Interactive ordering device, interactive ordering system, interactive ordering method, and program

Abstract

To provide an interactive ordering device, an interactive ordering system, an interactive ordering method, and a program capable of responding to an input from a user with high accuracy.SOLUTION: One aspect of an interactive ordering device of the present disclosure includes a reception unit that receives an input related to an order of a product from a user, a creation unit that creates a prompt according to the received input, a first estimation unit that inputs the created prompt to a generation model and estimates a first order candidate product to be an order candidate, and an output unit that outputs the estimated first order candidate product to the user.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present disclosure relates to an interactive ordering device, an interactive ordering system, an interactive ordering method, and a program. [Background technology]

[0002] In recent years, the introduction of IT (Information Technology) has progressed in the hospitality industry, including the food and beverage industry, and this technology is referred to as FoodTech.

[0003] For example, Patent Document 1 discloses a technology for a non-interactive self-ordering system in which a user selects and orders products from a menu displayed on a touch panel, in which the technology analyzes the content of the user's speech and the consumption status of the ordered products to suggest products to the user. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6535783 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the above-mentioned conventional technology is a non-interactive self-ordering system that merely recognizes the user's passive actions toward the system and suggests a product when the user takes an action that allows for a product suggestion.

[0006] For this reason, the above-mentioned conventional technology cannot be used for ordering systems that are expected to respond to ambiguous input from users in a way that captures their intentions, such as interactive ordering systems that accept orders from users in an interactive format.

[0007] The present disclosure has been made in consideration of the above circumstances, and aims to provide an interactive ordering device, an interactive ordering system, an interactive ordering method, and a program that are capable of responding accurately to input from a user. [Means for solving the problem]

[0008] One aspect of the interactive ordering device disclosed herein includes a reception unit that receives input from a user regarding a product order; a creation unit that creates a prompt in response to the received input; a first estimation unit that inputs the created prompt into a generative model to estimate a first order candidate product that will be an order candidate; and an output unit that outputs the estimated first order candidate product to the user.

[0009] One aspect of the interactive ordering system of the present disclosure includes a server device and an information processing device, and includes a reception unit that receives input from a user regarding a product order, a creation unit that creates a prompt in response to the received input, an estimation unit that inputs the created prompt into a generative model and estimates candidate order products that will be ordered, and an output unit that outputs the estimated candidate order products to the user.

[0010] One aspect of the interactive ordering method of the present disclosure includes a receiving step in which a receiving unit receives input regarding a product order from a user; a creating step in which a creating unit creates a prompt in response to the received input; an estimating step in which an estimating unit inputs the created prompt into a generative model to estimate candidate order products to be ordered; and an output step in which an output unit outputs the estimated candidate order products to the user.

[0011] One aspect of the program of the present disclosure is to cause a computer to function as a reception unit that receives input from a user regarding a product order, a creation unit that creates a prompt in response to the received input, an estimation unit that inputs the created prompt into a generative model and estimates candidate order products that will be ordered, and an output unit that outputs the estimated candidate order products to the user. [Effects of the Invention]

[0012] According to the present disclosure, it is possible to provide an interactive ordering device, an interactive ordering system, an interactive ordering method, and a program that are capable of responding accurately to input from a user. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram showing an example of the configuration of a system according to the first embodiment. [Figure 2] FIG. 2 is a block diagram illustrating an example of a hardware configuration of the information processing apparatus and the server apparatus according to the first embodiment. [Figure 3] FIG. 3 is a block diagram illustrating an example of the functional configuration of the system according to the first embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of user information stored in the user information storage unit of the first embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of product list information stored in the product list information storage unit according to the first embodiment. [Figure 6] FIG. 6 is a diagram showing an example of an order screen displayed on the input display device of the first embodiment. [Figure 7] FIG. 7 is a flowchart showing an example of an interactive self-ordering process performed in the interactive ordering system of the first embodiment. [Figure 8] FIG. 8 is a flowchart showing an example of the first estimation process performed in the interactive ordering system of the first embodiment. [Figure 9] FIG. 9 is a flowchart showing an example of the second estimation process performed in the interactive ordering system of the first embodiment. [Figure 10] FIG. 10 is a block diagram illustrating an example of the functional configuration of the system according to the second embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of a user's order history stored in the order history storage unit of the second embodiment. [Figure 12]FIG. 12 is a diagram showing an example of an order screen displayed on the input display device according to the second embodiment. [Figure 13] FIG. 13 is a flowchart showing an example of an interactive self-ordering process performed in the interactive ordering system of the second embodiment. [Figure 14] FIG. 14 is a flowchart showing an example of the specification process performed in the interactive ordering system of the second embodiment. [Figure 15] FIG. 15 is a block diagram showing an example of the functional configuration of a system according to the fourth modification. DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, an embodiment of the present disclosure (hereinafter simply referred to as "the present embodiment") will be described in detail with reference to the drawings. Note that the present disclosure is not limited to the following embodiment. Furthermore, the following embodiment and modified examples can be combined as appropriate.

