Location suggestion device, location suggestion method, and program

Abstract

To solve the problem in which: conventionally, products have sometimes been placed in inappropriate locations.SOLUTION: A location proposal device 1 comprises: a learning information storage unit 113 for storing learning information created using two or more pieces of actual result information having one or more product attribute values and sales information corresponding to a location identifier; a product information reception unit 121 for receiving one or more product attribute values; a sales information acquisition unit 132 for applying, for each of two or more locations, the one or more product attribute values received by the product information reception unit 121 to the learning information, and acquiring sales information for each of two or more pieces of location information; a location determination unit 133 for determining a location where products corresponding to the products attribute values received by the product information reception unit 121 are to be sold using the two or more pieces of sales information acquired by the sales information acquisition unit 132; and a location output unit 141 for outputting the location identifier of the location of interest or location information corresponding to the location identifier. Accordingly, the location proposal device is able to propose appropriate locations to which a plurality of products are placed using the past sales results of the products at each location.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a location proposal device that proposes locations where products should be placed, etc.

Background Art

[0002] In the prior art, there was an information processing device that could present the installation location of a vending machine at a position where commercial areas do not overlap and is an efficient sales location for products (see Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, conventionally, since a person determined the location where products were to be placed, products were sometimes placed in inappropriate locations.

[0005] Also, in the prior art, an appropriate location where products should be placed could not be proposed.

Means for Solving the Problems

[0006] The location suggestion device of the first invention comprises: a learning information storage unit that stores learning information created using two or more performance information items having one or more product attribute values ​​corresponding to a location identifier that identifies the place where a product was sold and sales information regarding the results of selling the product; a product information receiving unit that receives product information having one or more product attribute values; a sales information acquisition unit that applies one or more product attribute values ​​from the product information received by the product information receiving unit to the learning information for each of the two or more locations and acquires sales information for each of the two or more location information items; a location determination unit that determines a location where a product corresponding to the product information received by the product information receiving unit should be sold using the two or more sales information items acquired by the sales information acquisition unit; and a location output unit that outputs a location identifier for the location determined by the location determination unit, or location information corresponding to the location identifier.

[0007] This configuration allows us to suggest appropriate locations for introducing products by utilizing past sales performance data for multiple products at various locations.

[0008] Furthermore, the location suggestion device of the second invention further comprises a location information storage unit that stores two or more location information items, each having one or more location attribute values, associated with a location identifier that identifies a location that is a candidate for selling a product. The learning information is information created using two or more performance information items, each having one or more product attribute values ​​for a product, one or more location attribute values ​​corresponding to a location identifier that identifies a place where the product was sold, and sales information regarding the results of selling the product. The sales information acquisition unit applies one or more product attribute values ​​from the product information received by the product information receiving unit and one or more location attribute values ​​from each location information item to the learning information for each of the two or more location information items, and acquires sales information for each of the two or more location information items.

[0009] This configuration allows for the use of learning information that also takes location attribute values ​​into account, enabling the system to suggest more appropriate locations for product placement.

[0010] Furthermore, in addition to the first invention, the location suggestion device of this third invention stores learning information in the learning information storage unit for each of the two or more locations, which is created using two or more performance information sets having one or more product attribute values ​​for a product and sales information regarding the results of selling the product. The sales information acquisition unit applies one or more product attribute values ​​from the product information received by the product information receiving unit to the learning information corresponding to the location for each of the two or more locations, and acquires sales information.

[0011] This configuration allows us to suggest appropriate locations for introducing products by utilizing past sales performance data for multiple products at various locations.

[0012] Furthermore, the location suggestion device of this fourth invention is a location suggestion device in which, with respect to any one of the first to third inventions, the location determination unit acquires a location identifier or location information corresponding to the sales information that matches the selection criteria when the sales information corresponding to each of the two or more location pieces acquired by the sales information acquisition unit matches the selection criteria.

[0013] This configuration allows us to suggest appropriate locations for introducing products by utilizing past sales performance data for multiple products at various locations.

[0014] Furthermore, the location suggestion device of this fifth invention is a location suggestion device in which, for any one of the first to fourth inventions, the learning information is a learning model obtained by performing machine learning learning processing using one or more product attribute values ​​from two or more performance information as explanatory variables and sales information as the objective variable, and the sales information acquisition unit applies one or more product attribute values ​​from the product information received by the product information receiving unit to the learning model for each of the two or more locations, and acquires sales information for each of the two or more locations through machine learning prediction processing.

[0015] With this configuration, machine learning algorithms can be used to suggest appropriate locations for introducing products, based on past sales performance of multiple products at various locations.

[0016] Furthermore, the location suggestion device of the sixth invention, compared to the fifth invention, is a location suggestion device that acquires two or more performance information from a performance information storage unit which stores two or more performance information items having one or more product attribute values ​​for a product and sales information regarding the results of selling the product, associated with a location identifier that identifies the place where the product was sold; uses one or more product attribute values ​​as explanatory variables and sales information as the target variable to perform machine learning learning processing, acquires learning information which is a learning model, and stores the learning information in a learning information storage unit.

[0017] With this configuration, machine learning algorithms can be used to suggest appropriate locations for introducing products, based on past sales performance of multiple products at various locations.

[0018] Furthermore, the location suggestion device of the seventh invention is a location suggestion device in which, for any one of the first to sixth inventions, one or more product attribute values ​​include one or more pieces of information from the product category, product capacity, product contents, and container type.

[0019] This configuration allows us to suggest appropriate locations for introducing products by utilizing past sales performance data for multiple products at various locations.

[0020] Furthermore, the location suggestion device of the eighth invention is a location suggestion device in which, for any one of the first to seventh inventions, one or more location attribute values ​​include one or more pieces of information from among indoor / outdoor information indicating whether it is indoors or outdoors, a regional identifier specifying the area of ​​the location, location type information indicating the type of location where the vending machine is installed, and customer type information indicating the type of customer.

[0021] This configuration allows us to suggest appropriate locations for introducing products by utilizing past sales performance data for multiple products at various locations.

[0022] Furthermore, the location suggestion device of the ninth invention is a location suggestion device that, in relation to any one of the first to eighth inventions, uses sales information that is one or more of the following: sales indicators, sales composition ratios, or total sales amount.

[0023] With such a configuration, it is possible to propose an appropriate location for introducing a product by using the past sales performance of each location of a plurality of products.

Effects of the Invention

[0024] According to the location proposal device of the present invention, it is possible to propose an appropriate location for introducing a product by using the past sales performance of each location of a plurality of products.

