Guest proposal device

Abstract

This invention can accurately obtain specific customer service methods suitable for products to be sold by referring to purchased products in the purchase history, thereby efficiently increasing sales. A customer service proposal device includes: a unit that generates a machine learning model by learning from a dataset of successfully sold products, the purchase history of the products before the sale, and the customer service methods used when the products were sold, and outputs a customer service method by inputting the product to be sold and the purchase history; a unit that inputs a dataset of the product to be sold and the purchase history into the machine learning model to generate a customer service method; a unit that calculates the sales amount or sales volume related to the product for which the generated customer service method was implemented; a unit that determines whether the sales amount or sales volume is lower than a predetermined target value; a unit that enables the machine learning model to perform additional learning; and a unit that, if determined to be lower, inputs a dataset of the product to be sold and the purchase history into the machine learning model after additional learning to generate other customer service methods.

Description

Technical Field

[0001] This invention relates to a technique for proposing customer service methods when selling goods. Background Technology

[0002] In the context of selling goods (including services), the quality of customer service can lead to the failure to sell goods that should have been sold, or conversely, it can lead to the easy sale of goods that would otherwise be difficult to sell.

[0003] In addition, Patent Document 1 proposes a system that supports the provision of appropriate customer service and a convenient shopping environment by prompting store clerks with customer service methods appropriate to customers or by setting notification recipients based on the store clerks' job duties.

[0004] Existing technical documents Patent documents Patent Document 1: Japanese Patent Application Publication No. 2016-153935 Summary of the Invention

[0005] The technical problem that the invention aims to solve However, the aforementioned existing technology has the following problem: when operators sell goods they wish to sell to consumers, they cannot obtain specific methods to increase the probability of a successful sale of the goods.

[0006] Therefore, in view of the above problems, the present invention aims to provide a customer service proposal device that can accurately obtain specific customer service methods suitable for the products to be sold by referring to the purchased products in the purchase history, thereby efficiently increasing sales.

[0007] Technical solutions used to solve technical problems A disclosed customer service proposal device is characterized by comprising: a model generation unit, which learns from a dataset of successfully sold products, the purchase history of the products before their sale, and specific customer service methods used when selling the products, to generate a machine learning model that outputs the specific customer service method upon inputting the desired product and the purchase history; a model storage unit, which stores parameters defining the input and output characteristics of the machine learning model, which outputs the specific customer service method based on the input desired product and purchase history; and a customer service method generation unit, which inputs the data of the desired product and purchase history into the machine learning model. The system comprises: a customer service unit, which generates a specific customer service method; a customer service implementation unit, which implements the specific customer service method at the sales site; a verification information calculation unit, which calculates the sales revenue related to the products for which the specific customer service method was implemented; a target achievement determination unit, which determines whether the sales revenue is lower than a predetermined target value; an additional learning unit, which enables the machine learning model to perform additional learning; and a customer service method regeneration unit, which, if the target achievement determination unit determines that the sales revenue is lower than the target value, inputs a dataset of the products to be sold and their purchase history into the machine learning model after additional learning, so that it can regenerate other specific customer service methods.

[0008] Furthermore, another aspect of the hospitality proposal device disclosed herein is characterized by comprising: a model generation unit, which learns from a dataset of successfully sold products, the purchase history of the products prior to their sale, and specific hospitality methods used when selling the products, to generate a machine learning model that outputs the specific hospitality method upon inputting the desired product and the purchase history; a model storage unit, which stores parameters defining the input-output characteristics of the machine learning model, which outputs the specific hospitality method for the input desired product and purchase history; and a hospitality method generation unit, which inputs the desired product and purchase history into the machine learning model. The system comprises: a dataset for generating a specific customer service method; a customer service implementation unit for implementing the specific customer service method at the sales site; a verification information calculation unit for calculating the sales volume related to the products for which the specific customer service method was implemented; a target achievement determination unit for determining whether the sales volume is lower than a predetermined target value; an additional learning unit for enabling the machine learning model to perform additional learning; and a customer service method regeneration unit for, if the target achievement determination unit determines that the sales volume is lower than the target value, inputting a dataset of the expected products and purchase history into the additionally learned machine learning model to generate other specific customer service methods.

