. Due to the liberalization of electric power, even gas companies now have attribute information on customers who have contracts for both gas and electricity, or only for electricity.

This will be explained in detail using FIG. 5.
Figures (1) and (2) in Figure 5 combine building age and contract period. That is, in the case of (1), the building is 15 years or more old and the contract period is 3 years or more, and in the case of (2), the building is 10 years or more and the contract period is 3 years or more. This makes it possible to create a group that is likely to include condominiums that are relatively old and can increase the asset value of the properties.
Note that the contract period in (1) to (2) above is a gas contract period, and is different from, for example, (10) to (11) described later, which also includes an electric power contract period.

As an additional note, condominiums include cases of so-called actual demand where the purchaser or his/her related person resides, and so-called investment case where the purchaser or his/her related person does not reside and rents out the condominium, but it is difficult to narrow it down to actual demand. be.

In (3) to (6) of Figure 5, the building age, contract period, and closure request method are combined. In other words, there are four combinations: ``contract period is 3 years or more'', building age is ``5 years or more'' and ``15 years or more'', and the valve closing request method is ``web closing'' or ``telephone closing''. These are (3) to (6) above.

To explain further, in the case of (3) to (4) above, where the building is 15 years old or more and the contract period is 3 years or more, in addition to the case where the resident has lived in the building since it was newly built, the case where the tenant has lived in the building since the building was newly built, as well as those who have moved in after the purchase This may also include the case of a new resident whose residence has changed.
On the other hand, in the case of (5) to (6) above, where the building is 5 years or more old and the contract period is 3 years or more, compared to (3) to (4) above, if the tenant has moved in from the time of new construction, The proportion of residents living in the area will increase. Also, assuming that condominiums are more likely to be sold within 10 years than those built over 10 years ago, in the cases of (5) to (6) above, the condominiums must be built 5 years or older. Therefore, by combining this with information on requests to shut off the tap, it can be expected that the percentage of customers who are thinking of selling the condominium they currently occupy will be higher than in (3) and (4) above.

In (7) to (8) of Figure 5, "Building age is 5 years or more and contract period is 4 years or more" is separated for cases where the shutoff request method is "web shutoff" and "telephone shutoff". . Since the difference between the building year and the contract period is smaller than in cases (5) and (6) above, the percentage of residents who have been in the building since it was newly built is even higher. Generally speaking, people do not experience buying and selling real estate many times in their lives, so in cases (7) to (8) above, the percentage of people who have never bought or sold real estate for the first time is higher than (5) above. ) to (6).

In (9) of FIG. 5, in the case of "web closure", "the building is 15 years or more old and the contract period is 5 years or more", and the threshold of the contract period is two years longer than in (3) above. This excludes residents of rental apartments more than in case (3) above, and is expected to increase the proportion of real demand projects. In addition, since the above-mentioned (9) requires "web closure" as a condition, it is considered that the effect of Internet advertising is high, as described above.

(10) to (11) in Figure 5 indicate that among customers whose building is 5 years or older and whose contract period is 3 years or more, the contract is a gas contract or an electric power contract. '' case and ``telephone shut off'' case are separated.
Cases (10) to (11) above apply to subscribers who have not only gas contracts but also electric power contracts due to electricity liberalization. In general, customers who have a gas and electricity contract with a gas company tend to have higher trust in the gas company than customers who have a contract with only one or the other. When conducting telephone sales for real estate transactions using the name of a gas company, we can expect a high percentage of customers to be interested and listen to the explanation until the end.

To explain further, the age of construction and contract period in (10) to (11) above are the same as in (5) to (6) above, so (5) and (10) above or By combining (6) and (11), it is possible to list customers with only gas contracts, customers with both gas and electricity contracts, and customers with only electricity contracts. Additionally, if you are a customer with only an electricity contract, there is a high possibility that your property will be fully electric.
Also, if you have an electricity contract, the gas has been turned off, but if you have an electricity contract, you may still return to your home periodically, but it is likely that your home will be vacant in the future. . If the electricity and gas contracts have ended, there is a high possibility that the apartment building is currently vacant.

