Sales prediction device, sales prediction method, and sales prediction program

Abstract

Provided is a sell prediction device, etc., for predicting customers with potential to sell a house. The sell prediction device according to the present invention comprises: a sell prediction unit that predicts customers with potential to sell a house, on the basis of a prediction model that predicts customers with potential to sell a house and that is generated based on house / customer data, sell record data, and property appraisal data, and on the basis of a target customer's house / customer data and property appraisal data; and an output unit that outputs the predicted customers.

Description

[Technical field]

[0001] The present disclosure relates to a sales prediction device that predicts customers who are likely to sell a house. [Background technology]

[0002] By knowing which customers are likely to sell their homes, the sales representatives of housing manufacturers can create opportunities to contact new and existing customers and implement various measures to promote sales. However, the timing of selling a home differs for each customer.

[0003] As a system for supporting sales activities, Patent Document 1 discloses an information processing device that predicts the probability of closing a real estate transaction as a reference for determining the selling price of the real estate or adjusting the closing price. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2017-16321 A [Patent Document 2] US Patent Publication US2014 / 0222741 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology disclosed in Patent Document 1 predicts the probability of closing a real estate transaction for customers who have the intention to sell, and does not predict customers who are likely to sell a house. One of the purposes of the present disclosure is to provide a sales prediction device or the like that predicts customers who are likely to sell a house. [Means for solving the problem]

[0006] A sales prediction device according to a first aspect of the present disclosure is A prediction model for predicting customers who may sell a house, which is generated based on the house / customer data, sales performance data, and property appraisal data, and a sales prediction means for predicting customers who may sell a house, based on the house / customer data and property appraisal data of a target customer; An output means is provided for outputting the predicted customers.

[0007] The sales forecasting method according to the second aspect of the present disclosure includes: Based on a prediction model that predicts customers who are likely to sell a home, which is generated based on home / customer data, sales performance data, and property appraisal data, and based on the home / customer data and property appraisal data of the target customers, customers who are likely to sell a home are predicted and the predicted customers are output.

[0008] The third aspect of the present disclosure is a sales forecasting program. The computer is caused to execute a prediction model for predicting customers who are likely to sell a home, which is generated based on home / customer data, sales performance data, and property appraisal data, and to predict customers who are likely to sell a home based on the home / customer data and property appraisal data of the target customers, and to output the predicted customers.

[0009] The sales prediction program may be stored in a non-transitory recording medium that is readable and writable by a computer. Effect of the Invention

[0010] According to the present disclosure, there is provided a sales prediction device and the like that predicts customers who are likely to sell a house. [Brief description of the drawings]

[0011] [Figure 1] FIG. 2 is a block diagram showing an example of a configuration of a sales prediction device and a database according to the first embodiment. [Diagram 2] FIG. 13 is a diagram showing an example of a data structure of residence / customer data. [Diagram 3] FIG. 2 is a diagram illustrating an example of a data structure of sales performance data. [Figure 4] FIG. 2 is a diagram showing an example of a data structure of property appraisal data. [Diagram 5] FIG. 13 is a diagram illustrating an example of a prediction model. [Figure 6] FIG. 13 is a diagram illustrating an example of a prediction model, a conditional branch, and an influence degree. [Figure 7] 4 is a flowchart showing an example of an operation of the sales prediction device according to the first embodiment. [Figure 8] FIG. 11 is a block diagram showing an example of a configuration of a sales prediction device according to a second embodiment. [Figure 9] 10 is a flowchart showing an example of an operation of the sales prediction device according to the second embodiment. [Figure 10] FIG. 2 is a block diagram showing an example of a hardware configuration of a computer. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] <First embodiment> A sales prediction system according to a first embodiment will be described with reference to the drawings. FIG. 1 is a block diagram showing an example of the configuration of the sales prediction system according to the first embodiment. The sales prediction system shown in FIG. 1 includes a sales prediction device 10 and a database 20. An example of the sales prediction device 10 is a computer. An example of the database 20 is a memory or storage.

[0013] The sales prediction device 10 includes a learning unit 11, a sales prediction unit 12, and an output unit 13. The database 20 stores house and customer data 21, sales performance data 22, property appraisal data 23, and prediction model data 24.

[0014] (Database) Fig. 2 is a data structure showing an example of the residence / customer data 21. The residence / customer data 21 shown in Fig. 2 includes customer information, land information, and building information. The residence / customer data 21 is not limited to these pieces of information.