[0009] Invention Effects The customer service proposal device disclosed herein can accurately obtain specific customer service methods suitable for products to be sold by referring to the purchased products in the purchase history, thereby efficiently increasing sales. Attached Figure Description

[0010] Figure 1 This is a diagram showing an outline of the guest proposal device according to this embodiment.

[0011] Figure 2 This is a functional block diagram of the guest proposal device in this embodiment.

[0012] Figure 3 This is a diagram representing an example of the learning dataset used in this implementation.

[0013] Figure 4 This is a diagram illustrating an example of the hardware structure of the guest proposal device according to this embodiment.

[0014] Figure 5 This is a flowchart illustrating a processing example of the guest proposal device in this embodiment. Detailed Implementation

[0015] The embodiments for carrying out the present invention will be described with reference to the accompanying drawings.

[0016] (The working principle of the guest proposal device in this embodiment) use Figure 1 and Figure 2 The working principle of the guest proposal device (hereinafter referred to as "this device") 100 of this embodiment will be explained. Figure 1 This diagram illustrates the connection relationships between this device 100 and other devices. Figure 2 This is a functional block diagram of the device 100.

[0017] like Figure 1 As shown, the device 100 is connected to the store terminal 290 via a communication network 300. The communication network 300 can be wired or wireless. The store terminal 290 is a device that notifies the device 100 of the sales and inventory status of the goods sold in the store (including virtual stores), and can be, for example, a POS (Point of Sales) system.

[0018] like Figure 2As shown, the device 100 includes a learning data storage unit 110, a model storage unit 120, a model generation unit 130, a customer service method generation unit 140, a customer service implementation unit 150, a verification information calculation unit 160, a goal achievement determination unit 170, an additional learning unit 180, and a customer service method regeneration unit 190. Furthermore, storage units 110 and 120 are not strictly necessary for this device; alternatively, the device 100 may utilize storage units 110 and 120 provided by an external device.

[0019] The learning data storage unit 110 stores a learning dataset 240, which includes successfully sold products 210, the purchase history of products 210 before their sale (purchased products in the past) 220, and specific customer service methods 230 used when selling products 210. The learning dataset 240 is added to and updated over time. The specific customer service methods 230 include sales pitches to customers, advertisements displayed on screens (digital signage), coupons, etc., or a combination thereof.

[0020] Figure 3 An example of a learning data storage unit 110 is shown. For example... Figure 3 As shown, the learning data storage unit 110 may associate and store, for example, the following: Product 210: Instant miso soup; Purchase history 220: Product group with liver-protecting effects; Hospitality method 230: This product is very effective for hangovers! In addition, the learning data storage unit 110 may associate and store, for example, the following: Product 210: Cosmetics; Purchase history 220: None; Hospitality method 230: Famous actresses also use it! Furthermore, the information stored in the learning data storage unit 110 is not limited to this.

[0021] Model storage unit 120 stores parameters that define the input and output characteristics of the machine learning model 280, which will be described later.

[0022] The model generation unit 130 enables the machine learning model 280 to perform machine learning on the learning dataset 240, generating a machine learning model 280. This machine learning model 280 outputs a specific customer service method 270 by taking into input the desired product 250 and the target customer's purchase history 260. Furthermore, the present invention does not impose any particular limitation on the learning algorithm.

[0023] The customer service method generation unit 140 inputs a dataset of desired products 250 and target customers' purchase history 260 into the machine learning model 280, enabling it to generate a specific customer service method 270. The specific customer service method 270 includes customer-facing sales pitches, advertisements displayed on screens (digital signage), coupons, etc., or a combination thereof.

[0024] The customer service implementation unit 150 processes specific customer service methods 270 at the sales site of the product 250 to be sold. For example, the customer service implementation unit 150 displays the specific customer service methods 270 on an employee terminal or on an in-store display screen.

