Information processing apparatus, information processing method, and program

Abstract

To provide an information processor, an information processing method and a program for providing information useful for improving CX (customer experience / customer experience) by simulating the behavior of a person at the time of purchase in a store in the real world of various personas by utilizing a simulated customer.SOLUTION: An information processing device 2 includes a simulated-customer generation unit 10 that generates fictitious persona information of a customer, a store information setting unit 11 that sets store information as information in a store, a values estimation unit 12 that estimates values of the simulated customer regarding purchase of a product on the basis of the persona information, a behavior estimation unit 13 that estimates a purchased product and a purchase reason of the simulated customer on the basis of the store information, the persona information, and the values, and a VR purchase behavior estimation unit 16 that visualizes purchase behavior of the simulated customer calculated on the basis of the purchased product and the purchase reason of the simulated customer.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] CX (Customer Experience) refers to the customer experience from considering purchasing a product or service to purchasing and using it, and the value they gain from it. In recent years, for example, in the retail and service industries, there has been a great deal of emphasis on improving CX to gain customer support from various touchpoints in the purchasing and service implementation process, both in physical stores and online.

[0003] Patent Document 1 describes an information processing system in which an information providing device includes an acquisition unit that acquires actual usage data based on access to online content, a creation unit that creates a persona based on the actual usage data, an estimation unit that estimates the persona's usage of the online content, and an evaluation unit that performs a quantitative evaluation of a usability test of the online content when the usage status is estimated. The evaluation unit scores areas for improvement in the online content based on the evaluation index and feature level when the usage status is estimated. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-110843 Summary of the Invention [Problem to be solved by the invention]

[0005] For example, there is a need for initiatives to improve CX that will lead to increased revenue and performance based on an understanding of persona information such as customer hobbies, preferences, and purchasing situations, as well as a framework for carrying out operations more efficiently with limited human resources. Traditionally, understanding personas has been based on access to online content and is limited to online use. Therefore, it is not possible to understand personas while taking into account real-world human behavior, such as in-store purchasing behavior. Furthermore, understanding personas and considering measures to improve CX requires collecting large amounts of actual customer purchasing data. The present invention aims to provide an information processing device, information processing method, and program that can simulate the behavior of various personas when making purchases in a store in the real world using mock customers, and provide information useful for improving CX (customer experience). [Means for solving the problem]

[0006] To solve the above problems, the present invention provides an information processing device that includes a simulated customer generation unit that generates fictitious persona information of customers, a store information setting unit that sets store information as in-store information, a value estimation unit that estimates the simulated customer's values ​​regarding product purchases based on the persona information, a behavior estimation unit that estimates the simulated customer's purchased products and purchasing reasons based on the store information, persona information, and value estimation, and a visualization unit that visualizes the simulated customer's purchasing behavior calculated based on the simulated customer's purchased products and purchasing reasons.In this case, an information processing device can be provided that uses simulated customers to simulate the human behavior of various personas when purchasing in a store in the real world, and provides information useful for improving CX (customer experience).

[0007] Here, for example, the system further includes a 3D data conversion unit that generates 3D data of the store information and the human avatar of the simulated customer, and a human behavior calculation unit that calculates the purchasing behavior of the human avatar of the 3D data based on the purchased products and purchasing reasons of the simulated customer estimated by the behavior estimation unit. In this case, by converting the data into 3D data, the purchasing behavior of the simulated customer can be presented in an easy-to-understand manner. Furthermore, for example, the visualization unit visualizes the movement path, purchasing behavior, and product recognition process of the human avatar based on the purchasing behavior estimated by the human movement calculation unit. In this case, customer information can be obtained from various perspectives, and many insights can be gained for improving store sales. Furthermore, for example, the 3D data conversion unit includes a human avatar generation unit that generates human avatars, which are 3D data of the simulated customers generated by the simulated customer generation unit, and a store 3D data generation unit that generates 3D data of the store interior, shelf layout, product shelving, and products set by the store information setting unit. In this case, by using the 3D data of the human avatars and the store, it is possible to grasp purchasing behavior in a format close to that of a real store. Furthermore, the device may further include a causal inference unit that infers the causal effects of the purchasing behavior of the human avatar, thereby making it possible to understand the reasons why the human avatar performed such purchasing behavior. For example, the causal inference unit allows a user to check the purchasing behavior visualized by the visualization unit, input a question about the purchasing behavior, and answer the input question based on the estimated causal effects. In this case, the user can resolve any questions about the purchasing behavior. Furthermore, for example, the visualization unit visualizes touch points between the simulated customer and the store in association with the process from when the simulated customer obtains information about a product or service to when the simulated customer makes a purchase, allowing the user to deepen their understanding of purchasing behavior. Furthermore, for example, the visualization unit visualizes the difference between the purchasing behavior of target customers and the purchasing behavior of non-target customers, which makes it possible to obtain more insights that can lead to increased sales for non-target customers as well. Furthermore, the system further includes a store information generation unit that generates store information based on store image data, which is information about the store image that the user wants to realize. In this case, even if the user 5 does not have a specific image of the store that they want to realize, a detailed store that is closer to the image can be automatically generated and visualized, thereby supporting the creation of an ideal store and is expected to improve store sales and employee engagement. For example, the store information generating unit generates store information that matches the store image data based on the current store information set in the store information setting unit. In this case, it becomes easier to generate store information that is close to the store image data. The system may further include a similarity calculation unit that calculates the similarity between estimated data based on purchasing behavior and actual data, such as product purchase data at a physical store and data related to customer behavior in the physical store, and a learning unit that performs learning based on the similarity to reduce the difference between the estimated data and the actual data. In this case, it is possible to further improve the accuracy of estimating purchasing behavior.