Brief Description of the Drawings

[0025] [Figure 1] Conceptual diagram of the proposal system A in Embodiment 1 [Figure 2] Block diagram of the proposal system A [Figure 3] Flowchart for explaining an operation example of the location proposal device 1 [Figure 4] Flowchart for explaining the first example of the sales information acquisition process [Figure 5] Flowchart for explaining the second example of the sales information acquisition process [Figure 6] Flowchart for explaining the third example of the sales information acquisition process [Figure 7] Flowchart for explaining the fourth example of the sales information acquisition process [Figure 8] Flowchart for explaining the first example of the learning process [Figure 9] Flowchart for explaining the second example of the learning process [Figure 10] Diagram showing the location information management table [Figure 11] Diagram showing the product information management table [Figure 12] Diagram showing the performance information management table [Figure 13] Diagram showing the output example

Modes for Carrying Out the Invention

[0026] The following describes embodiments of the location suggestion device and the like with reference to the drawings. Note that components denoted by the same reference numerals in the embodiments perform similar operations, and therefore, further explanation may be omitted.

[0027] (Embodiment 1) In this embodiment, a location suggestion device that determines where a product should be placed uses learning information created using one or more product attribute values, one or more location attribute values, and sales information, as well as one or more product attribute values ​​for the product to be placed and one or more location attribute values ​​for each location.

[0028] In this specification, information X being associated with information Y means that information Y can be obtained from information X, or information X can be obtained from information Y, and the method of association is irrelevant. Information X and information Y may be linked, may reside in the same buffer, may information X be contained in information Y, or information Y may be contained in information X, and so on.

[0029] Figure 1 is a conceptual diagram of the proposed system A in this embodiment. The proposed system A comprises a location suggestion device 1 and one or more terminal devices 2.

[0030] Location suggestion device 1 is, for example, a so-called server, such as a cloud server or ASP server. Alternatively, location suggestion device 1 may be, for example, a personal computer, tablet device, or smartphone.

[0031] Terminal device 2 is a device used by the user. Terminal device 2 can be, for example, a personal computer, a tablet device, a smartphone, etc., and its type is not limited.

[0032] Figure 2 is a block diagram of the proposed system A in this embodiment.

[0033] The location suggestion device 1 comprises a storage unit 11, a reception unit 12, a processing unit 13, and an output unit 14. The storage unit 11 comprises a location information storage unit 111, a performance information storage unit 112, and a learning information storage unit 113. The reception unit 12 comprises a product information reception unit 121. The processing unit 13 comprises a learning unit 131, a sales information acquisition unit 132, and a location determination unit 133. The output unit 14 comprises a location output unit 141.

[0034] The terminal device 2 includes a terminal storage unit 21, a terminal receiving unit 22, a terminal processing unit 23, a terminal transmission unit 24, a terminal receiving unit 25, and a terminal output unit 26.

[0035] The storage unit 11, which constitutes the location suggestion device 1, stores various types of information. These types of information include, for example, location information (described later), performance information (described later), learning information (described later), a learning module that performs machine learning learning processing, and a prediction module that performs machine learning prediction processing.

[0036] The location information storage unit 111 stores two or more location information entries. Location information is information about a location. A location is where the product to be sold is placed. A location can also be described as a candidate location where the product is dispensed. A location is usually a vending machine that is a candidate for dispensing the product, or a location where one or more vending machines that are candidates for dispensing the product are installed. If a location is a place where two or more vending machines are installed, these two or more vending machines are usually installed adjacent to or near each other.

[0037] Location information is typically associated with a location identifier. Location information usually has one or more location attribute values. However, location information may consist only of a location identifier. A location identifier is information that identifies a location. Examples of location identifiers include the ID of a vending machine, the name of a vending machine, the ID of a location, and the name of a location.

[0038] Location attribute values ​​are attribute values ​​of a place, and can also be called location features or characteristics of a place. Location attribute values ​​are information corresponding to a location identifier that identifies the place where a product was sold. Examples of location attribute values ​​include vending machine attribute values, regional identifiers that specify the area of ​​the location, and addresses. Vending machine attribute values ​​are attribute values ​​of the vending machine. Examples of vending machine attribute values ​​include indoor / outdoor information, location type information, customer information, and number of columns. Indoor / outdoor information is information that indicates whether it is indoors or outdoors. Examples of indoor / outdoor information include "in-location" (e.g., value "1") indicating indoors and "out-location" (e.g., value "0") indicating outdoors. Location type information is information that indicates the type of location where the vending machine is installed. Examples of location type information include "corporate factory," "corporate office," "amusement facility," "hospital," "school," and "other." It is preferable for location type information to be hierarchical. Examples of hierarchical location type information include "office → company," "office → government office," and "other → sports." Note that the vending machine attribute value may be the string mentioned above, or it may be an ID (such as a number) corresponding to the string. Customer information refers to information about customers who use the vending machine. Customer information is, for example, customer type information that indicates the type of customer. Customer type information is, for example, "individual," "corporation," or "government agency."

[0039] Furthermore, if the learning information described later is prepared for each location, the location information storage unit 111 may not be necessary.

[0040] The performance information storage unit 112 stores two or more performance information items. Performance information refers to information about sales performance. Here, the performance information includes one or more product attribute values, one or more location attribute values, and sales information.

[0041] Product attribute values ​​are the attributes of a product, and can also be called product features or characteristics. Examples of product attribute values ​​include category, volume information, container type information, and contents information. Category information indicates the category of the product, for example, "water," "tea," "coffee," "carbonated beverage," and "alcoholic beverage." Volume information indicates the volume of the product, for example, "125ml," "250ml," "350ml," "500ml," and "650ml." Container type information indicates the type of container the product is in, for example, "PET bottle," "can," and "glass bottle." Contents information indicates the contents of the product, for example, "green tea," "oolong tea," "soda," "orange juice," "beer," and "shochu cocktail."

[0042] Sales information refers to information about the results of a product's sale. Examples of sales information include sales metrics, sales composition ratio, total sales amount, whether the sale was successful or not (e.g., "Good (1)" or "Poor (0)"), and a score.

[0043] A sales metric is an indicator that shows how well a product (for example, product A) is selling relatively in a given location. For example, a sales metric can be calculated as follows: "Sales metric = Sales amount of product A in a given location / Average sales amount of all products in a given location". For instance, if the sales amount of product A in a given location is 1000 yen and the average sales amount of all products in that location is 500 yen, then the "Sales metric for product A = 2.0". However, the calculation formula for the sales metric is not specified. In essence, a sales metric can be described as an indicator that corrects for the impact of the number of SKUs based on the sales amount composition ratio.

[0044] Furthermore, the sales composition ratio is the ratio of the sales amount of a particular product to the total sales amount of a single location.

[0045] Furthermore, the score is information obtained using one or more other sales information. The score is a value calculated by an increasing function with one or more of the following as parameters: sales indicators, sales composition ratios, and total sales amount.

[0046] The learning information storage unit 113 stores learning information. Learning information is information created using two or more pieces of performance information. Examples of learning information include a learning model and a correspondence table. The learning information storage unit 113 contains, for example, only one piece of learning information. Furthermore, the learning information in the learning information storage unit 113 exists for each of two or more locations. In such cases, each of the two or more pieces of learning information is associated with a location identifier that identifies the location.

[0047] The learning model is information constructed by the learning unit 131, described later, through machine learning learning processes.