[0015] The customer information is information indicating the attributes of the customer, and includes data items such as customer ID, gender, age, address, occupation, annual income, family structure, land ownership, purchase history (properties), customer rank, loan balance, etc. Note that the customer information is not limited to the above items, and may include other items that may affect the selling behavior of the customer, such as SNS (Social Network Service) history.

[0016] Land information is information that indicates the attributes of land, and includes data such as location, area, zoning, ground information, surrounding environment, building coverage ratio / floor area ratio, orientation, road surface, view / ventilation, etc. The surrounding environment is information about facilities related to life around the house. Facilities include, for example, commercial facilities, medical facilities, schools, parks, etc. Note that land information is not limited to the above items, and may include other items that may affect the selling behavior of customers.

[0017] The building information is information indicating the attributes of the building, and includes data items such as construction date (age), building structure, total floor area / building area, layout, housing performance information, equipment information, garden / garage information, remodeling information, and defects (physical, legal, psychological). The equipment information includes information on electrical equipment such as lighting, outlets, and intercoms, air conditioning equipment such as heating and cooling, and ventilation fans, and water supply and drainage equipment such as kitchens, toilets, and bathrooms. The garden / garage information includes information on the presence and size of gardens and garages. The remodeling information includes repairs to the interior and exterior of the building, and repairs to equipment such as baths, toilets, and kitchens, as well as the repair costs and dates of the repairs. The building information is not limited to the above items, and may include other items that may affect the selling behavior of customers.

[0018] FIG. 3 is a data structure showing an example of sales record data 22. Sales record data 22 shown in FIG. 3 includes sales information, customer information, land information, and building information. Sales information is information related to the sale of houses that have been traded in the past. Sales information includes data with items such as a negotiation ID, selling price, selling period, delivery date, and negotiation person ID. Note that the customer information, land information, and building information in the sales record data in FIG. 3 may be data of items used for the customer information, land information, and building information shown in FIG. 2.

[0019] Fig. 4 is a data structure showing an example of property appraisal data. The property appraisal data 23 shown in Fig. 4 includes appraisal information, land information, and building information. The appraisal information is information related to the appraisal of a property. The appraisal information includes data with items such as an appraisal ID, a customer ID, an appraisal date, an appraisal price (land price, building price), tax, and an appraiser ID.

[0020] The appraisal price is calculated by the transaction intermediary as the price at which the property is likely to be sold. For example, the appraisal price is calculated by comparing it with the most recent sales of properties with similar conditions to the property (house) being appraised. For this reason, the appraisal price varies depending on the transaction intermediary. The seller determines the selling price by referring to multiple appraisal prices. Note that the land information and building information in the property appraisal data in Figure 4 may be data for the items used in the land information and building information shown in Figure 2.

[0021] The prediction model data includes a plurality of prediction models generated by the learning unit 11. The prediction models are generated by a machine learning algorithm based on data related to home sales. The generated prediction models are models that predict customers who are likely to sell a home. The data related to home sales is, for example, sales performance data and property appraisal data.

[0022] (Sales prediction device) The sales prediction device of the first embodiment will be described with reference to the drawings. In FIG. 1, the sales prediction device 10 includes a learning unit 11, a sales prediction unit 12, and an output unit 13. The learning unit 11 acquires house / customer data 21, sales performance data 22, and property appraisal data 23 from a database 20. The learning unit 11 performs machine learning using the acquired house / customer data 21, sales performance data 22, and property appraisal data 23, and calculates (generates) a prediction model and various parameters related to the prediction model. The learning unit 11 performs learning using, for example, house / customer data 21 with sales performance and property appraisal data 23 as explanatory variables, and whether or not the property has been sold from the sales performance data 22 as a target variable, and generates a prediction model. The learning unit 11 registers the generated prediction model and various parameters in the database 20, and updates the prediction model data 24.

[0023] For example, the learning unit 11 generates a prediction model using heterogeneous mixture learning including FAB inference (Factorized Asymptotic Bayesian Inference) and the like. The heterogeneous mixture learning technique is disclosed in, for example, Patent Document 2. Specifically, the learning unit 11 generates a prediction model consisting of multiple prediction formulas, which are regression formulas, and selection conditions for the prediction formulas. In the learning unit 11, a learning algorithm is used to derive multiple regularities (prediction formulas) and condition data (house and customer data, sales record data, property appraisal data) that satisfy the regularities from the house and customer data 21, sales record data 22, and property appraisal data 23.