In (12) and (13) of Figure 5, "building age is 2 years or more and contract period is 2 years or more" is separated for cases where the shutoff request method is "web shutoff" and "telephone shutoff". . Since the year of construction and the contract period are the same, the percentage of residents who have lived in the buildings since they were newly built is even higher than in (7) to (8) above.

In (14) and (15) of FIG. 5, "electric power contract holders with a contract period of less than two years" are separated into cases where the shutoff request method is "web shutoff" and "telephone shutoff". (14) to (15) above differ from (12) to (13) above, which are gas contract holders, in that they do not include the building age and are electric power contract holders, so their viewpoints are different. It can be a condition.

In (16) and (17) of Figure 5, "building age is 12 years or more and contract period is 3 years or more" is separated for cases where the shutoff request method is "web shutoff" and "telephone shutoff". . The number of actual demand will be higher than in cases (3) to (4) above, where the building is 15 years or older.

Although not shown, other information included in the lifeline information will be explained.
Such other information includes, for example, the period from the last point of contact between the business operator and the building occupants to the present (when the business operator An example of this is the period since the most recent contact with the resident in the past.
In addition, other information includes the period from the final installation of various equipment (sold or provided by electricity, gas, and telecommunications carriers and installed in the home (e.g., terminal equipment, gas alarms)) to the present (installation An example of the period after the time when
Furthermore, regarding the area where the building is located, for example, statistical information on population changes such as the population outflow rate from the national census, which is conducted to understand the actual situation of people and households in the country, as well as the number of each age group and the proportion of household composition, etc. conditions using statistical information (an example of statistical information about people and/or households). Note that such statistical information may focus on changes in the same area by comparing statistics from a plurality of survey periods, or may focus on differences by comparing statistics between areas in a single survey period.

The following also focuses on sales of real estate transactions.
In other words, if the condominium list includes information indicating room size and/or floor plan, it may be determined whether the property is highly marketable based on the room size (m ² ) and floor plan information. . For example, it is conceivable to increase the weighting in the case of 50m ² to 80m ² , which is popular in actual demand, while decreasing the weighting in the case of studio rooms of 30m ² or less.
In addition, if the condominium list includes address information that directly or indirectly indicates the distance from the nearest station or the distance from the nearest station, for example, information on the number of walking minutes, the name of the nearest station or the number of walking minutes may be You may decide whether the property has high circulation and adjust the weighting. For example, since Yokohama Station has multiple lines and is popular as the nearest station, it may be possible to increase the weighting. Further, for example, since a 25-minute walk from the nearest station requires traveling by bus (bus area), it is assumed that the distribution is low, so it is possible to reduce the weighting.
Furthermore, if the condominium list includes information on past real estate transactions, past real estate transaction data (including external information) indicates that the profit margin per property is large from the closing price (or unit price per tsubo). Properties with high sales efficiency may be identified.

In addition, the customer attribute information includes contract information with energy suppliers and other service contract information that can be confirmed (an example of information indicating multiple types of infrastructure business contracts) as conditions for the contractor's contract. In this case, the contract information may be weighted based on the type or number of contract information. For example, it is thought that customers who have (or had) an electricity contract in addition to gas have a high loyalty to that energy company and therefore may have high sales efficiency, so they are given a higher weight. It is possible that In addition, it is possible to do the same thing if you have a contract for ancillary services provided by an energy provider.

Furthermore, if the customer attribute information includes information indicating the relationship between the energy contract holder and the energy usage suspension applicant, such as when the relationship between the contract holder and the applicant is different, especially if the surname is the same but the relationship is a child, etc. In this case, it is likely that the building was inherited, so the weighting may be increased. In other words, if we receive a gas shutoff request from someone other than the policyholder, and if the person is a relative of the policyholder or has a family relationship, the case may be weighted based on the possibility that the case has been inherited. It is possible to increase the Relatives here refer to blood relatives within the 6th degree of kinship, spouses, and relatives by marriage within the 3rd degree of kinship.