[0025] The verification information calculation unit 160 calculates the sales revenue or sales volume related to the product 250 to be sold. The verification information calculation unit 160 obtains the sales revenue or sales volume information of the product 250 to be sold from the store terminal 290 and performs a totaling operation.

[0026] The target achievement determination unit 170 determines whether the sales revenue or sales volume related to the product 250, calculated by the verification information calculation unit 160, is lower than the specified target value. The target value can be set appropriately.

[0027] The supplementary learning unit 180 enables the machine learning model 280 to perform supplementary learning using the supplemented and updated learning dataset 240. Furthermore, the input-output characteristics of the machine learning model 280 typically change before and after supplementary learning. Additionally, the information stored in the model storage unit 120 is updated during the processing by the supplementary learning unit 180.

[0028] If the target achievement determination unit 170 determines that the target achievement is below the target, the customer service method regeneration unit 190 inputs a dataset of expected sales products 250 and target customer purchase history 260 into the machine learning model 280 after additional learning, so that it can regenerate other specific customer service methods 270.

[0029] Based on the above working principle, this device 100 can refer to the purchased goods 260 in the purchase history to accurately obtain the specific customer service method 270 suitable for the desired sales of the goods 250, thereby efficiently increasing sales.

[0030] (Hardware structure of the guest proposal device in this embodiment) use Figure 4 The hardware structure of this device 100 will be described as an example. Figure 4 This diagram illustrates an example of the hardware structure of the device 100. (See figure below.) Figure 4 As shown, the device 100 includes a central processing unit (CPU) 510, a ROM (Read-Only Memory) 520, a RAM (Random Access Memory) 530, an auxiliary storage device 540, a communication I / F 550, an input device 560, a display device 570, and a storage medium I / F 580.

[0031] CPU 510 is a device that executes programs stored in ROM 520. It processes data loaded into RAM 530 according to program instructions and controls the overall operation of the device 100. ROM 520 stores the programs and data executed by CPU 510. When CPU 510 executes the program in ROM 520, RAM 530 loads the executed program and data and temporarily saves the calculation data during the operation.

[0032] The auxiliary storage device 540 is a device that stores the OS (Operating System), which is the basic software, the application program of this embodiment, and related data together. For example, it can include a learning data storage unit 110 and a model storage unit 120. The auxiliary storage device 540 can be, for example, an HDD (Hard Disk Drive) or flash memory.

[0033] The Communication I / F550 is used to connect to wired / wireless LAN (Local Area Network), Internet and other communication networks 300, and is an interface for sending and receiving data with other devices (POS systems, etc.) 290 that provide communication functions.

[0034] Input device 560 is a device for inputting data into device 100, such as a keyboard. Display device (output device) 570 is a device composed of an LCD (Liquid Crystal Display) or similar, serving as a user interface for users to use the functions of device 100 or to perform various settings. Storage medium I / F 580 is an interface for transmitting and receiving data with storage media 590 such as CD-ROM, DVD-ROM, and USB memory.

[0035] The various units included in this device 100 can also be implemented by the CPU 510 executing the program corresponding to each unit stored in the ROM 520 or the auxiliary storage device 540. Alternatively, the processing of each unit can be implemented in hardware. The program of this invention can also be read from an external server device via communication I / F 550, or from storage medium 590 via storage medium I / F 580, and then executed on this device 100.

[0036] (A processing example of the guest proposal device in this embodiment) use Figure 5 The processing example of this device 100 will be described. Figure 5 This is a flowchart illustrating the processing procedure of this device 100.

[0037] In S10, the model generation unit 130 enables the machine learning model 280 to perform machine learning on the learning dataset 240, generating a machine learning model 280. This machine learning model 280 outputs a specific customer service method 270 by taking as input the desired product 250 and the target customer's purchase history (purchased products) 260. Furthermore, the learning algorithm of the machine learning model 280 is not particularly limited.