[0008] The present invention also provides an information processing method in which a processor executes a program stored in a memory to generate fictitious persona information of customers, set store information as in-store information, estimate values ​​related to product purchases of the simulated customers based on the persona information, estimate the products and purchasing reasons of the simulated customers based on the store information, persona information, and values, and visualize the purchasing behavior of the simulated customers calculated based on the products and purchasing reasons of the simulated customers.In this case, an information processing method can be provided that uses the simulated customers to simulate the human behavior of various personas when purchasing in a store in the real world, and provides information useful for improving CX (customer experience).

[0009] Furthermore, the present invention is a program for causing a computer to implement the following functions: generating fictitious persona information of customers; setting store information as in-store information; estimating values ​​related to product purchases of simulated customers based on the persona information; estimating products and purchasing reasons of the simulated customers based on store information, persona information, and values; and visualizing the purchasing behavior of the simulated customers calculated based on the products and purchasing reasons of the simulated customers. In this case, the computer can implement the functions of simulating the human behavior of various personas when purchasing in a store in the real world using the simulated customers and providing information useful for improving CX (Customer Experience). [Effects of the Invention]

[0010] According to the present invention, it is possible to provide an information processing device, an information processing method, and a program that can simulate the human behavior of various personas when making purchases in a store in the real world using mock customers, and provide information useful for improving CX (customer experience). [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a functional configuration of an information processing system according to a first embodiment. [Figure 2] FIG. 2 is a block diagram showing the functional configuration of a 3D data conversion unit. [Figure 3] 4 is a flowchart showing a processing procedure according to the first embodiment. [Figure 4] FIG. 1 is a diagram illustrating a hardware configuration of an information processing apparatus according to an embodiment of the present invention. [Figure 5] FIG. 10 is a block diagram showing the functional configuration of an information processing system according to a second embodiment. [Figure 6] FIG. 10 is a block diagram showing the functional configuration of an information processing system according to a third embodiment. [Figure 7] FIG. 10 is a diagram showing the purchasing behavior of a simulated customer created by the person movement calculation unit. [Figure 8] 10(a) to 10(c) are diagrams showing examples in which the VR behavior estimation unit visualizes the purchasing behavior of a simulated customer. DETAILED DESCRIPTION OF THE INVENTION

[0012] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the accompanying drawings, with reference to first to third embodiments.

[0013] [First embodiment] First, a first embodiment of the information processing system 1 will be described.

[0014] <Overall explanation of Information Processing System 1> FIG. 1 is a block diagram showing the functional configuration of an information processing system 1 according to the first embodiment. 1 shows only the functional configuration related to this embodiment selected from among the various functions of the information processing system 1. The illustrated information processing system 1 estimates customer purchasing behavior in a store. The information processing system 1 includes an information processing device 2, a display unit 3, and an input unit 4. The information processing device 2 is, for example, a desktop PC (Personal Computer) or a notebook PC, and performs processing to estimate the purchasing behavior of a simulated customer using information input by a system user 5 via an input unit 4. The display unit 3 displays the results of the estimation of purchasing behavior by the information processing device 2. The display unit 3 is, for example, a display device such as a liquid crystal display or an organic EL display. The input unit 4 is a user interface such as a mouse, keyboard, etc. Also, by using a touch panel, it is possible to make it function as both the display unit 3 and the input unit 4.

[0015] The information processing device 2 is connected to the display unit 3 and the input unit 4 by wire or wirelessly. The method of connection is not particularly limited, and examples of wired connection methods that can be used include a wired local area network (LAN), a universal serial bus (USB), and RS-232C. Also, examples of wireless connection methods that can be used include Wi-Fi (Wireless Fidelity, registered trademark), Bluetooth (registered trademark), ZigBee (registered trademark), and Ultra-Wide Band (UWB).

[0016] The display unit 3 and the input unit 4 may also be information processing terminals such as a tablet terminal or a smart device. In this case, the tablet terminal or smart device is connected to the information processing device 2 via a communication network. The communication network may be a wired communication line such as an optical communication line or a public telephone line. Alternatively, a wireless communication line such as a mobile phone line or Wi-Fi (Wireless Fidelity, registered trademark) may also be used. Furthermore, the information processing device 2 may also be connected to the input data of the input unit 4 via a server device, a cloud system, or the like.

[0017] The information processing system 1 generally operates as follows. A system user 5, who is a user of the information processing system 1, inputs customer persona information and store information such as store, shelf, product information, and product sales via the input unit 4. In the information processing system 1, the information processing device 2 performs processing to estimate the purchasing behavior of a simulated customer based on the persona information, store information, etc. The results output from the information processing device 2 are then displayed on the display unit 3 and visually presented to the system user 5. This will be explained in detail below.

[0018] <Basic operation> In this embodiment, the information processing device 2 in the information processing system 1 estimates customer behavior in a text space, and when estimating, visualizes in a virtual space the results of a simulation of the purchasing behavior, such as the movement path of a simulated customer's human avatar in a store and the product recognition process. Here, a persona having detailed attribute information of the target customer is used. A persona is a fictitious user image (user image) who is or is thought to be interested in a company's services or products. When setting a persona, the fictitious user image is specified in terms of age, gender, occupation, annual income, residential area, family structure, and so on.