[0048] The learning model is, for example, information obtained by performing a machine learning training process using product attribute values ​​(one or more) and location attribute values ​​(one or more) as explanatory variables, and sales information as the dependent variable.

[0049] Furthermore, the product attribute values ​​of the explanatory variables may be product attribute values ​​obtained using one or more product attribute values ​​from the actual data. Similarly, the location attribute values ​​of the explanatory variables may be location attribute values ​​obtained using one or more location attribute values ​​from the actual data. The sales information, which is the dependent variable, may be sales information obtained using sales information from the actual data.

[0050] The learning model is, for example, information obtained by performing machine learning training using one or more product attribute values ​​from two or more performance data points corresponding to one location identifier as explanatory variables, and sales information corresponding to one location identifier as the target variable. In such cases, the learning model is associated with the location identifier.

[0051] A learning model is the information used in the prediction process of machine learning. A learning model can also be called a learner, classifier, or classification model. The machine learning algorithm can be deep learning, random forest, decision tree, SVM, SVR, etc. Furthermore, various machine learning functions and existing libraries can be used in machine learning, such as the TensorFlow library, the R language's random forest module, fastText, and TinySVM.

[0052] A correspondence table is a table that contains two or more pieces of correspondence information. Correspondence information is, for example, information used to establish a correspondence between a vector whose elements are one or more product attribute values ​​and one or more location attribute values, and sales information.

[0053] Furthermore, the product attribute values ​​that are elements of the vector may be product attribute values ​​obtained using one or more product attribute values ​​that exist in the performance information. Similarly, the location attribute values ​​that are elements of the vector may be location attribute values ​​obtained using one or more location attribute values ​​that exist in the performance information. The sales information that the vector contains may be sales information obtained using sales information that exists in the performance information.

[0054] The correspondence table may, for example, be a table corresponding to a single location identifier. In this case, the correspondence information is, for example, information for establishing a correspondence between a vector whose elements are one or more product attribute values ​​and sales information. The correspondence information has, for example, the above vector and sales information. In this case, the correspondence table is associated with the location identifier.

[0055] The reception desk 12 receives various types of information and instructions. These types of information and instructions include, for example, suggestion instructions, learning instructions, product information, location information, and performance information.

[0056] A proposal instruction is an instruction to propose a location for introducing the product to be sold. Proposal instructions usually include product information. A learning instruction, on the other hand, is an instruction to create learning information.

[0057] Here, "reception" typically refers to the reception of information transmitted via wired or wireless communication lines, but it may also be a concept that includes the reception of information input from input devices such as keyboards, mice, and touch panels, as well as the reception of information read from recording media such as optical discs, magnetic discs, and semiconductor memory.

[0058] If the reception is the reception of information entered from an input device, the input means can be anything, such as a touch panel, keyboard, mouse, or menu screen. Such a reception unit 12 can be implemented using device drivers for input means such as touch panels and keyboards, or control software for menu screens.

[0059] The product information receiving unit 121 receives product information that has one or more product attribute values. The product information receiving unit 121 receives product information from, for example, the terminal device 2. Note that the product information received by the product information receiving unit 121 may consist only of one or more product attribute values.

[0060] The processing unit 13 performs various processes. These processes include, for example, those performed by the learning unit 131, the sales information acquisition unit 132, and the location determination unit 133.

[0061] The learning unit 131 acquires two or more performance information items from the performance information storage unit 112, uses these two or more performance information items to acquire learning information, and stores this learning information in the learning information storage unit 113.

[0062] The learning unit 131, for example, acquires two or more performance information items from the performance information storage unit 112, uses one or more product attribute values ​​acquired using the performance information as explanatory variables, and uses sales information acquired using the performance information as the target variable to perform machine learning training, acquires training information which is the learning model, and stores the training information in the learning information storage unit 113.

[0063] The learning unit 131, for example, acquires two or more performance information from the performance information storage unit 112, uses one or more product attribute values ​​and one or more location identifiers acquired using each of the performance information as explanatory variables, and uses sales information as the target variable to perform machine learning training, acquires training information which is the learning model, and stores the training information in the learning information storage unit 113.

[0064] The learning unit 131, for example, acquires two or more performance information items from the performance information storage unit 112, and constructs a correspondence table for each of the two or more performance information items, which has a vector with one or more product attribute values ​​acquired using the performance information as elements, and correspondence information that associates the sales information contained in the performance information or the sales information acquired using said sales information, and stores it in the learning information storage unit 113.

[0065] The learning unit 131, for example, acquires two or more performance information items from the performance information storage unit 112, acquires a vector whose elements are one or more product attribute values ​​acquired using each performance information item, and one or more location attribute values ​​acquired using each performance information item, and constructs a correspondence table for each of the two or more performance information items that associates the vector with the sales information contained in the performance information item or the sales information acquired using that sales information item, and stores this correspondence table in the learning information storage unit 113.

[0066] The sales information acquisition unit 132 applies one or more product attribute values ​​from the product information received by the product information reception unit 121 to the learning information for each of the two or more locations, and acquires sales information for each of the two or more location locations. Applying one or more product attribute values ​​to the learning information can be thought of as using one or more product attribute values ​​and the learning information.

[0067] The sales information acquisition unit 132, for example, applies one or more product attribute values ​​from the product information received by the product information receiving unit 121 and one or more location attribute values ​​from each location information to the learning information for each of the two or more location information stored in the location information storage unit 111, and acquires sales information for each of the two or more location information.

[0068] The sales information acquisition unit 132 acquires sales information, for example, through machine learning prediction processing. Alternatively, the sales information acquisition unit 132 acquires sales information using, for example, a correspondence table. The following describes each of these two processes in detail. In the case of machine learning (1-1) When using one learning model

[0069] The sales information acquisition unit 132 acquires a learning model from the learning information storage unit 113. The sales information acquisition unit 132 also constructs a vector for each of the two or more location information stored in the location information storage unit 111 using one or more product attribute values ​​from the product information received by the product information receiving unit 121 and one or more location attribute values ​​from each location information. One or more elements included in the vector include product attribute values ​​from the product information, product attribute values ​​acquired from one or more product attribute values ​​from the product information, location attribute values ​​from the location information, or location attribute values ​​acquired from one or more location attribute values ​​from the location information.

[0070] Next, the sales information acquisition unit 132 provides the acquired learning model and vector to the machine learning prediction module for each of the two or more location information, executes the prediction module, and acquires sales information for each of the two or more location information. (1-2) When using two or more learning models for each location

[0071] The sales information acquisition unit 132 acquires two or more learning models for each location from the learning information storage unit 113. Furthermore, for each of the two or more locations, the sales information acquisition unit 132 constructs a vector using one or more product attribute values ​​from the product information received by the product information receiving unit 121, for each of the two or more location information stored in the location information storage unit 111.