[0024] FIG. 5 is a diagram showing an example of a prediction model. The prediction model shown in FIG. 5 is expressed by a tree structure based on a plurality of classified prediction formulas (patterns 1, 2, 3, 4, and 5) and selection conditions. Conditional formulas are assigned to the internal nodes of the tree structure, and each prediction formula is assigned to a leaf node. For example, one conditional formula using a feature such as "Has construction at a specific location been completed?" is assigned to the internal node. Note that, in the above description, an example in which the learning unit 11 generates a prediction model by heterogeneous mixture learning, which is one type of white-box type machine learning, has been described, but this is not limiting, and decision tree learning and general linear models may be used. For example, the learning unit 11 may generate a prediction model by using other white-box type machine learning. Alternatively, the prediction model generated by the learning unit 11 may use other known machine learning methods such as deep learning and neural networks.

[0025] The sales prediction unit 12 acquires a prediction model from the prediction model data 24 in the database 20, and inputs information on the target customer's house / customer data 21 and property appraisal data 23 into the acquired prediction model to obtain an output value. For example, in the case of the above prediction model, the sales prediction unit 12 calculates a score by multiplying the degree of influence (weight varies depending on the pattern) of each item (building age, etc.) by the actual value (Y years since construction, etc.) according to each classified pattern (prediction formula). The sales prediction unit 12, for example, predicts customers who may sell a house using house / customer data and property appraisal data with no sales history. Note that house / customer data and property appraisal data with a sales history may be used to predict customers who may sell a house.

[0026] FIG. 6 is a diagram showing an example of a prediction model and the degree of influence on sales. In the prediction model shown in FIG. 6, customers are classified into one of patterns 6, 7, and 8 using "XX construction carried out" and "age of building" (see FIG. 6). The sales prediction unit 21 calculates the score of the customer in each classified pattern using parameters based on the house and customer data 21 of the customer. For example, if the parameters in the classified pattern are YY inspection carried out, warranty construction experience, population density of the construction site, and customer satisfaction, the sales prediction unit 21 calculates the value of each parameter by normalization processing (processing the value so that the average is 0 and the variance is 1). The sales prediction unit 21 sums up the calculated values ​​to obtain the score of the customer. At this time, the calculated parameter values ​​may be weighted, and the weighted values ​​may be summed up to obtain the score of the customer. Note that the above calculation method is an example, and scores may be calculated using more parameters.

[0027] In addition, data on the results of sales prediction (such as whether customers on the predicted customer list actually sold properties) may be fed back to the prediction model.

[0028] The sales prediction unit 12 generates a customer list by arranging the customers with the highest scores in descending order, regarding the customers with higher scores as customers more likely to sell their homes. The generated customer list is delivered to the output unit 13.

[0029] The sales prediction unit 12 extracts customers who are likely to sell a house based on the output value. When extracting a customer, the sales prediction unit 12 preferably also extracts parameters that are the reason why the customer was extracted, that is, the factors that cause the customer to have a high score (the basis for prediction).

[0030] The output unit 13 outputs customer candidates who satisfy a predetermined condition in terms of relevance to sales as a result of prediction by the sales prediction unit 12. A customer who satisfies a predetermined condition means, for example, that the score calculated by the sales prediction unit 12 is equal to or greater than a predetermined value.

[0031] When the output unit 13 receives a customer list of candidate customers generated by the sales prediction unit 12, it displays it on the display unit (not shown) of the sales prediction device 10 or transmits it to the sales representative's terminal (not shown).

[0032] The output unit 13 outputs, for example, a customer list. One example of a customer list is data with items including information on customer ranking, customer ID, and prediction factors (basis for calculation). The customer ranking is a ranking of customers arranged in descending order of score. The sales prediction factors are parameters that have a greater influence on the calculation of the score. For example, when the age of the building, renovation information, and floor plan in the building information, and further the age and family structure in the customer information are parameters that have a greater influence, these are output as sales prediction factors linked to each customer (for example, customer ID, etc.).

[0033] Next, the operation of the sales prediction device 10 of the first embodiment will be described with reference to the drawings. FIG. 7 is a flow chart showing an example of the operation of the sales prediction device 10. The learning unit 11 of the sales prediction device 10 acquires the house / customer data 21, the sales record data 22, and the property appraisal data 23 from the database 20. The learning unit 11 generates a prediction model for predicting customers who may sell a house based on the house / customer data 21, the sales record data 22, and the property appraisal data 23 (step S11). The learning unit 11 registers the generated prediction model in the prediction model data 24 of the database 20.