[0015] The system of this embodiment will be described below using a self-order system at a restaurant as an example. However, the system of this embodiment is not limited to self-order systems at restaurants, but can also be applied to systems in which users (customers) place orders themselves, such as self-order systems in hospitality businesses other than food and beverage establishments, and food delivery services.

[0016] (First embodiment) First, the configuration of the system according to the first embodiment will be described.

[0017] Fig. 1 is a block diagram showing an example of the configuration of a system 1 according to the first embodiment. As shown in Fig. 1, the system 1 includes an interactive ordering system 10, a generative model 40, and a store terminal 50. The interactive ordering system 10 includes an information processing device 20 and a server device 30. The information processing device 20, the server device 30, the generative model 40, and the store terminal 50 are connected via a network 2. The network 2 can be realized by, for example, at least one of the Internet and a LAN (Local Area Network).

[0018] The interactive ordering system 10 accepts orders from users while communicating with the users in an interactive format such as chat, and is configured to include the information processing device 20 and the server device 30 as described above.

[0019] The information processing device 20 is used when a user places an order interactively. For example, the information processing device 20 may be a portable device or terminal such as a smartphone, tablet terminal, or notebook PC (Personal Computer) owned by the user. Furthermore, for example, the information processing device 20 may be a tablet ordering terminal placed on a table or the like in a restaurant, or a fixed ordering device installed at the entrance or the like of a restaurant.

[0020] If the information processing device 20 is a portable device or terminal owned by the user, an order using the information processing device 20 may be placed through a web page or through an application that allows self-ordering. Examples of applications that allow self-ordering include, but are not limited to, official applications for restaurants and chat applications that also support self-ordering.

[0021] The server device 30 receives an order while interacting with the information processing device 20 (user) in an interactive format, processes the received order, and notifies the store terminal 50 of the processed order. In the first embodiment, the server device 30 uses a generative model 40 to communicate with the information processing device 20 (user) so that it can respond to ambiguous input from the user while understanding their intention. The server device 30 may be a server operated by a restaurant or a server operated by a company that accepts orders on behalf of the user. The server device 30 in the first embodiment is an example of an interactive ordering device.

[0022] The generative model 40 is a natural language processing model known as generative AI (Generative Artificial Intelligence), and provides a service using natural language processing. In the first embodiment, the generative model 40 will be described taking as an example a case where an AI chat service is provided by fine-tuning a large language model (LLM: Large Language Models), but is not limited to this.

[0023] The store terminal 50 (an example of an order receiving terminal) is notified of the order processed by the server device 30. Based on this notification, the restaurant staff starts serving (preparing, etc.) the food, drinks, etc. Note that the store terminal 50 may be a stationary device installed in the kitchen or the like within the restaurant, or may be a portable terminal carried by the restaurant staff.

[0024] FIG. 2 is a block diagram showing an example of the hardware configuration of the information processing device 20 and the server device 30 according to the first embodiment.

[0025] First, a description will be given of the hardware configuration of the server device 30. As shown in Fig. 2, the server device 30 includes a control device 31, a main memory device 32, an auxiliary memory device 33, a communication device 34, and various buses 35. The control device 31, the main memory device 32, the auxiliary memory device 33, and the communication device 34 are connected via the various buses 35. In this way, the server device 30 of the first embodiment has a general hardware configuration using a normal computer.

[0026] The control device 31 controls the overall operation of the server device 30. The control device 31 may be, for example, at least one of a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit), but is not limited to these. There may be any number of CPUs or GPUs as long as they are one or more, and they may be single-core or multi-core.