Next, an application example using the prediction result by the sales prediction unit 16 will be described.
An example of such an application is a case where the output unit 17 (see FIG. 3) makes a determination regarding guidance based on the prediction result. Such guidance includes the presence or absence of guidance on services that can be provided due to the vacant house, and/or the degree of guidance when guidance is provided.

Determination regarding guidance can be made using the probability of the prediction result. For example, if the prediction result has a high probability, the level of guidance should be a detailed explanation, and if the prediction result has a low probability, the level of guidance should be a simple explanation. This is a simple explanation.
Additionally, it may be determined whether or not to provide guidance. In this case, when it is determined that guidance is to be provided, the degree of guidance described above is determined.

The guidance in the application example is provided when a customer applies for closure. There are two ways to apply for closure: by telephone (telephone closure) and by the web (web closure). Hereinafter, the determination based on telephone closure will be described as an application example 1, and the determination based on web closure will be described as application example 2.
In addition, there are two ways to make a judgment: one is to make a judgment in advance and respond based on the preliminary judgment results when an application is received, and the other is to make a judgment about a customer when an application is received and respond based on the judgment result. Although there are corresponding cases, application examples will be described without distinguishing between these cases.

6A and 6B are diagrams showing examples of screens output and displayed on the business operator's terminal 30 (see FIG. 1) in the case of a telephone reception, where (a) shows an example of a reception screen for a shutoff request for a telephone operator, (b) partially shows another example of the reception screen, and (c) partially shows another example of the reception screen. That is, (b) and (c) show only the area 62 in (a). Note that the shutoff request in FIG. 6 is used to mean a stoppage or suspension of gas use, but is not limited to this, and the same applies to requests for other infrastructure such as the stoppage or suspension of electricity use.
A reception screen 60 shown in FIG. 6(a) has an area 61 for displaying customer information, and an area 62 located below the area 61 for displaying information related to guidance.

In the area 61 of the reception screen 60, from the top, "customer information for the closure request" is displayed, an input area 61a for inputting the subscriber number, and a display corresponding to the subscriber number input in the input area 61a. There is an area for displaying the "name" and "address" of the person. As mentioned above, the contractor number or name here is an example of specific information that identifies the contractor.
In the example shown in FIG. 6(a), the telephone operator who is answering the phone has entered the subscriber number "○○○○○" obtained from the customer using the telephone T30 (see FIG. 1) into the input area 61a. . Then, in the area 61, "□□□ □□" as "Name" corresponding to the subscriber number "○○○○○" and "△△△Prefecture△△△City△△Ward△" are displayed as "Address". △1-1-1” is displayed.

The "name" and "address" displayed in area 61 are used by the telephone operator to verify the customer's identity, and the telephone operator checks whether the name and address conveyed by the customer match those displayed. Verify your identity.

In the area 62 of the reception screen 60, from the top, "Information on services that can be provided due to the vacant house" is displayed, and in the area 61, the judgment result for the customer with the subscriber number "○○○○○" is displayed. There is a region 63. The area 63 indicating the determination result includes a display 63a indicating whether or not to provide service guidance, and a display 63b indicating the degree of guidance in the case of guidance.

In FIG. 6A, the display 63a of the area 63 indicating the determination result is "Presence or absence of guidance: Yes", and the display 63b is "Extent of guidance: detailed". Therefore, after confirming the identity of the customer, the telephone operator refers to the displays 63a and 63b in the area 63 showing the determination result, and thoroughly explains to the customer during the call the services that can be provided due to the vacant house.
For example, we explain that we provide real estate purchasing or brokerage services, and then ask questions such as whether they are thinking of selling real estate or whether they are interested in this service. Explaining the benefits of the service to the customer and answering questions from the customer. Furthermore, depending on the customer's wishes, materials such as pamphlets can be sent to the customer's home or an appointment can be made to have a sales person visit the customer.