[0034] The sales prediction unit 12 acquires house and customer data 21 and prediction model data 24 including a prediction model from the database 20. The sales prediction unit 12 predicts customers who are likely to sell a house based on a prediction model that predicts customers who are likely to sell a house and the house and customer data of the target customers (step S12). The sales prediction unit 12 delivers the customers who are likely to sell a house to the output unit 13.

[0035] The output unit 13 outputs candidate customers who satisfy a predetermined condition from among the predicted customers (step S13). The output unit 13 may output a list of candidate customers as a customer list. For example, the output unit 13 transmits the customer list to a terminal (illustrated) of a sales representative. The sales representative can increase sales efficiency by giving priority to customers included in the customer list in sales activities.

[0036] (Effects of the First Embodiment) According to the first embodiment, the learning unit 11 generates a prediction model for predicting customers who may sell a house based on house and customer data, sales performance data, and property appraisal data, and the sales prediction unit 12 predicts customers who may sell a house based on the generated prediction model and the house and customer data and property appraisal data of the target customer. The output unit 13 outputs customer candidates who satisfy predetermined conditions from among the predicted customers. This makes it possible to predict customers who are likely to sell a house.

[0037] <Second embodiment> A second embodiment of the present disclosure will be described with reference to the drawings. A sales prediction device 30 according to the second embodiment has a configuration excluding the learning unit 11 of the sales prediction device 10 according to the first embodiment. The sales prediction device 30 of the second embodiment uses prediction model data 24 stored in the database 20 of the first embodiment. The prediction model data 24 includes a prediction model generated by the learning unit 11 of the first embodiment, etc. Fig. 8 is a block diagram showing an example of the configuration of a sales prediction device according to the second embodiment. The sales prediction device 30 shown in Fig. 8 includes a sales prediction unit 32 and an output unit 33. The sales prediction unit 32 and the output unit 33 have the same functions as the sales prediction unit 12 and the output unit 13 of the first embodiment.

[0038] The sales prediction unit 32 predicts customers who may sell a house based on a prediction model generated based on the house / customer data, sales performance data, and property appraisal data, and the house / customer data and property appraisal data of the target customers. Here, the sales performance data, property appraisal data, and house / customer data use the same data as the sales performance data 22, property appraisal data 23, and house / customer data 21 described in the first embodiment.

[0039] The output unit 33 outputs customers predicted to sell a house. In the sales prediction device 10 of the first embodiment, the output unit 13 outputs customer candidates who satisfy predetermined conditions from among the predicted customers. The output unit 33 of the sales prediction device 30 of the second embodiment removes the constraints imposed by the output unit 13 due to the predetermined conditions.

[0040] FIG. 9 is a flow chart showing an example of the operation of the sales prediction device 30 of the second embodiment. As shown in FIG. 9, according to the sales prediction device 30 of the second embodiment, the sales prediction unit 32 acquires the prediction model data 24, the house / customer data 21, and the property appraisal data 23 from the database 20. The prediction model data 24 includes a prediction model for predicting a customer who may sell a house, which is generated based on the house / customer data 21, the sales performance data 22, and the property appraisal data 23. The sales prediction unit 32 predicts a customer who may sell a house based on the house / house data 21, the property appraisal data 23, and the prediction model (step S22). The output unit 33 outputs the predicted customer (step S22).

[0041] The sales prediction device 30 of the second embodiment can predict customers who are likely to sell a house, similar to the sales prediction device 10 of the first embodiment.

[0042] (Hardware configuration) In each embodiment, some or all of the components in the sales prediction device shown in Figures 1 and 8 can be realized, for example, by using any combination of a computer 60 and a program shown in Figure 10. The computer 60 includes, as an example, the following configuration.

[0043] CPU61 ROM62 RAM63 · Storage device 65 for storing programs 64 and other data A drive device 67 for reading and writing the recording medium 66 Communication Interface 68 Input / output interface 69 for inputting and outputting data Each component of the sales prediction device in each embodiment of the present application is realized by the CPU 61 acquiring and executing a program 64 that realizes these functions. The program 64 that realizes the functions of each component of the sales prediction device is stored in the storage device 65 or the RAM 63 in advance, for example, and is read by the CPU 61 as necessary. The program 64 may be supplied to the CPU 61 via a communication network, or may be stored in the recording medium 66 in advance, and the drive device 67 may read the program and supply it to the CPU 61.