[0027] Examples of the main memory device 32 include, but are not limited to, a ROM (Read Only Memory) and a RAM (Random Access Memory). The ROM stores various programs, such as a program for controlling the server device 30 and a program for accepting orders interactively. The RAM is used as a working area when the control device 31 performs various controls based on the programs stored in the ROM.

[0028] The auxiliary storage device 33 stores the various programs described above, as well as various data such as user information, product list information, and order history. The various programs described above may be stored in at least one of the main storage device 32 and the auxiliary storage device 33. Examples of the auxiliary storage device 33 include, but are not limited to, existing storage devices capable of magnetic, electrical, or optical storage, such as a hard disk drive (HDD), a solid state drive (SSD), and a digital versatile disc (DVD). The auxiliary storage device 33 may be built into the server device 30 or externally connected to the server device 30 via an interface such as a universal serial bus (USB). The auxiliary storage device 33 may also be a network-attached storage (NAS) connected via a network such as a LAN or a wide area network (WAN).

[0029] The communication device 34 is used to communicate with the information processing device 20, the generative model 40, the store terminal 50, etc. via the network 2. Examples of the communication device 34 include, but are not limited to, a communication device for a wired LAN and a wireless communication device for a wireless LAN.

[0030] In addition to the above configuration, the server device 30 may further include hardwired circuits such as an IC (Integrated Circuit), an ASIC (Application Specific Integrated Circuit), and an FPGA (Field-Programmable Gate Array) that are specific to the server device 30.

[0031] Next, a hardware configuration of the information processing device 20 will be described. As shown in Fig. 2, the information processing device 20 includes a control device 21, a main memory device 22, an auxiliary memory device 23, an input / display device 24, an audio input device 25, an audio output device 26, a sensor group 27, a communication device 28, and various buses 29. The control device 21, the main memory device 22, the auxiliary memory device 23, the input / display device 24, the audio input device 25, the audio output device 26, the sensor group 27, and the communication device 28 are connected via the various buses 29. As such, the information processing device 20 of the first embodiment has a general hardware configuration of electronic devices such as smartphones and tablet terminals.

[0032] The control device 21 controls the overall operation of the information processing device 20. The method of realizing the control device 21 is the same as that of the control device 31. However, since the control device 21 can also perform audio playback and synthesis, in addition to the control device 21, hardwired circuits such as ICs, ASICs, and FPGAs specific to the above processes may also be provided.

[0033] The implementation method of the main memory device 22 is the same as that of the main memory device 32, and therefore detailed description thereof will be omitted. The ROM of the main memory device 22 stores various programs such as a program for controlling the information processing device 20 and a program for placing an order interactively.

[0034] The auxiliary storage device 23 stores the various programs and data described above. The various programs described above may be stored in at least one of the main storage device 22 and the auxiliary storage device 23. The method for realizing the auxiliary storage device 23 is similar to that of the auxiliary storage device 33, and therefore a detailed description thereof will be omitted.

[0035] The input display device 24 displays and updates a chat-style order screen and inputs character strings for ordering in chat style, and serves as a user interface. In the first embodiment, the input display device 24 is described as a touch panel display, but is not limited to this. For example, the input display device 24 may be realized by dividing it into a display device that displays the order screen and an input device that inputs character strings for ordering.

[0036] The voice input device 25 inputs the user's voice and can be realized by, for example, a microphone. The voice output device 26 outputs the voice generated by the control device 21 and can be realized by, for example, a speaker.

[0037] The information processing device 20 may be configured to allow the user to input an order using a spoken voice via the voice input device 25, instead of or in addition to inputting an order character string via the input display device 24. Furthermore, the information processing device 20 may be configured to output the contents of the interactive exchange displayed on the order screen via the voice output device 26, instead of or in addition to displaying and updating the order screen via the input display device 24.

[0038] The sensor group 27 detects various information related to the information processing device 20 and the user who uses the information processing device 20. Examples of the sensor group 27 include, but are not limited to, a GPS (Global Positioning System) sensor that detects the position of the information processing device 20, an acceleration sensor or gyro sensor that detects the tilt, acceleration, etc. of the information processing device 20, an image sensor (digital camera) that captures images and videos around the information processing device 20, and a biosensor that detects biometric information of the user.

[0039] The communication device 28 is used to communicate with the server device 30 via the network 2. The method for realizing the communication device 28 is similar to that of the communication device 34, and therefore a detailed description thereof will be omitted.