[0038] Furthermore, the parameters of the input-output characteristics of the machine learning model 280 generated by the processing in S10 are specified to be stored in the model storage unit 120.

[0039] In S20, the customer service method generation unit 140 inputs a dataset of desired sales products 250 and target customers' purchase history 260 into the machine learning model 280, enabling it to generate a specific customer service method 270. The specific customer service method 270 includes customer-facing sales pitches, advertisements displayed on screens (digital signage), coupons, etc., and may also be a combination of these.

[0040] Then, in S20, the customer service implementation unit 150 processes the implementation of the specific customer service method 270 at the sales site. For example, the customer service implementation unit 150 displays the specific customer service method 270 on an employee terminal or on an in-store display screen.

[0041] Furthermore, after the processing in S20, the supplementary learning unit 180 performs supplementary learning so that the machine learning model 280 uses the supplemented and updated learning dataset 240. In addition, generally, the input-output characteristics of the machine learning model 280 change before and after supplementary learning. During the processing in the supplementary learning unit 180, the information stored in the model storage unit 120 is also updated.

[0042] In S30, the verification information calculation unit 160 calculates the sales amount or sales volume related to the product 250 to be sold. The verification information calculation unit 160 obtains information related to the sales amount or sales volume of the product 250 to be sold from the store terminal 290 and performs a totaling operation.

[0043] Next, in S30, the target achievement determination unit 170 determines whether the sales revenue or sales volume related to the product 250 calculated in S30 is lower than the specified target value. The target value can be set appropriately.

[0044] In S30, when the determination is "lower than", in S40, the customer service method regeneration unit 190 inputs a dataset of expected sales products 250 and target customer purchase history 260 into the machine learning model 280 after additional learning, so that it regenerates other specific customer service methods 270.

[0045] By performing the above processing, the device 100 can accurately obtain the specific customer service method 270 suitable for the desired product 250 by referring to the purchased products 260 in the purchase history, thereby efficiently increasing sales.

[0046] The embodiments of the present invention have been described in detail above, but the present invention is not limited to this specific embodiment. Various modifications and alterations can be made within the scope of the spirit of the present invention as set forth in the claims.

[0047] Explanation of reference numerals in the attached figures 100 Guest Suggestion Device 110 Learning Data Storage Unit 120 model storage units 130 model generation units 140 Guest Service Method Generation Unit 150 Guest Reception Implementation Units 160 Verification Information Calculation Unit 170 Target Achievement Judgment Unit 180 Additional Learning Units 190 Regeneration Unit for Guest Reception Methods 210 Successfully Sold Products 220 Purchase history of successfully sold products before the sale (purchased products in the previous period). 230 Specific customer service methods when selling goods 240 Learning Dataset 250 Items expected to sell 260. Purchase history of target customers (purchased items) 270 Specific hospitality methods generated by machine learning models 280 Machine Learning Models 290 Store Terminal (POS System) 300 communication network 510 CPU 520 ROM 530 RAM 540 Auxiliary storage device 550 communication interface 560 Input Device 570 Output Device 580 Storage Media Interface 590 Storage Media

Claims

1. A guest proposal device, characterized in that, have: The model generation unit learns from a dataset of successfully sold products, the purchase history of those products before they were sold, and the specific customer service methods used when selling those products, and generates a machine learning model that outputs the specific customer service methods when given input products to be sold and the purchase history. A model storage unit stores parameters that define the input and output characteristics of the machine learning model, which outputs the specific customer service method in response to the input of the desired products for sale and purchase history. The customer service method generation unit inputs a dataset of the desired products and purchase history into the machine learning model, so that it generates a specific customer service method. The customer service implementation unit is responsible for implementing a specific customer service method at the sales site. The verification information calculation unit calculates the sales revenue related to the goods for which the specific customer service method was implemented. The target achievement determination unit determines whether the sales amount is lower than the specified target value; An additional learning unit is added, which enables the machine learning model to perform additional learning; as well as The customer service method regeneration unit, when the target achievement determination unit determines that the target achievement is below the target, inputs a dataset of the expected products and purchase history into the machine learning model after additional learning, so that it can regenerate other specific customer service methods.