[0072] Next, the sales information acquisition unit 132 provides the acquired learning model and vectors to the machine learning prediction module for each of the two or more locations, executes the prediction module, and acquires sales information for each of the two or more locations. (2) When using a correspondence table (2-1) When using one correspondence table (2-1-1) When using correspondence information for maximum similarity

[0073] The sales information acquisition unit 132 constructs a vector for each of the two or more location information stored in the location information storage unit 111, using one or more product attribute values ​​from the product information received by the product information receiving unit 121 and one or more location attribute values ​​from each location information. Next, the sales information acquisition unit 132 calculates the similarity between the constructed vector and the vectors from the two or more corresponding information in the correspondence table for each of the two or more location information.

[0074] Next, the sales information acquisition unit 132 acquires the sales information that corresponds to the vector with the highest calculated similarity from the correspondence information contained in the correspondence table. (2-1-2) When using correspondence information that matches the similarity criteria

[0075] The sales information acquisition unit 132 constructs a vector for each of the two or more location information stored in the location information storage unit 111, using one or more product attribute values ​​from the product information received by the product information receiving unit 121 and one or more location attribute values ​​from each location information. Next, the sales information acquisition unit 132 calculates the similarity between the constructed vector and the vectors from the two or more corresponding information in the correspondence table for each of the two or more location information.

[0076] Next, the sales information acquisition unit 132 retrieves sales information from the correspondence table that corresponds to one or more vectors that satisfy the similarity conditions among the calculated similarity scores. Then, the sales information acquisition unit 132 retrieves representative values ​​(e.g., mean, median) of the one or more sales information that it has acquired. The similarity conditions are, for example, that the similarity is the maximum, that the similarity is above a threshold, that the similarity is greater than a threshold, or that the similarity is within the top N (N is a natural number). (2-2) When using a correspondence table for each of the two or more locations (2-2-1) When using correspondence information for maximum similarity

[0077] The sales information acquisition unit 132 obtains correspondence tables for each of the two or more locations from the learning information storage unit 113. The sales information acquisition unit 132 also constructs a vector for each of the two or more locations using one or more product attribute values ​​from the product information received by the product information receiving unit 121. Next, the sales information acquisition unit 132 calculates the similarity between the constructed vector and the vectors from the two or more correspondence information contained in the acquired correspondence table for each of the two or more locations.

[0078] Next, the sales information acquisition unit 132 acquires the sales information that corresponds to the vector with the highest calculated similarity from the correspondence information contained in the correspondence table. (2-2-2) When using correspondence information that matches the similarity criteria

[0079] The sales information acquisition unit 132 acquires a correspondence table for each of two or more locations from the learning information storage unit 113. The sales information acquisition unit 132 also constructs a vector for each of the two or more locations stored in the location information storage unit 111, using one or more product attribute values ​​that correspond to each location's product information. Next, the sales information acquisition unit 132 calculates the similarity between the constructed vector and the vectors that correspond to each of the two or more locations in the correspondence table.

[0080] Next, the sales information acquisition unit 132 retrieves sales information from the correspondence table that corresponds to one or more vectors that satisfy the similarity conditions among the calculated similarity scores. Then, the sales information acquisition unit 132 retrieves representative values ​​(e.g., mean, median) of the one or more sales information that it has acquired.

[0081] The location determination unit 133 uses two or more sales information acquired by the sales information acquisition unit 132 to determine the location where the product corresponding to the product information received by the product information reception unit 121 should be sold. For example, the location determination unit 133 determines the location corresponding to the best sales information among the two or more sales information acquired by the sales information acquisition unit 132.

[0082] The location determination unit 133, for example, acquires a location identifier or location information corresponding to one or more sales information that matches the selection criteria when the sales information corresponding to each of the two or more location information acquired by the sales information acquisition unit 132 matches the selection criteria.

[0083] The selection criteria are the conditions for selecting sales information. For example, the selection criteria might be that the sales information is the best available, that the sales information (e.g., total sales) is above a certain threshold, or that the amount of sales information is greater than a certain threshold.

[0084] The output unit 14 outputs various types of information. These types of information include, for example, a location identifier or location information corresponding to the location identifier.

[0085] Output usually refers to transmission to terminal device 2, but it may also be a concept that includes display on a screen, projection using a projector, printing with a printer, sound output, storage on a recording medium, and transfer of processing results to other processing devices or other programs.

[0086] The location output unit 141 outputs the location identifier of the location determined by the location determination unit 133, or location information corresponding to the location identifier. When location information is output, it is stored, for example, in the location information storage unit 111.

[0087] The terminal storage unit 21, which constitutes the terminal device 2, stores various types of information. These types of information include, for example, product information, user identifiers, location information, and performance information.

[0088] The terminal reception unit 22 receives various types of information and instructions. These types of information and instructions include, for example, suggestion instructions and learning instructions.

[0089] The means of inputting various information and instructions can be anything, such as a touch panel, keyboard, mouse, or menu screen.

[0090] The terminal processing unit 23 performs various processes. These processes include, for example, converting received information and instructions into information and instructions for transmission. Other processes include, for example, converting received information into information for output.

[0091] The terminal transmission unit 24 transmits various information and instructions to the location suggestion device 1. These various information and instructions include, for example, location suggestion instructions.

[0092] The terminal receiving unit 25 receives various types of information from the location suggestion device 1. These types of information include, for example, a location identifier and location information.

[0093] The terminal output unit 26 outputs various types of information. These types of information include, for example, a location identifier and location information.

[0094] The storage unit 11, location information storage unit 111, performance information storage unit 112, learning information storage unit 113, and terminal storage unit 21 are preferably made of non-volatile recording media, but can also be made of volatile recording media.

[0095] The process by which information is stored in the storage unit 11, etc. is not relevant. For example, information may be stored in the storage unit 11, etc. via a recording medium, information transmitted via a communication line, etc. may be stored in the storage unit 11, etc., or information input via an input device may be stored in the storage unit 11, etc.

[0096] The reception unit 12 and the product information reception unit 121 are preferably implemented by wireless or wired communication means, but may also be implemented by means of receiving broadcasts, device drivers for input means such as touch panels and keyboards, or control software for menu screens.

[0097] The processing unit 13, learning unit 131, sales information acquisition unit 132, location determination unit 133, and terminal processing unit 23 can typically be implemented using a processor, memory, etc. The processing procedures of the processing unit 13, etc., are usually implemented in software, and this software is recorded on a recording medium such as ROM. However, it may also be implemented in hardware (dedicated circuitry). The processor can be a CPU, MPU, GPU, etc., and the type is not limited.

[0098] The output unit 14, the location output unit 141, and the terminal transmission unit 24 are usually implemented by wireless or wired communication means, but may also be implemented by broadcasting means.

[0099] The terminal reception unit 22 can be implemented using device drivers for input means such as touch panels and keyboards, or control software for menu screens, etc.

[0100] The terminal receiving unit 25 is usually implemented by wireless or wired communication means, but it may also be implemented by means of receiving broadcasts.

[0101] The terminal output unit 26 may or may not be considered to include output devices such as a display or speakers. The terminal output unit 26 can be implemented using driver software for an output device, or driver software for an output device and an output device.