[0044] There are various variations in the method of implementing each device. For example, the sales prediction device may be implemented by any combination of separate information processing devices and programs for each component. In addition, multiple components included in the sales prediction device may be implemented by any combination of a single computer 60 and programs.

[0045] In addition, some or all of the components of the sales prediction device are realized by other general-purpose or dedicated circuits, processors, etc., or a combination of these. These may be configured by a single chip, or may be configured by multiple chips connected via a bus.

[0046] Some or all of the components of the sales prediction device may be realized by a combination of the above-mentioned circuits, etc. and programs.

[0047] When some or all of the components of the sales prediction device are realized by a plurality of information processing devices, circuits, etc., the plurality of information processing devices, circuits, etc. may be centrally arranged or distributed. For example, the information processing devices, circuits, etc. may be realized as a client-server system, a cloud computing system, etc., in a form in which each is connected via a communication network.

[0048] Although the present invention has been described with reference to the present embodiment, the present invention is not limited to the above embodiment. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. [Explanation of symbols]

[0049] 10 Sales Prediction Device 11 Learning Department 12 Sales Forecast Department 13 Output section 20 Database 21 Residences · Customer Data 22 Sales performance data 23 Property appraisal data 24 Prediction model data 30 Sales Prediction Device 32 Sales Forecast Department 33 Output section 60 Computer

Claims

1. a sales prediction means for predicting customers who are likely to sell a home by inputting target home / customer data and property appraisal data with no sales history into a prediction model generated by performing machine learning using home / customer data, sales history data, and property appraisal data related to past home sales, and predicting customers who are likely to sell a home among those with no sales history; and an output means for outputting the predicted customers, The sales prediction device, in which the home / customer data includes at least one of the following customer attributes: customer ID, gender, age, address, occupation, annual income, family composition, land ownership, purchasing history, customer rank, and remaining loan balance.

2. The sales prediction device of claim 1 , further comprising a learning means for performing machine learning using house / customer data, sales performance data, and property appraisal data relating to sales of houses that have been traded in the past, to generate the prediction model.

3. 3. The sales prediction device of claim 2, wherein the prediction model generated by the learning means is configured in a tree structure with a plurality of prediction formulas for predicting customers who are likely to sell a house, and selection conditions in which conditional formulas are assigned to internal nodes and the prediction formulas are assigned to leaf nodes.

4. The sales prediction device according to claim 3, wherein the learning means uses the house / customer data, the sales record data, or the property appraisal data used in machine learning as condition data, and generates the prediction model by deriving the prediction formula and condition data for which the prediction formula is valid from the candidates for the prediction model using a learning algorithm in machine learning.

5. The sales prediction device of claim 4, wherein the sales prediction means extracts, as prediction factors, items that influenced the calculation of a customer score indicating the likelihood of a customer who may sell a house in the prediction model used for the predicted customer.

6. The sales prediction device according to claim 5 , wherein the output means outputs a customer list including the predicted customers and the prediction factors related to the predicted customers.

7. The sales prediction device according to claim 1 , wherein the output means outputs customer candidates who satisfy a predetermined condition from among the predicted customers.

8. A prediction model is generated by performing machine learning using home / customer data, sales performance data, and property appraisal data related to past home sales, to predict customers who are likely to sell a home. The model inputs the target home / customer data and property appraisal data with no sales history into the model, which predicts customers with no sales history who are likely to sell a home. Output the predicted customers, A sales prediction method, in which the home / customer data includes at least one of the following customer attributes: customer ID, gender, age, address, occupation, annual income, family composition, land ownership, purchasing history, customer rank, and remaining loan balance.

9. A prediction model is generated by performing machine learning using home / customer data, sales performance data, and property appraisal data related to past home sales, to predict customers who are likely to sell a home. The model inputs the target home / customer data and property appraisal data with no sales history into the model, which predicts customers with no sales history who are likely to sell a home. and outputting the predicted customers. A sales prediction program in which the home / customer data includes at least one of the following customer attributes: customer ID, gender, age, address, occupation, annual income, family composition, land ownership, purchasing history, customer rank, and remaining loan balance.

Images