[0019] To achieve this, as shown in FIG. 1, the information processing device 2 includes a simulated customer generation unit 10, a store information setting unit 11, a value estimation unit 12, a behavior estimation unit 13, a 3D (three-dimensional) data conversion unit 14, a person movement calculation unit 15, a VR (Virtual Reality) purchasing behavior estimation unit 16, and a causal inference unit 17.

[0020] As shown in FIG. 1, in the information processing device 2, a system user 5 uses an input unit 4 to input or select information about a fictitious target customer image. The mock customer generation unit 10 generates fictitious customer persona information based on information about a fictitious customer image. The mock customer generation unit 10 generates multiple pieces of persona information (mock customer information, mock customer data) corresponding to the input or selected information. For example, when the system user 5 sets the age, gender, occupation, etc. of a target customer, the mock customer generation unit 10 generates a large amount of persona information related to multiple personas with the same attributes but different lifestyles, preferences, time of day when they visit the store, and situations. This persona information is, for example, text data.

[0021] The store information setting unit 11 sets store information as information within the store. The store information includes the layout of shelves within the store, product allocations, and product information to be arranged on the shelves. To set the store information, the system user 5 uses the input unit 4 to input information about the shelves and product groups of the target store. The store information set by the store information setting unit 11 can use information about actual stores and products as is, or it can set information about shelf allocations and products in a fictitious store. This store information is, for example, text data.

[0022] The value estimation unit 12 estimates the values ​​of the simulated customer regarding product purchases based on the persona information. That is, the value estimation unit 12 estimates the values ​​that the persona holds regarding the products set by the store information setting unit 11. Values ​​refer to customer desires regarding product purchases, etc. Specifically, the value estimation unit 12 estimates the values ​​that the persona holds and the values ​​associated with each product based on information such as sales information for actual products for stores and products and social trends. More specifically, for example, when customers are divided into segments by age (e.g., 20s, 30s, etc.), the purchasing tendencies of customers in each segment can be determined based on the household survey published by the Statistics Bureau of the Ministry of Internal Affairs and Communications, and the values ​​associated with products can be estimated. Furthermore, the desires (e.g., desire for survival, desire for social status) that drive product purchases can be determined, and the values ​​associated with products can be estimated.

[0023] The behavior estimation unit 13 estimates the products purchased by the simulated customer and the reasons for those purchases based on the store information, persona information, and the results of the estimated values. For example, the behavior estimation unit 13 receives input of persona information such as age, gender, occupation, hobbies and preferences, concerns, and the situation at the time of the store visit. The behavior estimation unit 13 then utilizes generative AI such as LLM (Large Language Model) to output the products purchased and the reasons for purchase.

[0024] The purchasing behavior simulated by the behavior estimation unit 13 is basically carried out in text space, making it difficult to imagine the specific behavior of customers in a store. Therefore, by simulating the purchasing behavior of a persona in a virtual space, the behavioral, cognitive, and thought processes leading up to a purchase in the store can be visualized and reproduced. This makes it possible to obtain customer information from various perspectives and gain many insights for improving store sales.

[0025] The 3D data conversion unit 14 converts store information and simulated customers, which are primarily set in text space, into 3D data. As a result, the 3D data conversion unit 14 generates 3D data for the store information and human avatars of the simulated customers. Store information such as store layout and shelf allocation may be expressed not only as text data but also as image data. Even in this case, the store information is converted into 3D data based on the image and text data to generate store 3D data, and simulated customers linked to personas are converted into human avatars in 3D space.

[0026] The human movement calculation unit 15 calculates the purchasing behavior of the 3D data human avatar based on the purchased products and purchasing reasons of the simulated customer estimated by the behavior estimation unit 13. Specifically, the human movement calculation unit 15 calculates detailed actions and movements of the human avatar as purchasing behavior, such as the route and movement within the store, and what products and advertising information the person recognized before making a purchase, based on the purchased products and purchasing reasons of the persona calculated by the behavior estimation unit 13, the 3D store data generated by the 3D data conversion unit 14, and the human avatar. The human movement calculation unit 15 creates the purchasing behavior, for example, as text data.

[0027] The VR purchasing behavior estimation unit 16 is an example of a visualization unit, and visualizes the purchasing behavior of a simulated customer, calculated based on the products purchased by the simulated customer and their reasons for purchase. More specifically, the VR purchasing behavior estimation unit 16 visualizes the movement path, purchasing behavior, and product recognition process of a human avatar based on the purchasing behavior estimated by the human movement calculation unit 15. The VR purchasing behavior estimation unit 16 visualizes purchasing behavior within a store in a virtual space based on the 3D data of the store layout, shelves, products, and persona human avatars created by the 3D data conversion unit 14, and the detailed purchasing behavior of the human avatars estimated by the human movement calculation unit 15. By recreating consumer purchasing behavior in a virtual space as human movements, which is difficult to express using text alone, it is possible to obtain more information about purchasing behavior, increasing the likelihood of identifying issues and measures that can lead to increased sales.