In FIG. 6B, the display 63a of the area 63 indicating the determination result is "Presence or absence of guidance: Yes", and the display 63b is "Level of guidance: simple". The telephone operator briefly explains to the customer during the call the services that can be provided as the house becomes vacant.
For example, just explain that your company provides real estate purchasing or brokerage services, and ask them to contact you if they are interested.

In FIG. 6(c), the display 63a of the area 63 indicating the determination result is "Presence or absence of guidance: None", and the display 63b is "-". The telephone operator only accepts a request to close the tap without explaining to the customer during the call the services that can be provided due to the vacant house.

In this way, since the presence or absence of guidance and the degree of guidance are displayed in the area 63 indicating the determination result, the telephone operator can respond to the call according to the display in the area 63 to make a call requesting a closure. It is possible to reliably provide guidance on real estate transactions to customers who are determined to need guidance on real estate transactions.
In addition, when providing guidance, for example, by providing guidance according to the degree of guidance determined according to the estimation result of the high or low possibility of vacant houses, we can provide thorough explanations to customers whose houses are likely to be vacant, and Real estate transactions can be conducted more efficiently by providing clear explanations to customers with low potential.

7 and 8 are diagrams showing examples of screens that are output and displayed on the user's terminal 20 (see FIG. 1) in the case of web reception, and FIG. /n). Moreover, FIG. 7(b) shows an example of the web reception screen (2/n) for a stopper application, FIG. 8(a) shows another example of the reception screen (2/n), and (b) , shows another example of the reception screen (2/n). "n" here is an integer of 2 or more.
The reception screen 70 shown in FIG. 7A includes messages to the customer such as "We accept your application for closure" and "Please enter your contract number", and an area 71 for displaying customer information.

The area 71 of the reception screen 70 includes an input area 71a for inputting the contractor number, and an area for displaying the "name" and "address" associated with the contractor number input in the input area 71a. be. In the example shown in FIG. 7(a), the same "name" and "address" as shown in FIG. 6(a) described above are displayed.
The user confirms the content, and if it is correct, operates the OK button. This causes the screen to proceed to the screen shown in FIG. 7(b).

For security purposes, a predetermined personal identification number or password may be entered, or the screen shown in FIG. 7(a) may be displayed after the login operation. A password and information input when logging in are examples of specific information that identifies a subscriber.

The reception screen 70 shown in FIG. 7B includes an area 72 for the desired closing date and an explanation display area 73 for displaying an explanation of the real estate transaction service.
The area 72 includes, from the top, a display that says "Please enter the desired date for closing the cap," and an area 72a for the user to input the desired date for closing the cap. In the example shown in FIG. 7(b), the area 72a is a blank field with no input.

The explanation display area 73 includes, from the top, a display of "Information on real estate transaction services" and an area 73a surrounded by a dashed line to indicate the display range of the guidance. To explain further, the example shown in FIG. 7B is a case where the determination results for the user are "Presence or absence of guidance: Yes" and "Level of guidance: Detailed", and the text of the guidance is long. For this reason, a scroll bar 73b is displayed in the area 73a showing the content of the guidance, which is used to shift the notation in the explanation display area 73 up and down to display a portion that is not displayed.

The reception screen 70 shown in FIG. 8A includes an area 72 for the desired closing date and an explanation display area 73 for displaying an explanation of the real estate transaction service. The region 72 is the same as shown in FIG. 7(b), and its explanation will be omitted.
As in the case shown in FIG. 7B, the explanation display area 73 includes, from the top, a display of "Information on real estate transaction services" and an area 73a. Another example shown in FIG. 8(a) is a case where the determination result for the user is "Presence or absence of guidance: Yes" and "Level of guidance: Simple", and when the text of the guidance is as shown in FIG. 7(b). It's relatively short. Therefore, the scroll bar 73b displayed in FIG. 7(b) is not displayed in the area 73a showing the content of the guidance.