[0040] 3 is a block diagram showing an example of the functional configuration of the system 1 of the first embodiment, and mainly shows an example of the functional configuration of the information processing device 20 and the server device 30 included in the interactive ordering system 10 of the first embodiment. As shown in FIG. 3, the information processing device 20 includes an input / output control unit 201, and the server device 30 includes a reception unit 301, a recognition unit 303, a user information storage unit 305, a response unit 307, a product list information storage unit 309, an output unit 317, and an order history storage unit 319. The response unit 307 includes a creation unit 311, a first estimation unit 313, and a second estimation unit 315.

[0041] The input / output control unit 201 can be realized by, for example, the control device 21, main memory device 22, and communication device 28 described in FIG. 1. For example, the control device 21 reads out a program for interactive ordering stored in the main memory device 22 (ROM) or the auxiliary memory device 23 and loads it into the main memory device 22 (RAM). The control device 21 realizes the above-mentioned input / output control unit 201 by executing various processes in accordance with the loaded program. Here, the case where the input / output control unit 201 is realized as software has been described as an example, but the input / output control unit 201 may also be realized as hardware. In this case, the input / output control unit 201 may be realized by, for example, the above-mentioned hardwired circuit. The above-mentioned input / output control unit 201 may also be realized by a combination of software and hardware.

[0042] The reception unit 301, the recognition unit 303, the response unit 307, the creation unit 311, the first estimation unit 313, the second estimation unit 315, and the output unit 317 can be realized, for example, by the control device 31, the main memory device 32, and the communication device 34 described in FIG. 2 . For example, the control device 31 reads a program for receiving orders in an interactive format stored in the main memory device 32 (ROM) or the auxiliary memory device 33 and loads it into the main memory device 32 (RAM). The control device 31 executes various processes in accordance with the loaded program to realize each of the above-mentioned functional units. Here, the description has been given using an example in which each of the above-mentioned functional units is realized as software, but at least some of the above-mentioned functional units may also be realized as hardware. In this case, the functional units realized as hardware may be realized, for example, by the above-mentioned hardwired circuit. Furthermore, any of the above-mentioned functional units may be realized by a combination of software and hardware.

[0043] The user information storage unit 305, the product list information storage unit 309, and the order history storage unit 319 can be realized by, for example, the auxiliary storage device 33 described with reference to FIG.

[0044] In the first embodiment, it is not necessary for the server device 30 to have all of the above-described functional units as essential components, and at least some of the functional units can be omitted. For example, it is also possible to omit at least one of the recognition unit 303, the user information storage unit 305, the second estimation unit 315, and the order history storage unit 319.

[0045] The input / output control unit 201 controls input and output for interactive communication such as chat with the server device 30. For example, the input / output control unit 201 displays a chat-style order screen on the input / display device 24, accepts a character string input by a user from the input / display device 24, displays the accepted character string on the order screen to update the order screen, and notifies the server device 30 of the accepted character string. Also, for example, when the input / output control unit 201 is notified of a character string such as a response from the server device 30, the input / output control unit 201 accepts the notified character string and displays the accepted character string on the order screen on the input / display device 24 to update the order screen.

[0046] When receiving an input using a user's utterance from the voice input device 25, the input / output control unit 201 performs voice recognition on the received utterance and converts it into a character string, thereby treating the input voice as a character string. The input / output control unit 201 may also generate a voice of the character string notified by the server device 30 by voice synthesis and output the voice from the voice output device 26.

[0047] The receiving unit 301 receives various inputs from the user. Specifically, the receiving unit 301 receives inputs related to a user authentication request and a product order from the user (information processing device 20).

[0048] For example, before opening the order screen on the input / output control unit 201, the input / output control unit 201 receives input of authentication information for user authentication from the user and transmits the received authentication information to the server device 30. As a result, the receiving unit 301 receives the authentication information as a request for user authentication.

[0049] The authentication information may be any information that can be used to authenticate a user. For example, if the recognition unit 303 described below recognizes a user through password authentication, the authentication information may be account information that is a set of ID (identification) and password. For example, if the recognition unit 303 described below recognizes a user through facial authentication, the authentication information may be a facial image of the user captured by the sensor group 27 (image sensor). For example, if the recognition unit 303 described below recognizes a user through code authentication, the authentication information may be information read from a QR (Quick Response) Code (registered trademark) by the sensor group 27 (image sensor). In addition to these, biometric information such as the user's fingerprint information may also be used as authentication information.