[0028] The causal inference unit 17 infers the causality of the purchasing behavior of the person avatar. This allows the user to understand the reasons for the purchasing behavior. The VR purchasing behavior estimation unit 16 can interact with the system user 5 via the input unit 4 by using the causal inference unit 17. For example, while the purchasing behavior of a persona person avatar is being reproduced in a virtual space, information about a product that the person avatar recognized and became interested in is presented on the display unit 3. The system user 5 can inquire of the information processing system 1 at any time about why the person avatar became interested in that product and obtain a response. In this case, the causal inference unit 17 can also be said to allow the user to check the purchasing behavior visualized by the VR purchasing behavior estimation unit 16 on the display unit 3, input a question about the purchasing behavior via the input unit 4, and answer the input question based on the inferred causality. This allows the system user 5 to resolve any questions about purchasing behavior. In this way, by utilizing simulation interactively, it is expected that many insights can be gained that will lead to useful ideas that will increase sales.

[0029] <Detailed explanation of the behavior estimation unit 13> The simulated customer generation unit 10 generates multiple simulated customers corresponding to the persona information of the targeted customers. The behavior estimation unit 13 outputs a large amount of information on the purchased products and purchasing reasons of multiple simulated customers with personas having similar attributes, and performs processing that links to analyzing customer purchasing behavior in various ways based on the output data. For example, the purchased products and purchasing reasons of the large amount of output personas can be statistically processed to determine the most frequently purchased products and the most common reasons for behavior. Furthermore, it becomes possible to calculate the characteristic behavior of each persona from the differences between these statistically most frequently purchased products and purchasing reasons and the purchased products and purchasing reasons of each persona. In this way, by estimating the most likely purchases, reasons for purchase, and characteristic behaviors from multiple persona information of target customers, it is possible to gain more insights into store measures that will lead to increased sales and improved CX.

[0030] <Detailed explanation of 3D data conversion unit 14> FIG. 2 is a block diagram showing the functional configuration of the 3D data conversion unit 14. As shown in FIG. As shown in FIG. 2, the 3D data conversion unit 14 is made up of a human avatar generation unit 20 and a store 3D data generation unit 21.

[0031] The human avatar generation unit 20 generates a human avatar, which is 3D data of the simulated customer generated by the simulated customer generation unit 10. Specifically, the human avatar generation unit 20 uses the persona information generated by the simulated customer generation unit 10 to generate 3D data (3D human avatar) of a virtual human character, such as height, physique, sex, clothing, and skeletal information.

[0032] The store 3D data generation unit 21 generates 3D data (store 3D data) of the store interior, shelf layout, product shelving plan, and products set by the store information setting unit 11. By using 3D data of people's avatars and stores, it is possible to understand purchasing behavior in a manner similar to that of a physical store.

[0033] <Detailed explanation of the human movement calculation unit 15> The person behavior calculation unit 15 estimates not only the person's behavior and actions, but also the purchasing behavior and thought process, such as how the person thought when recognizing an interesting product or advertisement during the purchasing behavior, and how the recognition result during the purchasing behavior affected the purchasing result.

[0034] <How to display estimated VR behavior> A method for presenting the simulation results of the VR purchasing behavior estimation unit 16 on the display unit 3 will be described. To review the simulation results visualized on the display unit 3 and consider measures to enhance customer experience, it is important to understand customer purchasing behavior at touchpoints between the store and the customer. In this context, a touchpoint refers to a point of contact between the store and the customer that may cause some change or influence. Examples of touchpoints include when a customer looks at a product, picks up a product, or pauses in front of a product. To achieve this, it is possible to display the simulation results using an AIDMA model, which classifies the psychological processes consumers go through from obtaining information about a product or service to purchasing it into five steps: Attention, Interest, Desire, and Memory. Specifically, the touchpoints with the customer in the simulation performed by the VR purchasing behavior estimation unit 16 are assigned to corresponding steps in the AIDMA model. The system user 5 can then focus on customer touchpoints at specific steps within the five steps to deepen their understanding of purchasing behavior. In addition to the AIDMA model, customer journey models or other models that consider more detailed and complex processes may also be used. In this case, it can be said that the VR purchasing behavior estimation unit 16 visualizes the touch points between the simulated customer and the store in association with the process from when the simulated customer obtains information about a product or service to when the customer purchases it.

[0035] The VR purchasing behavior estimation unit 16 can also visualize and present on the display unit 3 the differences between the estimated purchasing behavior of the customer persona set as a target by the system user 5 and the estimated purchasing behavior of customer personas other than the target. For example, the VR purchasing behavior estimation unit 16 presents information on the differences between the estimated results for each persona, such as showing that the results of a purchasing behavior simulation in which the target customer persona is expected to show a favorable reaction show that a different customer persona shows the most favorable reaction. By also presenting the estimated purchasing behavior of customers other than the target customer persona, it becomes possible to obtain more insights that can lead to increased sales for customers other than the target customer.

[0036] <Processing Procedure> Next, the processing procedure of the information processing system 1 will be described with reference to FIG. FIG. 3 is a flowchart showing a processing procedure according to the first embodiment. In step S101, the system user 5 accesses the information processing device 2 via the input unit 4 and sets persona information of a target customer and store information such as the area, store, and product. Next, in step S102, the simulated customer generating unit 10 generates a plurality of pieces of persona information as simulated customer data that match the persona information set by the system user 5.

[0037] In step S103, the store information setting unit 11 sets the shelf layout of the store, the shelf allocation of products, etc. so as to match the store information set by the system user 5 in step S101. In step S104, the value estimation unit 12 estimates values ​​such as the target customer and the product desire level based on information such as the simulated customer data, store information, store product sales information, and social trends set by the system user 5 in step S101.