2. A guest proposal device, characterized in that, have: The model generation unit learns from a dataset of successfully sold products, the purchase history of those products before they were sold, and the specific customer service methods used when selling those products, and generates a machine learning model that outputs the specific customer service methods when given input products to be sold and the purchase history. A model storage unit stores parameters that define the input and output characteristics of the machine learning model, which outputs the specific customer service method in response to the input of the desired products for sale and purchase history. The customer service method generation unit inputs a dataset of the desired products and purchase history into the machine learning model, so that it generates a specific customer service method. The customer service implementation unit is responsible for implementing a specific customer service method at the sales site. The verification information calculation unit calculates the sales volume related to the goods for which the specific customer service method described above was implemented; The target achievement determination unit determines whether the sales volume is lower than the specified target value; An additional learning unit is added, which enables the machine learning model to perform additional learning; as well as The customer service method regeneration unit, when the target achievement determination unit determines that the target achievement is below the target, inputs a dataset of the expected products and purchase history into the machine learning model after additional learning, so that it can regenerate other specific customer service methods.

3. A customer service proposal method, executed in a computer equipped with a model storage unit, wherein the model storage unit stores parameters defining the input-output characteristics of a machine learning model, the machine learning model performing machine learning on a dataset of successfully sold products, the purchase history of the products before their sale, and specific customer service methods used when selling the products, and outputting the specific customer service methods by inputting the desired products to be sold and the purchase history, characterized in that, The method of presenting ideas to guests includes the following steps: The model generation unit learns from a dataset of successfully sold products, the purchase history of those products before the sale, and the specific customer service methods used when selling those products to generate the machine learning model. This machine learning model outputs the specific customer service methods by taking the desired products and purchase history as input. The step of the hospitality method generation unit inputting a dataset of the desired products and purchase history into the machine learning model to generate a specific hospitality method; The customer service implementation unit performs the steps of implementing a specific customer service method at the sales site; The verification information calculation unit calculates the sales revenue related to the goods for which the specific customer service method was implemented; The step of the target achievement determination unit to determine whether the sales amount is lower than the specified target value; The step of adding a learning unit to enable the machine learning model to perform additional learning; and If the target achievement determination unit determines that the target is below the target, the customer service method regeneration unit inputs a dataset of the desired products and purchase history into the machine learning model after additional learning, so that it can regenerate other specific customer service methods.

4. A customer service proposal method, executed in a computer equipped with a model storage unit, wherein the model storage unit stores parameters defining the input-output characteristics of a machine learning model, the machine learning model performing machine learning on a dataset of successfully sold products, the purchase history of the products prior to their sale, and specific customer service methods used when selling the products, and outputting the specific customer service methods by inputting the desired products to be sold and the purchase history, characterized in that, The method of presenting ideas to guests includes the following steps: The model generation unit learns from a dataset of successfully sold products, the purchase history of those products before the sale, and the specific customer service methods used when selling those products to generate the machine learning model. This machine learning model outputs the specific customer service methods by taking the desired products and purchase history as input. The step of the hospitality method generation unit inputting a dataset of the desired products and purchase history into the machine learning model to generate a specific hospitality method; The customer service implementation unit performs the steps of implementing a specific customer service method at the sales site; The verification information calculation unit calculates the sales volume related to the goods for which the specific customer service method was implemented; The step of the target achievement determination unit to determine whether the sales volume is lower than the specified target value; The step of adding a learning unit to enable the machine learning model to perform additional learning; and If the target achievement determination unit determines that the target is below the target, the customer service method regeneration unit inputs a dataset of the desired products and purchase history into the machine learning model after additional learning, so that it can regenerate other specific customer service methods.

5. A proposal procedure for entertaining guests, characterized in that, Used to enable a computer to perform the guest proposal method as described in claim 3 or claim 4.

Images