[0102] Next, we will explain an example of the operation of the proposed system A. First, we will explain an example of the operation of the location suggestion device 1 using the flowchart in Figure 3.

[0103] (Step S301) The product information receiving unit 121 determines whether or not it has received a proposal instruction containing product information. If it has received a proposal instruction, it proceeds to step S302; if it has not received a proposal instruction, it proceeds to step S307. At this point, the product information receiving unit 121 receives a proposal instruction, for example, from terminal device 2.

[0104] (Step S302) The sales information acquisition unit 132 acquires one or more product attribute values ​​included in the proposal instruction received in step S301.

[0105] (Step S303) The sales information acquisition unit 132 performs sales information acquisition processing using one or more product attribute values ​​obtained in step S302. Four examples of sales information acquisition processing will be explained using the flowcharts in Figures 4, 5, 6, and 7.

[0106] (Step S304) The location determination unit 133 determines one or more locations using the sales information for each location obtained in step S303. Here, the location determination unit 133 obtains, for example, one or more location identifiers corresponding to sales information that matches the selection criteria.

[0107] (Step S305) The location output unit 141 acquires information to be output, which is information corresponding to each of the one or more locations determined in step S304. The information to be output is, for example, a location identifier for each location, location information paired with the location identifier for each location, and location information with the location identifier for each location. For example, the location output unit 141 acquires location information corresponding to each of the one or more locations determined in step S304 from the location information storage unit 111.

[0108] (Step S306) The location output unit 141 outputs the information acquired in step S305. Return to step S301. The location output unit 141 also transmits the information to, for example, the terminal device 2 that sent the suggestion instruction.

[0109] (Step S307) The reception unit 12 determines whether or not it has received a learning instruction. If it has received a learning instruction, it proceeds to step S308; if it has not received a learning instruction, it proceeds to step S309.

[0110] (Step S308) The learning unit 131 performs the learning process. The process returns to step S301. An example of the learning process will be explained using the flowcharts in Figures 8 and 9.

[0111] (Step S309) The reception unit 12 determines whether or not it has received the information. If it has received the information, it proceeds to step S310; if it has not received the information, it returns to step S301.

[0112] (Step S310) The processing unit 13 stores the information received in step S309 in the storage unit 11. The process returns to step S301. The information here may include, for example, location information and performance information.

[0113] In the flowchart in Figure 3, processing is terminated by power-off or processing termination interrupts.

[0114] Next, we will explain the first example of the sales information acquisition process in step S303 using the flowchart in Figure 4. In this example, the sales information acquisition process uses a single learning model to acquire sales information through machine learning prediction.

[0115] (Step S401) The sales information acquisition unit 132 acquires the learning model from the learning information storage unit 113.

[0116] (Step S402) The sales information acquisition unit 132 assigns 1 to counter i.

[0117] (Step S403) The sales information acquisition unit 132 determines whether the i-th location identifier exists in the location information storage unit 111. If the i-th location identifier exists, the process proceeds to step S404; otherwise, it returns to the higher-level process.

[0118] (Step S404) The sales information acquisition unit 132 acquires one or more location attribute values ​​that are paired with the i-th location identifier from the location information storage unit 111.

[0119] (Step S405) The sales information acquisition unit 132 constructs a vector using one or more product attribute values ​​acquired in step S302 and one or more location identifiers acquired in step S403.

[0120] (Step S406) The sales information acquisition unit 132 provides the learning model acquired in step S401 and the vector constructed in step S405 to a machine learning prediction module, executes the module, and acquires the predicted sales information.

[0121] (Step S407) The sales information acquisition unit 132 temporarily stores the sales information acquired in step S406 in a buffer (not shown) paired with the i-th location identifier.

[0122] (Step S408) The sales information acquisition unit 132 increments counter i by 1. Return to step S403.

[0123] Next, we will explain a second example of the sales information acquisition process in step S303 using the flowchart in Figure 5. In this case, the sales information acquisition process uses two or more learning models for each location to acquire sales information through machine learning prediction.

[0124] (Step S501) The sales information acquisition unit 132 assigns 1 to counter i.

[0125] (Step S502) The sales information acquisition unit 132 determines whether the i-th location identifier exists in the location information storage unit 111. If the i-th location identifier exists, the process proceeds to step S503; otherwise, it returns to the higher-level processing.

[0126] (Step S503) The sales information acquisition unit 132 acquires the learning model that is paired with the i-th location identifier from the learning information storage unit 113.

[0127] (Step S504) The sales information acquisition unit 132 constructs a vector using one or more acquired product attribute values.

[0128] (Step S505) The sales information acquisition unit 132 provides the learning model acquired in step S503 and the vector constructed in step S504 to a machine learning prediction module, executes the module, and acquires the predicted sales information.

[0129] (Step S506) The sales information acquisition unit 132 temporarily stores the sales information acquired in step S406 in a buffer (not shown) paired with the i-th location identifier.

[0130] (Step S507) The sales information acquisition unit 132 increments counter i by 1. Return to step S502.

[0131] Next, we will explain a third example of the sales information acquisition process in step S303 using the flowchart in Figure 6. In this example, the sales information acquisition process uses a single correspondence table to acquire sales information.

[0132] (Step S601) The sales information acquisition unit 132 assigns 1 to counter i.

[0133] (Step S602) The sales information acquisition unit 132 determines whether the i-th location identifier exists in the location information storage unit 111. If the i-th location identifier exists, the process proceeds to step S603; otherwise, it returns to the higher-level processing.

[0134] (Step S603) The sales information acquisition unit 132 acquires one or more location attribute values ​​that are paired with the i-th location identifier from the location information storage unit 111.

[0135] (Step S604) The sales information acquisition unit 132 constructs a vector using one or more product attribute values ​​acquired and one or more location identifiers acquired in step S603.

[0136] (Step S605) The sales information acquisition unit 132 assigns 1 to counter j.

[0137] (Step S606) The sales information acquisition unit 132 determines whether the j-th correspondence information exists in the correspondence table of the learning information storage unit 113. If the j-th correspondence information exists, the unit proceeds to step S607; otherwise, the unit proceeds to step S610.

[0138] (Step S607) The sales information acquisition unit 132 acquires the vector of the j-th corresponding information in the correspondence table of the learning information storage unit 113.

[0139] (Step S608) The sales information acquisition unit 132 calculates the similarity between the vector acquired in step S604 and the vector acquired in step S607, associates this similarity with the j-th corresponding information, and temporarily stores it in a buffer (not shown).

[0140] (Step S609) The sales information acquisition unit 132 increments counter j by 1. Return to step S606.

[0141] (Step S610) The sales information acquisition unit 132 determines vectors (corresponding information) corresponding to one or more similarities that satisfy the similarity conditions from the similarities accumulated in step S608.

[0142] (Step S611) The sales information acquisition unit 132 acquires sales information corresponding to each of the one or more vectors determined in step S609 from the correspondence table and temporarily stores it in a buffer (not shown).

[0143] (Step S612) The sales information acquisition unit 132 increments counter i by 1. Return to step S602.