[0038] In step S105, the behavior estimation unit 13 estimates the products purchased by the simulated customer and the reasons for those purchases, based on the simulated customer data generated in step S102, the store information set in step S103, and the values ​​estimated in step S104. In step S106, the 3D data conversion unit 14 converts the simulated customer into a human avatar in a 3D space based on the persona information.

[0039] In step S107, the 3D data conversion unit 14 converts the store layout and shelf allocation information set by the store information setting unit 11 mainly in the form of text data or two-dimensional image data into 3D data to generate store 3D data. In step S108, the person movement calculation unit 15 calculates detailed purchasing behavior in the store and person movements during the purchasing process based on the persona's purchased items and purchasing reasons calculated by the behavior estimation unit 13, and the store 3D data and person avatar generated by the 3D data conversion unit 14.

[0040] In step S109, the VR purchasing behavior estimation unit 16 displays on the display unit 3 the purchasing behavior of a simulated customer in a store in a 3DCG (Computer Graphics) virtual space based on the 3D data of the store layout, shelves, products, and persona avatars created by the 3D data conversion unit 14, and the detailed purchasing behavior and movements of the avatars estimated by the human movement calculation unit 15, thereby visualizing the simulation results. In step S110, the VR purchasing behavior estimation unit 16 uses the causal inference unit 17 to interact with the system user 5. That is, the system user 5 inputs a question about the purchasing results and purchasing process obtained by the simulation via the input unit 4. In response, the VR purchasing behavior estimation unit 16 uses the causal inference unit 17 to obtain an answer to the question and presents it to the system user 5 on the display unit 3.

[0041] In the processing procedure of Fig. 3, after the behavior of the human avatar at the time of purchase is confirmed by simulating the purchasing behavior of a mock customer in a store in a virtual space in step S109, instructions are executed using the causal inference unit 17 in step S110. However, during step S109, that is, while the behavior of the human avatar is being confirmed, the system user 5 may input a question or the like and execute the processing of S110 in which the system user 5 answers the question. In this case, since interaction becomes possible while the purchasing behavior simulation is being reproduced and confirmed, there is a high possibility that the opportunities and frequency of interaction between the system user 5 and the information processing system 1 will increase, making it easier to obtain insights that will lead to improvements in store sales and CX.

[0042] <Hardware configuration> FIG. 4 is a diagram showing the hardware configuration of the information processing device 2 according to this embodiment. The information processing device 2 according to this embodiment is realized by a computer 1000 having a configuration as shown in FIG.

[0043] A computer 1000 is connected to an output device 1001 and an input device 1002, and has a configuration in which an arithmetic unit 1003, a primary storage device 1004, a secondary storage device 1005, an output I / F (Interface) 1006, an input I / F 1007, and a network I / F 1008 are connected via a bus 1009. For example, in the information processing system 1, the output device 1001 can be regarded as the display unit 3, and the input device can be regarded as the input unit 4.

[0044] The arithmetic device 1003 operates and executes various processes based on programs stored in the primary storage device 1004 and the secondary storage device 1005, programs read from the input device 1002, etc. The arithmetic device 1003 is realized by a processor such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field Programmable Gate Array).

[0045] The primary storage device 1004 is a memory device such as RAM (Random Access Memory) that temporarily stores data used for various calculations by the arithmetic device 1003. The secondary storage device 1005 is a storage device in which data used for various calculations by the arithmetic device 1003 and various databases are registered, and is realized by a ROM (Read Only Memory), HDD (Hard Disk Drive), SSD (Solid State Drive), etc.

[0046] The secondary storage device 1005 is realized by an internal storage device or an external storage device. The secondary storage device 1005 may also be a storage medium such as a USB memory or an SD (Secure Digital) card. The secondary storage device 1005 may also be an online storage device such as a cloud storage device, a NAS (Network Attached Storage), a file server, or the like.

[0047] The output I / F 1006 is an interface for transmitting information to be output to an output device 1001 that outputs various information such as a display, projector, or printer, and is realized by a connector conforming to a standard such as USB (Universal Serial Bus) or HDMI (registered trademark) (High Definition Multimedia Interface).

[0048] The input I / F 1007 is an interface that receives information from various input devices 1002 such as a mouse, keyboard, keypad, and buttons, and is realized by, for example, a USB. The output I / F 1006 and the input I / F 1007 may be connected wirelessly to the output device 1001 and the input device 1002, respectively. The output device 1001 and the input device 1002 may be integrated into one device, such as a touch panel. In this case, the output I / F 1006 and the input I / F 1007 may also be integrated into one device as an input / output I / F.

[0049] The network I / F 1008 receives data from other devices via the network 1010 and sends it to the arithmetic device 1003, and also transmits data generated by the arithmetic device 1003 to other devices via the network 1010. The arithmetic unit 1003 controls the output device 1001 and the input device 1002 via the output I / F 1006 and the input I / F 1007. For example, the arithmetic unit 1003 loads a program from the input device 1002 or the secondary storage device 1005 onto the primary storage device 1004 and executes the loaded program.