[0050] Furthermore, for example, the input / output control unit 201 accepts an input of a character string related to an order from a user on an order screen displayed on the input display device 24, and transmits the accepted character string to the server device 30. In this way, the accepting unit 301 accepts an input related to a product order.

[0051] The recognition unit 303 recognizes the user who orders the product. Specifically, the recognition unit 303 checks the authentication information received by the reception unit 301 against the user information (master information for authentication) stored in the user information storage unit 305, thereby recognizing whether the user who orders the product is a user registered in the interactive ordering system 10.

[0052] FIG. 4 is a diagram showing an example of user information stored in the user information storage unit 305 of the first embodiment. In the example shown in FIG. 4, the user information includes, in addition to an ID and a password corresponding to authentication information, age, gender, and address, but is not limited to these. In particular, the part corresponding to authentication information may include information according to the authentication method adopted by the recognition unit 303. The user information stored in the user information storage unit 305 is registered by the user via the information processing device 20, for example, when the user uses the interactive ordering system 10 for the first time.

[0053] When the reception unit 301 receives the authentication information, the recognition unit 303 compares the authentication information with each piece of user information stored in the user information storage unit 305. If the recognition unit 303 compares user information containing information that matches the authentication information (an ID and a password in the example shown in FIG. 4), the recognition unit 303 recognizes that the user ordering the product is the user of the compared user information. On the other hand, if the recognition unit 303 does not compare user information containing information that matches the authentication information, the recognition unit 303 recognizes that the user ordering the product is not a user registered in the interactive ordering system 10. Note that if the recognition unit 303 recognizes that the user ordering the product is not a user registered in the interactive ordering system 10, subsequent processing may be stopped, or the processing may be continued by treating the user as a one-time user (guest user).

[0054] When the receiving unit 301 receives an input related to a product order from a user, the response unit 307 analyzes whether the input is an ambiguous order. An ambiguous order is, for example, an order in which the product to be ordered cannot be uniquely identified. For example, the response unit 307 performs natural language processing on the input (character string) received by the receiving unit 301 and interprets the meaning of the character string, thereby analyzing whether the order uniquely identifies a product included in the product list information stored in the product list information storage unit 309.

[0055] FIG. 5 is a diagram showing an example of product list information stored in the product list information storage unit 309 of the first embodiment. In the example shown in FIG. 5, each product included in the product list information is information including, but not limited to, an ID, a product name, a main genre, and a sub-genre. The main genre corresponds to a major classification of the product, and classifies the product as either food or drink. The sub-genre corresponds to a minor classification of the product, and if the product is food, it classifies it into a dish group such as a main dish or a side dish, and if the product is a drink, it classifies it into a type of drink.

[0056] If the analysis result shows that the order does not uniquely specify the product name of a product included in the product list information stored in the product list information storage unit 309, the response unit 307 determines that the order is ambiguous. For example, if the input regarding a product order from a user received by the reception unit 301 is an abstract order that does not specifically specify a product, such as "I want something that arrives quickly," the response unit 307 analyzes the order as ambiguous. On the other hand, if the order uniquely specifies the product name of a product included in the product list information, the response unit 307 determines that the order is clear (unambiguous).

[0057] In the first embodiment, the server device 30 uses the generative model 40 to respond to the user so that the server device 30 can respond to the user's intention even when the order is ambiguous. Hereinafter, a method for responding to the user using the generative model 40 will be specifically described with reference to the creation unit 311 and the first estimating unit 313.

[0058] The creation unit 311 creates a prompt in response to an input related to a product order from a user that is accepted by the acceptance unit 301. Specifically, when the response unit 307 analyzes the order as being ambiguous, the creation unit 311 creates a prompt in response to the input that is accepted by the acceptance unit 301. A prompt is an instruction or question that is input to the generative model 40 (AI chat service).