[0144] Next, a fourth example of the sales information acquisition process in step S303 will be explained using the flowchart in Figure 7. In this example, the sales information acquisition process acquires sales information using a correspondence table for two or more locations.

[0145] (Step S701) The sales information acquisition unit 132 assigns 1 to counter i.

[0146] (Step S702) The sales information acquisition unit 132 determines whether the i-th location identifier exists in the location information storage unit 111. If the i-th location identifier exists, the process proceeds to step S703; otherwise, it returns to the higher-level processing.

[0147] (Step S703) The sales information acquisition unit 132 acquires a correspondence table from the learning information storage unit 113 that is paired with the i-th location identifier.

[0148] (Step S704) The sales information acquisition unit 132 constructs a vector using one or more acquired product attribute values.

[0149] (Step S705) The sales information acquisition unit 132 assigns 1 to counter j.

[0150] (Step S706) The sales information acquisition unit 132 determines whether the j-th correspondence information exists in the correspondence table acquired in step S703. If the j-th correspondence information exists, the unit proceeds to step S707; otherwise, the unit proceeds to step S710.

[0151] (Step S707) The sales information acquisition unit 132 acquires the vector of the j-th correspondence information in the correspondence table.

[0152] (Step S708) The sales information acquisition unit 132 calculates the similarity between the vector acquired in step S704 and the vector acquired in step S707, associates this similarity with the j-th corresponding information, and temporarily stores it in a buffer (not shown).

[0153] (Step S709) The sales information acquisition unit 132 increments counter j by 1. Return to step S706.

[0154] (Step S710) The sales information acquisition unit 132 determines vectors (corresponding information) corresponding to one or more similarities that satisfy the similarity conditions from the similarities accumulated in step S708.

[0155] (Step S711) The sales information acquisition unit 132 acquires sales information corresponding to each of the one or more vectors determined in step S709 from the correspondence table acquired in step S703. Here, if the sales information acquisition unit 132 acquires two or more sales information, it acquires a representative value of those two or more sales information as the sales information corresponding to the i-th location identifier. Next, the sales information acquisition unit 132 temporarily stores the acquired sales information, associated with the i-th location identifier, in a buffer (not shown).

[0156] (Step S712) The sales information acquisition unit 132 increments counter i by 1. Return to step S702.

[0157] Next, we will explain the first example of the learning process in step S308 using the flowchart in Figure 8. The first example of the learning process is the case where one learning model is obtained.

[0158] (Step S801) The learning unit 131 assigns 1 to counter i.

[0159] (Step S802) The learning unit 131 determines whether the i-th performance information exists in the performance information storage unit 112. If the i-th performance information exists, the unit proceeds to step S803; otherwise, the unit proceeds to step S807.

[0160] (Step S803) The learning unit 131 obtains one or more product attribute values ​​and one or more location attribute values ​​from the i-th performance information. The learning unit 131 obtains two or more explanatory variables using one or more product attribute values ​​and one or more location attribute values. The explanatory variables may be product attribute values ​​or location attribute values, or information obtained from one or more product attribute values ​​or one or more location attribute values.

[0161] (Step S804) The learning unit 131 obtains the sales information contained in the i-th performance information. If the sales information is not the target variable, the learning unit 131 uses the sales information to obtain the target variable. The target variable is information obtainable from sales information or one or more sales information, and is related to sales. The target variable is usually related to the volume of sales.

[0162] (Step S805) The learning unit 131 constructs a vector using two or more explanatory variables obtained in step S803 and the target variable obtained in step S804. This vector can also be called training data.

[0163] (Step S806) The learning unit 131 increments counter i by 1. Return to step S802.

[0164] (Step S807) The learning unit 131 provides the two or more vectors configured in step S805 to a learning module that performs machine learning learning processing, executes the learning module, and obtains a learning model.

[0165] (Step S808) The learning unit 131 stores the learning model acquired in step S807 in the learning information storage unit 113. It then returns to the higher-level processing.

[0166] In addition, in the flowchart of Figure 8, instead of steps S807 and S808, the learning unit 131 may construct a correspondence table in which each of the two or more vectors constructed in step S805 is the corresponding information, and store the correspondence table in the learning information storage unit 113.

[0167] Next, we will explain a second example of the learning process in step S308 using the flowchart in Figure 9. The first example of the learning process is when a learning model is obtained for each location.

[0168] (Step S901) The learning unit 131 assigns 1 to counter i.

[0169] (Step S902) The learning unit 131 determines whether the i-th location identifier exists. If the i-th location identifier exists, the unit proceeds to step S903; otherwise, it returns to the higher-level processing.

[0170] (Step S903) The learning unit 131 obtains the performance information corresponding to the i-th location identifier from the performance information storage unit 112. The amount of performance information obtained is usually two or more.

[0171] (Step S904) The learning unit 131 assigns 1 to counter j.

[0172] (Step S905) The learning unit 131 determines whether or not the j-th performance information exists among the performance information obtained in step S903. If the j-th performance information exists, the unit proceeds to step S906; otherwise, the unit proceeds to step S910.

[0173] (Step S906) The learning unit 131 obtains one or more product attribute values ​​from the i-th performance information. The learning unit 131 obtains one or more explanatory variables using one or more product attribute values. The explanatory variables may be product attribute values ​​or information obtained from product attribute values.

[0174] (Step S907) The learning unit 131 obtains the sales information contained in the j-th performance information. If the sales information is not the target variable, the learning unit 131 uses the sales information to obtain the target variable. The target variable is the sales information or information obtainable from the sales information, and is related to sales.

[0175] (Step S908) The learning unit 131 constructs a vector using one or more explanatory variables obtained in step S906 and the target variable obtained in step S907. This vector can also be called training data.

[0176] (Step S909) The learning unit 131 increments counter j by 1. Return to step S905.

[0177] (Step S910) The learning unit 131 provides the two or more vectors configured in step S908 to a learning module that performs machine learning learning processing, executes the learning module, and obtains a learning model.

[0178] (Step S911) The learning unit 131 stores the learning model acquired in step S910 in the learning information storage unit 113, paired with the i-th location identifier.

[0179] (Step S912) The learning unit 131 increments counter i by 1. Return to step S902.

[0180] In the flowchart of Figure 9, instead of steps S910 and S911, the learning unit 131 may construct a correspondence table using the two or more vectors constructed in step S908 as corresponding information, and store it in the learning information storage unit 113 in pairs with the i-th location identifier. In this case, a correspondence table for each location identifier will be stored.

[0181] Next, an example of the operation of terminal device 2 will be described. The terminal reception unit 22 receives various instructions and information, such as learning instructions and suggestion instructions. Next, the terminal processing unit 23 converts the received instructions into instructions for transmission. Next, the terminal transmission unit 24 transmits the various instructions to the location suggestion device 1. Next, in response to the transmission of instructions, the terminal reception unit 25 receives various information (for example, information about suggested locations) from the location suggestion device 1. Next, the terminal processing unit 23 converts the received information into information for output. Next, the terminal output unit 26 outputs the information.