[0050] For example, when the computer 1000 functions as the information processing device 2, the arithmetic unit 1003 of the computer 1000 executes a program loaded onto the primary storage device 1004 to realize each process and function of the information processing system 1 as shown in Fig. 1. The arithmetic unit 1003 of the computer 1000 may also load a program acquired from another device via the network I / F 1008 onto the primary storage device 1004 and execute the loaded program. The arithmetic unit 1003 of the computer 1000 may also cooperate with the other device via the network I / F 1008 to call and use the functions and data of a program from another program of the other device.

[0051] [Second embodiment] Next, a second embodiment of the information processing system 1 will be described.

[0052] <Overall explanation of Information Processing System 1> The information processing system 1 may be implemented in various different forms other than the first embodiment described above. Below, a method for automatically generating store information based on information about a desired store and then simulating purchasing behavior will be described. When a system user 5 inputs conditions related to the desired store into the information processing system 1, store information and 3D data that match the conditions are automatically generated, and purchasing behavior in the store is simulated in a virtual space.

[0053] <Basic operation> FIG. 5 is a block diagram showing the functional configuration of an information processing system 1 according to the second embodiment. 5 includes an information processing device 2, a display unit 3, and an input unit 4, similar to the information processing system 1 described in Fig. 1. On the other hand, the information processing system 1 illustrated in Fig. 5 differs in that it includes an automatic store information generation unit 101 that inputs store image data 100 to the information processing device 2. Therefore, the following explanation will mainly focus on this difference.

[0054] As shown in FIG. 5, in the information processing device 2, a system user 5 inputs into an input unit 4 store image data 100, which is conditions relating to the store image that the system user 5 wants to realize, such as the layout and interior of the store. The automatic store information generation unit 101 is an example of a store information generation unit, and generates store information based on store image data 100, which is information about the store image that the user wants to realize. The automatic store information generation unit 101 generates store information that matches the store image data 100 based on the current store information set in the store information setting unit 11. That is, the automatic store information generation unit 101 generates store information such as a store layout, interior design, and product packaging that matches the store image data 100 based on the current store, shelf layout, product information, and the like set in the store information setting unit 11. This makes it easier to generate store information that is close to the store image data.

[0055] Even if system user 5 does not have a specific image of the store they want to create, the system can automatically generate and visualize a detailed store that is closer to their image, supporting them in creating their ideal store and helping to improve store sales and employee engagement.

[0056] The store information generated by the automatic store information generation unit 101 is input to the 3D data conversion unit 14 in the same manner as in the first embodiment, and is used to generate 3D data for a purchasing behavior simulation in a virtual space.

[0057] After the 3D data conversion unit 14 generates 3D data such as stores, shelves, products, and human avatars using the persona information generated by the simulated customer generation unit 10 and the store information generated by the automatic store information generation unit 101, the processing by the human movement calculation unit 15, the VR purchasing behavior estimation unit 16, and the purchasing behavior causal inference unit 17 is the same as in the first embodiment, so a description thereof will be omitted.

[0058] <Detailed Description of the Store Information Automatic Generation Unit 101> The automatic store information generation unit 101 can use an image generation AI such as Stable Diffusion. A system user 5 inputs conditions related to the store image that the automatic store information generation unit 101 wants to realize into the input unit 4 in text form, and the automatic store information generation unit 101 outputs store information such as the store layout and interior design as image data. In addition, the automatic store information generation unit 101 can output store information with modifications to the interior design, advertising, layout, etc. as image data, using the store's shelf layout and product shelving set in the store information setting unit 11 as a framework. Furthermore, the automatic store information generation unit 101 can specify conditions related to the decoration of some shelves in the store and the design of some products, and output the desired store information as image data.

[0059] The automatic store information generation unit 101 performs machine learning based on feedback on whether the generated store information matches the conditions presented by the system user 5, thereby enabling it to output store information that matches a specific situation, environment, or the preferences of a specific system user 5 with greater accuracy.

[0060] Furthermore, by utilizing the difference between the ideal store image envisioned by the system user 5 and the store information generated by the automatic store information generation unit 101, issues in creating an ideal store can be identified, leading to further refinement of the ideal store conditions and improvements to the automatically generated store information. Specifically, after the system user 5 checks the store information generated by the automatic store information generation unit 101 displayed on the display unit 3, the system user 5 modifies part of the automatically generated store information. For example, if the system user 5 feels there is a problem with the layout of a particular shelf in the store, the system user 5 modifies the conditions for the store to be realized so as to change only the shelf layout in the store image data 100, and the automatic store information generation unit 101 generates store information with the improved shelf layout. In this way, the process of inputting the store image data 100, automatically generating store information, finding issues, modifying the store image data 100 again, and automatically generating store information is repeated. This makes it possible to more specifically realize the image of the store to be realized while clearly being aware of the challenges in creating the store, even for issues and problems that were not noticed when the initial store conditions were input.

[0061] [Third embodiment] Next, a third embodiment of the information processing system 1 will be described.

[0062] <Overall explanation of Information Processing System 1> In this embodiment, the information processing system 1 performs a purchasing behavior simulation using a simulated customer in a virtual space, as in the first and second embodiments, while also explaining a method for improving the estimation accuracy of the purchasing behavior simulation. Below, we explain a method for improving the estimation accuracy of the behavioral purchasing simulation based on a comparison of purchasing information and customer sensing data from an actual store with the estimation results thereof.