[0059] Any method for creating a prompt may be used, and the input received by the receiving unit 301 may be used as the prompt as is. In this case, for example, if the input received by the receiving unit 301 is "I want something that arrives quickly," the input "I want something that arrives quickly" may be used as the prompt. However, generally, to obtain an appropriate response or result from an AI chat, a clear and specific prompt is required. For this reason, the creating unit 311 may create a prompt by adding a precondition, such as that the input is a product order from a user at a restaurant, as a standard phrase in addition to the input received by the receiving unit 301. In this case, the precondition portion of the prompt may be, for example, "We have received the following order from a customer as a restaurant order. Please tell us the restaurant product that the following order is intended for." and the question portion of the prompt may be, "I want something that arrives quickly."

[0060] The creation unit 311 may also create a prompt using RAG (Retrieval Augmented Generation). RAG is a technique for causing the generative model 40 to respond based on the contents of a pre-specified document or database. For example, the creation unit 311 may use RAG to create a prompt that causes the generative model 40 to respond to an input received by the reception unit 301 based on the product list information stored in the product list information storage unit 309. In this case, the creation unit 311 creates a prompt including a context obtained by searching product list information that includes a list of products. Specifically, the creation unit 311 performs a full-text search of the product list information stored in the product list information storage unit 309, generates the full text of the product list information as a context, and creates a prompt that instructs the generative model 40 to respond to the input received by the reception unit 301 based on the context. In this case, the prompt can be created by setting the prerequisite part of the prompt to, for example, "We have received the following order from a customer for a restaurant. Considering the content of the context below, please tell us the restaurant product that the following order is intended for.", the context part of the prompt to be the generated context (the full text of the product list information), and the question part of the prompt to be, "I would like something that arrives quickly." Note that the context may be a vectorized version of the full text of the product list information. Furthermore, if the product genre, etc. is clear from the order input accepted by the accepting unit 301, the creating unit 311 may generate a context by searching only that genre rather than searching the full text of the product list information.

[0061] The first estimation unit 313 inputs the prompt created by the creation unit 311 into the generative model 40 and estimates a first order candidate product that will be an order candidate. Specifically, the first estimation unit 313 acquires an answer (answer sentence) to the input prompt from the generative model 40, and estimates a product extracted from the acquired answer as the first order candidate product. Note that if the answer includes multiple products, the first estimation unit 313 may extract all of these products and estimate them as the first order candidate product. For example, suppose that in response to a prompt such as "I want something that arrives quickly," the generative model 40 acquires an answer such as "Would you like edamame, cucumber, and cold tofu?" In this case, the first estimation unit 313 extracts "edamame," "cucumber," and "hiyayakko" from the answer and estimates them as the first order candidate product.

[0062] The first estimation unit 313 may also be configured to confirm whether the estimated first order candidate products are products that can be accepted as orders. For example, the first estimation unit 313 may be configured to confirm whether the estimated first order candidate products are included in the product list information stored in the product list information storage unit 309. For example, the first estimation unit 313 references the product list information shown in FIG. 5 and confirms that the estimated first order candidate products "edamame," "cucumber," and "hiyayakko" are included. In this way, even if the generative model 40 responds with a product that is not included in the product list information due to an incomplete prompt or the like, it is possible to prevent such a product from being suggested to the user.

[0063] Furthermore, for example, out-of-stock information indicating whether an item is out of stock may be added to the product list information stored in the product list information storage unit 309, and the first estimation unit 313 may check whether the estimated first order candidate item is out of stock. In this way, it is possible to prevent an out-of-stock item from being suggested to the user.

[0064] If the first order candidate product estimated by the first estimation unit 313 is not a product that can be accepted as an order, the second estimation unit 315 estimates, from product list information including a list of products that can be accepted as orders, a product whose similarity with the input accepted by the acceptance unit 301 exceeds a threshold as a third order candidate product that will be an order candidate. In this way, even if a response to the user using the generative model 40 does not go well, a minimum response to the user can be made.

[0065] Specifically, the second estimation unit 315 breaks down the input (character string) received by the receiving unit 301 into words, performs vector conversion, and leaves words such as adjectives and nouns that are likely to be related to the order. The second estimation unit 315 calculates the similarity between each genre of the product list information stored in the product list information storage unit 309 and the left-over words, and estimates products in a genre where the calculated similarity exceeds a threshold as third order candidate products. Note that the break down into words can be performed using a known method such as morphological analysis, the vector conversion can be performed using a known method such as Word2Vec, and the similarity can be calculated using a known method such as cosine similarity.