[0182] The following describes a specific example of the operation of the proposed system A in this embodiment. In this specific example, the proposed location is a vending machine that sells a newly introduced product, or a location where multiple vending machines that sell a newly introduced product are installed.

[0183] Assume that the location information storage unit 111 of the location suggestion device 1 stores the location information management table shown in Figure 10. The location information management table is a table that manages two or more location information entries. The location information management table manages two or more records that have an "ID," a "location identifier," and "location attribute values." In this case, the "location attribute values" include "vending machine attribute values" and "regional identifiers." In this case, the "vending machine attribute values" include "indoor / outdoor information," "location type information," "customer type information," and "number of columns."

[0184] "ID" is information that identifies a record. Among the "location identifiers," location identifiers containing "V," such as "V001," are information that identifies a single vending machine, and location identifiers containing "P," such as "P001," are information that identifies a location where two or more vending machines are installed. Indoor / outdoor information "1" indicates that the vending machine is installed indoors. Indoor / outdoor information "0" indicates that the vending machine is installed outdoors. For example, location type information "P1" indicates that the vending machine is installed in a "corporate factory." Location type information "P2" indicates that the corresponding vending machine is installed in a "corporate office," location type information "P3" indicates that it is installed in an "amusement facility," location type information "P4" indicates that it is installed in a "hospital," and location type information "P5" indicates that it is installed in a "school." Location type information "P6" indicates that the corresponding vending machine is installed in an "other" location. Customer type information "K1" indicates that the customer is a "factory worker." Customer type information "K2" indicates that the customer is an "office worker". Customer type information "K3" indicates that the customer is a "general user". Each piece of "regional identifier" information (A1, A2, A3, ...) here identifies, for example, a city, town, or ward.

[0185] Furthermore, the storage unit 11 stores the product information management table shown in Figure 11. The product information management table is a table that manages information on two or more products. The product information management table manages two or more records that have "ID", "product identifier", "product name", and "product attribute value". In this case, the "product attribute value" includes "category", "capacity information", "container type information", and "contents information". "ID" is information that identifies the record.

[0186] Furthermore, the performance information storage unit 112 stores the performance information management table shown in Figure 12. The performance information management table is a table that manages a large amount of performance information. The performance information management table manages a large number of records that have "ID" and "performance information". In this case, the "performance information" includes "location identifier", "product identifier", "time information", and "sales information". In this case, the "sales information" includes "sales amount", "sales indicator", "sales composition ratio", and "score".

[0187] "ID" is information that identifies a record. "Time information" is information indicating the time, in this case, the year, month, and day. "Sales amount" is, in this case, the sales amount for the three months immediately preceding the date indicated by the time information. "Sales indicator" and "Sales composition ratio" are also information obtained based on the sales status for the three months immediately preceding the date indicated by the time information. "Score" is, in this case, information calculated by an increasing function "f" that takes two or more of the following as parameters: sales amount, sales indicator, and sales composition ratio. For example, the calculation formula is "score = f(sales amount, sales indicator, sales composition ratio)". It goes without saying that while the performance information management table has a "location identifier" and a "product identifier", the performance information can also be considered to have one or more location attribute values ​​paired with the "location identifier" and one or more product attribute values ​​paired with the "product identifier".

[0188] In this context, the following two specific examples will be explained. Specific example 1 is an example of the process of building a learning model. Specific example 2 is an example of the process of suggesting locations for products to be introduced in the future.

[0189] (Specific example 1) Assume that, at the user's instruction, a learning instruction was transmitted to the location suggestion device 1 via terminal device 2.

[0190] Next, the reception unit 12 of the location suggestion device 1 receives a learning instruction. Then, the learning unit 131 performs the learning process as follows: Specifically, for each record in the performance information management table (Figure 12), the learning unit 131 obtains location attribute values ​​("Indoor / Outdoor Information", "Location Type Information", "Customer Type Information", "Number of Columns", ... "Regional Identifier", ...) that correspond to the location identifier from the location information management table (Figure 10). The learning unit 131 also obtains product attribute values ​​("Category", "Capacity Information", "Container Type Information", "Contents Information", ...) that correspond to the product identifier from the product information management table (Figure 11) for each record in the performance information management table. The learning unit 131 also obtains the "Score", which is sales information, for each record in the performance information management table (Figure 12). Here, we assume that the score is used as the target variable. However, other types of sales information such as sales amount, sales indicator, and sales composition ratio may be used as the target variable.

[0191] Next, the learning unit 131 constructs training data for each record in the performance information management table, using multiple location attribute values ​​and multiple product attribute values ​​as explanatory variables, and the sales information "score" as the objective variable. The training data is, for example, a vector (indoor / outdoor information, location type information, customer type information, number of columns, ..., regional identifier, ..., category, capacity information, container type information, contents information, ..., score).

[0192] Next, the learning unit 131 provides the configured training data to a machine learning module, executes the learning module, and obtains a learning model. Then, the learning unit 131 stores the learning model in the learning information storage unit 113.

[0193] In summary, this specific example demonstrates that the learning unit 131 was able to construct a learning model that learned from performance data. This learning model uses a vector (indoor / outdoor information, location type information, customer type information, number of columns, ..., regional identifier, ..., category, capacity information, container type information, contents information, ...) which is a set of explanatory variables, to obtain the target variable, which is the score.

[0194] (Specific example 2) Suppose the user has now entered a suggestion into terminal device 2 to determine the placement location for a new product. This suggestion includes the product name "Oolong A", product information "<Category> Tea <Volume Information> 350 <Container Type Information> PET bottle> <Contents Information> Oolong tea...", and the number of placement locations "5". Terminal device 2 then receives this suggestion and transmits it to location suggestion device 1.

[0195] Next, the product information receiving unit 121 of the location suggestion device 1 receives the suggestion instruction, which includes product information, from the terminal device 2.

[0196] Next, the sales information acquisition unit 132 acquires the learning model constructed by the processing in Specific Example 1 from the learning information storage unit 113.

[0197] Next, the sales information acquisition unit 132 acquires location attribute values ​​"indoor / outdoor information, location type information, customer type information, number of columns, ..., regional identifier, ..." from each record in the location information management table (Figure 10). Then, for each record in the location information management table, the sales information acquisition unit 132 uses the product attribute values ​​"<category> tea <capacity information> 350 <container type information> PET bottle> <contents information> oolong tea ..." from the received product information and the location attribute values ​​of each record in the location information management table to construct a vector (indoor / outdoor information, location type information, customer type information, number of columns, ..., regional identifier, ..., category, capacity information, container type information, contents information, ...) for each record in the location information management table.

[0198] Next, the sales information acquisition unit 132 provides the vector corresponding to each record in the location information management table and the acquired learning model to the machine learning prediction module, executes the prediction module, and obtains a score. Next, the sales information acquisition unit 132 temporarily stores the acquired score, associating it with the location identifier for each record in the location information management table.