[0063] <Basic operation> FIG. 6 is a block diagram showing the functional configuration of an information processing system 1 according to the third embodiment. 6 includes an information processing device 2, a display unit 3, and an input unit 4, similar to the information processing system 1 described in Fig. 1. On the other hand, the information processing system 1 shown in Fig. 5 differs in that it also includes a similarity calculation unit 202 that inputs actual purchase data 200 and in-store customer sensing data 201 to the information processing device 2, and a learning unit 203. Therefore, the following explanation will mainly focus on this difference.

[0064] The actual purchase data 200 is, for example, sales data (purchase data) of products in a physical store. Customer sensing data 201 is data on customer behavior in a physical store. For example, the customer sensing data 201 can be generated by analyzing video from a surveillance camera in the physical store and using the results of behavior recognition based on the customer's movements or gaze detection. By using behavior recognition, the customer sensing data 201 can include the path taken in the physical store, the amount of time spent in front of each shelf, and the products picked up. Furthermore, based on gaze detection information, the customer sensing data 201 can also include the products recognized and the gaze duration.

[0065] The similarity calculation unit 202 calculates the similarity between the estimated data based on purchasing behavior and the actual data by comparing the estimated data based on purchasing behavior with actual purchase data 200, which is actual data obtainable in an actual store, and customer sensing data 201 in the actual store. The estimated data based on purchasing behavior includes data such as the actions and movements of the human avatar, and what products the customer paid attention to and purchased.

[0066] The learning unit 203 learns based on the similarity to reduce the difference between the estimated data and the actual data, and reflects the learning result in the VR purchasing behavior estimation unit 16. This makes it possible to improve the accuracy of estimating purchasing behavior.

[0067] The similarity calculation unit 202 calculates the similarity between actual sales data of a customer and sales data (purchase data) estimated for the persona corresponding to that customer. It also calculates the similarity between the customer sensing data 201 and the results of estimated purchasing behavior of a human avatar. In calculating the similarity, the similarity is calculated for each actual customer and simulated customer, and the similarity for each customer segment is statistically calculated for each persona (customer segment) classified by age, gender, etc., and can be used for learning together with the similarity of the sales data. Alternatively, the similarity may be calculated by comparing the average purchasing behavior of multiple customers for each customer segment with the average estimated purchasing behavior of simulated customers. In this way, utilizing actual customer sensing data 201 can further improve the estimation accuracy of purchasing behavior simulations.

[0068] In this way, by using real data such as purchasing data and customer sensing data that can be obtained from physical stores, it is possible to specialize the simulation to the area of ​​the store where the simulation is implemented, respond to fluctuations in product sales due to trends, etc., and adaptively learn the VR purchasing behavior estimation unit 16 to improve the accuracy of the simulation.

[0069] FIG. 7 is a diagram showing the purchasing behavior of a simulated customer created by the person movement calculation unit 15. As described above, the human behavior calculation unit 15 creates the purchasing behavior of a simulated customer, for example, using text data. FIG. 7 shows an example of this content. In this case, a male in his twenties is assumed as the target customer. From this, multiple persona information items shown in the persona list are created, and the purchasing behavior of this simulated customer is shown. This persona information and purchasing behavior are composed of job, worries, situation at the time of visiting the store, purchased product, and keywords. In the case of the simulated customer shown as #1, these are (engineer), (tired from work), (morning after arriving at work...), (nuts), (tired, perfect, coffee,...). The reason for purchasing nuts is also displayed.

[0070] 8(a) to 8(c) are diagrams showing examples in which the VR purchasing behavior estimation unit 16 visualizes the purchasing behavior of a simulated customer. Of these, Figure 8(a) shows an example of displaying the path of a human avatar within a store. Here, the human avatar's walking movement within the store and route information are displayed. In addition, the right column displays a list of recognized products. Here, it is displayed that cookies, chocolate, and potato chips were recognized, and that chocolate was selected. In addition, the bottom column displays the result of the interaction with system user 5. Here, system user 5 inquired, "Why didn't you select cookies?" and the response was, "I wanted something sweet, so I selected chocolate."

[0071] FIG. 8(b) shows a visualization of the interest level of a simulated customer in a product. Here, it is shown that the human avatar showed interest in the product located in the area of ​​interest indicated by the circle.

[0072] FIG. 8(c) shows a visualization of the product purchase results and purchase reasons. Here, the reasons for purchasing the chocolates are shown to be, "It's convenient because it has a zipper," and "I want to share it with my colleagues."

[0073] Comparing FIG. 7 and FIG. 8, by converting text data into 3D data, the purchasing behavior of the simulated customer can be presented to the system user 5 in an easy-to-understand manner.

[0074] According to the information processing device 2 described above in detail, it is possible to provide an information processing device that can simulate the human behavior of various personas when making purchases in a store in the real world using mock customers, and provide information that is useful for improving CX (customer experience). Furthermore, the information processing device 2 is not limited to online use, as in the prior art. Therefore, it is possible to understand personas by taking into account real-world human behavior, such as purchasing behavior in a store. Furthermore, there is no need to collect a large amount of actual customer purchasing data.

[0075] In the above example, the VR purchasing behavior estimation unit 16 visualizes the simulation results using VR, but the visualization method is not particularly limited.