The reception screen 70 shown in FIG. 8(b) has an area 72 for the desired date of closure. The region 72 is the same as shown in FIG. 7(b), and its explanation will be omitted.
To explain further, another example shown in FIG. 8(b) is a case where the determination result for the user is "Presence or absence of guidance: None" and "Level of guidance: -", and the notation of guidance is omitted. Ru. Therefore, the reception screen 70 shown in FIG. 8(b) does not have the explanation display area 73 shown in FIG. 7(b) and FIG. 8(a). In the explanation display area 73, the lengths of the explanations are made different (length and shortness of the explanation).

In this way, real estate transaction guidance can be reliably provided to those customers who have made a Web request and are determined to need real estate transaction guidance in the area 73a of the explanation display area 73. Further, when providing guidance, the content of the guidance is changed depending on the degree of guidance. That is, although we will guide customers who are unlikely to have a vacant house due to moving, we will provide a simple explanation assuming that the customer may not have any plans to move. On the other hand, customers whose homes are likely to become vacant due to relocation are likely to be interested in real estate transaction guidance, so we provide detailed explanations while summarizing the main points. As a result, efficient real estate transaction guidance can be provided on the web request screen. Note that although the focus is on the possibility that the house will become vacant due to moving, the degree of guidance may also be determined based on the age group of the customer.
By omitting guidance on real estate transactions for customers who are unlikely to move, it is possible to reduce the burden of screen confirmation associated with customer operations.

Although the embodiments of the present invention have been described above, the technical scope of the present invention is not limited to the range described in the above embodiments. It is clear from the claims that various changes or improvements made to the above-described embodiments are also included within the technical scope of the present invention.

10... Prediction device, 13... Lifeline information acquisition unit, 14... General information acquisition unit, 16... Sales prediction unit

Claims (8)

a lifeline information acquisition unit that acquires information regarding the customer's lifeline;
a general information acquisition department that acquires general information regarding condominiums;
By performing machine learning using information regarding the customer's lifeline, the general information, and information indicating whether the customer has sold a condominium as training data, it is possible to sell the property where the customer resides. a sales prediction department that predicts
Equipped with
If the residence belongs to a condominium, the general information includes at least information regarding the number of sales and rental transactions ;
The sale prediction unit determines the possibility that the property is an investment property based on the information regarding the number of sales and rental transactions .
Prediction device. The information regarding the lifeline includes information regarding at least one of the type, name, model number, installation location, payment method, and service contract of the equipment used for provision by the lifeline.
The prediction device according to claim 1. The general information includes information indicating at least one of the property's size, location, zoning, whether or not it is rented, residence name, condominium company, construction company, and management company.
The prediction device according to claim 1. The prediction device according to claim 1, wherein the general information includes information indicating whether or not the property is rented. 5. The prediction device according to claim 4, wherein the information indicating whether the property is rented includes information indicating at least one of the size of the property, the name of the residence, a sales company, a construction company, and a management company. If the residence belongs to a condominium, the general information further includes information regarding whether the seller is a real estate agent or an individual;
The prediction device according to claim 1. The information regarding the lifeline includes information indicating at least one of the reason for starting use, reception medium, payment method, and contract name/type;
The prediction device according to claim 1. In the information processing device,
A lifeline information acquisition function that acquires information about a customer's lifeline;
A general information acquisition function that acquires general information regarding condominiums;
By performing machine learning using information regarding the customer's lifeline, the general information, and information indicating whether the customer has sold a condominium as training data, it is possible to sell the property where the customer resides. A sales prediction function that predicts the
Realize,
If the residence belongs to a condominium, the general information includes at least information regarding the number of sales and rental transactions ;
The sales prediction function determines the possibility that the property is an investment property based on the information regarding the number of sales and rental transactions .
program.