[0199] Next, the sales information acquisition unit 132 acquires five location identifiers corresponding to the top five scores from the acquired scores, in accordance with the selection condition "top 5 scores" indicated by the number of product placement locations "5". Then, the location output unit 141 transmits information including the five location identifiers acquired by the sales information acquisition unit 132 to the terminal device 2.

[0200] Next, terminal device 2 receives and outputs information containing five location identifiers. An example of such output is shown in Figure 13. In Figure 13, underlined parts such as 1301 are anchors (links), and when instructed, location information corresponding to the location identifier is output.

[0201] In summary, this specific example demonstrates that by considering not only product attribute values ​​but also location attribute values ​​for each location, it was possible to propose suitable locations for product placement.

[0202] As described above, according to this embodiment, it is possible to propose an appropriate location for introducing a product by using past sales performance data for multiple products at various locations.

[0203] Furthermore, according to this embodiment, since learning information that also takes into account location attribute values ​​(characteristics of a location) is used, it is possible to suggest a more appropriate location for placing products.

[0204] In this embodiment, the location suggestion device 1 may be a standalone device. In this case, the reception unit 12 and the product information reception unit 121 receive input such as information and instructions from the user. The output unit 14 and the location output unit 141 output information such as a location identifier to a display, for example.

[0205] Furthermore, the processing in this embodiment may be implemented in software. This software may be distributed via software download or the like. Alternatively, this software may be recorded on a recording medium such as a CD-ROM and distributed. This also applies to other embodiments described herein. The software that implements the location suggestion device 1 in this embodiment is the following program. In other words, this program is a program that causes a computer that can access a learning information storage unit, which stores learning information created using two or more performance information items having one or more product attribute values ​​corresponding to a location identifier that identifies the place where a product was sold, and sales information regarding the results of selling the product, to function as a product information receiving unit that receives product information having one or more product attribute values; a sales information acquisition unit that applies the one or more product attribute values ​​of the product information received by the product information receiving unit to the learning information for each of the two or more locations and acquires sales information for each of the two or more location information items; a location determination unit that uses the two or more sales information items acquired by the sales information acquisition unit to determine where the product corresponding to the product information received by the product information receiving unit should be sold; and a location output unit that outputs the location identifier of the location determined by the location determination unit, or location information corresponding to the location identifier. [Industrial applicability]

[0206] As described above, the location suggestion device 1 according to the present invention has the effect of suggesting an appropriate location for placing a product using past sales records of multiple products at each location, and is useful as a server or the like for suggesting locations for placing products. [Explanation of Symbols]

[0207] 1. Location suggestion device 2 Terminal devices 11 Storage Unit 12 Reception Department 13 Processing Unit 14 Output section 21 Terminal storage section 22 Terminal Reception Section 23 Terminal Processing Unit 24 Terminal transmission unit 25 Receiving part of the terminal 26 Terminal output section 111 Location Information Storage Unit 112 Performance Information Storage Unit 113 Learning Information Storage Unit 121 Product Information Reception Department 131 Learning Department 132 Sales Information Acquisition Department 133 Location Determination Section 141 Location Output Section

Claims

1. A learning information storage unit stores learning information created using two or more performance data sets, each having one or more product attribute values ​​corresponding to a location identifier that identifies the place where the product was sold, and sales information relating to the results of the sale of the product. A product information receiving unit that receives product information having one or more product attribute values, A sales information acquisition unit acquires sales information by applying the one or more product attribute values ​​that the product information has received from the product information receiving unit to the learning information, for each of the two or more location information stored in the location information storage unit which stores two or more location information associated with a location identifier that identifies a location that is a candidate for selling the product, A location determination unit acquires sales information that matches the selection criteria, which are conditions for selecting sales information, from among the two or more sales information acquired by the sales information acquisition unit, and acquires a location identifier or location information corresponding to the sales information. The system comprises a location determination unit and a location output unit that outputs the location identifier or location information acquired by the location determination unit, The aforementioned location is, A vending machine from which the product is to be dispensed, and a location where one or more vending machines from which the product is to be dispensed are installed. The one or more product attribute values ​​mentioned above are: This includes one or more pieces of information from the following: product category, product volume, product contents, and product container type. The aforementioned sales information is, A location suggestion device that provides a score using one or more of the following pieces of information: a sales indicator, which is an indicator showing how well a product is selling relatively in a given location; a sales composition ratio, which is the ratio of the product's sales amount to the total sales amount in a given location; total sales amount; or whether sales are good or not.

2. The location information storage unit includes: The system stores two or more location information entries, each having one or more location attribute values, associated with a location identifier that identifies a potential location for selling the product. The aforementioned learning information is information created using two or more performance data sets, each having one or more product attribute values ​​for a product, one or more location attribute values ​​corresponding to a location identifier that identifies the place where the product was sold, and sales information relating to the results of the sale of the product. The aforementioned sales information acquisition unit, The location suggestion device according to claim 1, wherein for each of the two or more location information stored in the location information storage unit, the device applies the one or more product attribute values ​​of the product information received by the product information receiving unit and the one or more location attribute values ​​of each location information to the learning information, and acquires sales information for each of the two or more location information.

3. The learning information storage unit includes: For each of the two or more locations, learning information is stored that is created using two or more pieces of performance data, each containing one or more product attribute values ​​for a product and sales information relating to the results of selling that product. The aforementioned sales information acquisition unit, The location suggestion device according to claim 1, wherein for each of the two or more locations, the device applies the one or more product attribute values ​​of the product information received by the product information receiving unit to the learning information corresponding to the location, and acquires the sales information.

4. The aforementioned learning information is a learning model obtained by performing a machine learning learning process using the one or more product attribute values ​​possessed by the two or more performance data as explanatory variables and the sales information as the target variable. The aforementioned sales information acquisition unit, The location suggestion device according to claim 1, wherein for each of two or more locations, the device applies the one or more product attribute values ​​of the product information received by the product information receiving unit to the learning model, and acquires sales information for each of the two or more locations through predictive processing of machine learning.

5. The location suggestion device according to claim 4, further comprising: a learning unit that acquires two or more performance information from a performance information storage unit which stores two or more performance information having one or more product attribute values ​​for a product and sales information relating to the results of selling the product, associated with a location identifier that identifies the place where the product was sold; a learning unit that uses each of the one or more product attribute values ​​as an explanatory variable and the sales information as an objective variable to perform machine learning learning processing, acquires learning information which is a learning model, and stores the learning information in the learning information storage unit.

6. The location suggestion device according to claim 2, wherein the one or more location attribute values ​​include one or more pieces of information from among indoor / outdoor information indicating whether it is indoors or outdoors, a regional identifier that identifies the area of ​​the location, location type information indicating the type of location where the vending machine is installed, and customer type information indicating the type of customer.

7. A method for causing a computer to perform all the processing that the location suggestion device described in any one of Claims 1 to 6 would perform.

8. A computer, A program for causing a location suggestion device to function as described in any one of claims 1 to 6.

Images