[0076] <Explanation of information processing method> The above-described processing performed by the information processing device 2 is realized by the cooperation of software and hardware resources. That is, a processor such as a CPU provided in the information processing device 2 loads into a main memory and executes a program that realizes each function of the information processing device 2, thereby realizing each function. Therefore, the processing performed by the information processing device 2 described above can be considered to be an information processing method in which, by having a processor execute a program recorded in memory, a fictitious persona information of a customer is generated, store information is set as in-store information, values ​​regarding product purchases of the simulated customer are estimated based on the persona information, the products purchased by the simulated customer and their reasons for purchase are estimated based on the store information, persona information, and values, and the purchasing behavior of the simulated customer calculated based on the products purchased and their reasons for purchase can be visualized. This makes it possible to provide an information processing method that uses simulated customers to simulate the human behavior of various personas when purchasing in a store in the real world and provides information useful for improving CX (customer experience).

[0077] Furthermore, the program running on information processing device 2 can be considered to be a program that causes a computer to realize the following functions: generating fictitious persona information of customers; setting store information as in-store information; estimating values ​​related to product purchases of simulated customers based on persona information; estimating products and purchasing reasons of simulated customers based on store information, persona information, and values; and visualizing the purchasing behavior of simulated customers calculated based on the products and purchasing reasons of the simulated customers. This allows a computer to realize the functions of simulating the human behavior of various personas when purchasing in a store in the real world using simulated customers and providing information useful for improving CX (customer experience).

[0078] The program for realizing this embodiment can be provided not only by communication means but also by being stored on a recording medium such as a CD-ROM.

[0079] Although the present embodiment has been described above, the technical scope of the present invention is not limited to the scope described in the above embodiment. It is clear from the claims that various modifications and improvements to the above embodiment are also included in the technical scope of the present invention. [Explanation of symbols]

[0080] 1...information processing system, 2...information processing device, 3...display unit, 4...input unit, 10...simulated customer generation unit, 11...store information setting unit, 12...value estimation unit, 13...behavior estimation unit, 14...3D data conversion unit, 15...person movement calculation unit, 16...VR purchasing behavior estimation unit, 17...causal inference unit, 100...store image data, 101...automatic store information generation unit, 200...actual purchase data, 201...customer sensing data, 202...similarity calculation unit, 203...learning unit, 1003...arithmetic device

Claims

1. a simulated customer generation unit that generates fictitious persona information of customers; a store information setting unit that sets store information as in-store information; a value estimation unit that estimates values ​​regarding product purchases of the simulated customer based on the persona information; a behavior estimation unit that estimates the purchased items and purchase reasons of the simulated customers based on the store information, the persona information, and the values; a visualization unit that visualizes the purchasing behavior of the simulated customer, the purchasing behavior being calculated based on the purchased products and the purchasing reasons of the simulated customer; An information processing device comprising:

2. a 3D data conversion unit that generates 3D data of the store information and the human avatars of the simulated customers; a human behavior calculation unit that calculates the purchasing behavior of a human avatar of 3D data based on the purchased products and purchasing reasons of the simulated customer estimated by the behavior estimation unit; The information processing device according to claim 1 , further comprising:

3. The information processing device according to claim 2 , wherein the visualization unit visualizes a movement path, a purchasing behavior, and a product recognition process of a human avatar based on the purchasing behavior estimated by the human movement calculation unit.

4. The 3D data conversion unit a person avatar generation unit that generates a person avatar, which is 3D data of the simulated customer generated by the simulated customer generation unit; a store 3D data generation unit that generates 3D data of the store interior, shelf layout, product shelving, and products set by the store information setting unit; The information processing device according to claim 2 , comprising:

5. The information processing device according to claim 1 , further comprising a causal inference unit that infers causality of purchasing behavior of the human avatar.

6. The information processing device according to claim 5 , wherein the causal inference unit allows a user to check the purchasing behavior visualized by the visualization unit, input a question about the purchasing behavior, and answers the input question based on an estimated causal effect.

7. The information processing device according to claim 1 , wherein the visualization unit visualizes touch points between the simulated customer and the store in association with the process from when the simulated customer obtains information about a product or service to when the simulated customer purchases the product or service.

8. The information processing device according to claim 1 , wherein the visualization unit visualizes a difference between the purchasing behavior of customers set as a target and the purchasing behavior of customers other than the target.

9. The information processing device according to claim 1 , further comprising a store information generating unit that generates the store information based on store image data that is information about a store image that a user wants to realize.

10. The information processing device according to claim 9 , wherein the store information generating unit generates store information that matches the store image data based on current store information set in the store information setting unit.

11. a similarity calculation unit that calculates a similarity between the estimated data based on the purchasing behavior and actual data, such as product purchase data in a physical store and data related to customer behavior in the physical store; a learning unit that performs learning based on the similarity so as to reduce a difference between the estimated data and the actual data; The information processing device according to claim 1 , further comprising:

12. The processor executes the program stored in the memory. Generate fictitious customer personas, Set store information as in-store information, Based on the persona information, values ​​regarding product purchases of the simulated customer are estimated, Inferring the products and reasons for purchase of the simulated customers based on the store information, the persona information, and the values; Visualizing the purchasing behavior of the simulated customer calculated based on the purchased products and the purchasing reasons of the simulated customer; Information processing methods.

13. On the computer, The ability to generate fictional customer personas, The ability to set store information as in-store information, A function of estimating values ​​regarding product purchases of simulated customers based on the persona information; a function of estimating the products and reasons for purchase of the simulated customers based on the store information, the persona information, and the values; A function of visualizing the purchasing behavior of the simulated customer calculated based on the purchased product and the purchase reason of the simulated customer; A program to achieve this.

Images