[0066] For example, in the case of an input of "I want something that will arrive soon," the words "soon," "comes," and "thing" remain, and the "speedy menu" sub-genre of the product list information shown in Fig. 5 has a high similarity to these words (exceeds the threshold). Therefore, the second estimation unit 315 estimates that "edamame," "cucumber," and "hiyayakko" are the third candidate order products.

[0067] The output unit 317 outputs the first order candidate product estimated by the first estimation unit 313 to the user (information processing device 20). For example, if the first order candidate product estimated by the first estimation unit 313 is a product that can be accepted as an order, the output unit 317 outputs the first order candidate product to the user (information processing device 20). For example, if the first order candidate product estimated by the first estimation unit 313 is not a product that can be accepted as an order, the output unit 317 outputs a third order candidate product to the user (information processing device 20). Note that if the input related to a product order from the user accepted by the accepting unit 301 is analyzed by the responding unit 307 as a clear order, the output unit 317 outputs a product that is uniquely identified from the product list information to the user (information processing device 20).

[0068] When the input / output control unit 201 is notified of order candidate products as a response from the server device 30, it accepts the notified order candidate products, displays the accepted order candidate products on the order screen on the input / display device 24, and updates the order screen.

[0069] 6 is a diagram showing an example of an order screen 401 displayed on the input display device 24 of the first embodiment. The order screen 401 shown in FIG. 6 displays a chat-style exchange between the user and the server device 30, in which a message 411 from the server device 30 saying "Please place your order" is displayed, and a message 413 (order) from the user saying "I want something that will arrive soon" is displayed. In response to this, the server device 30 uses the generative model 40 to display a message 415 saying "How about these?", as well as options 417 for "edamame," 418 for "cucumber," and 419 for "hiyayakko," which are first order candidate products.

[0070] As described above, according to the first embodiment, the server device 30 uses the generative model 40 to respond to the user, and can therefore respond in a way that captures the user's intention in ordering, even if the order is ambiguous.

[0071] After this, when the user selects an item to be ordered from options 417 to 419 on the order screen 401, the selected item is entered into the input field 405, and when the user presses the send button 407, the input / output control unit 201 sends the entered item to be ordered to the server device 30.

[0072] The receiving unit 301 further receives an input from the user (information processing device 20) specifying an order target product to be ordered from among the order candidate products output to the user.

[0073] The response unit 307 processes the ordered products received by the reception unit 301 as an order, and stores the ordered products processed as an order in the order history storage unit 319 as an order history, in association with the user recognized by the recognition unit 303.

[0074] The output unit 317 outputs the ordered items received by the reception unit 301 to the store terminal 50. As a result, the restaurant staff starts providing (cooking, etc.) the ordered items.

[0075] Next, the operation of the system of the first embodiment will be described.

[0076] FIG. 7 is a flowchart showing an example of an interactive self-ordering process performed in the interactive ordering system 10 of the first embodiment.

[0077] First, when the authentication information is accepted by the accepting unit 301, the recognition unit 303 compares the authentication information with the user information (master information for authentication) stored in the user information storage unit 305, thereby recognizing that the user ordering the product is a user registered in the interactive ordering system 10 (step S101).

[0078] Next, when the input / output control unit 201 recognizes that the user ordering the product is a user registered in the interactive ordering system 10 by the server device 30, it displays a chat-style ordering screen on the input display device 24 along with a message from the server device 30 such as "Please place your order" (step S103).

[0079] Next, the input / output control unit 201 accepts the order string input by the user from the input display device 24, displays the accepted string on the order screen to update the order screen, and notifies the server device 30 of the accepted string (step S105).

[0080] Next, when the receiving unit 301 receives an input of a character string related to a product order from the user, the response unit 307 analyzes whether the input order is ambiguous or not (step S107).

[0081] If the response unit 307 determines that the order is ambiguous (Yes in step S107), the creation unit 311 and the first estimation unit 313 perform a first estimation process to estimate first order candidate products (step S109). Details of the first estimation process will be described later with reference to FIG. 8. On the other hand, if the response unit 307 determines that the order is not ambiguous (No in step S107), the process proceeds to step S117.

[0082] Next, the first estimation unit 313 checks whether the products that can be ordered can be estimated as first order candidate products (step S111). If the products that can be ordered can be estimated as first order candidate products (Yes in step S111), the process proceeds to step S117. On the other hand, if the products that can be ordered cannot be estimated as first order candidate products (No in step S111), the second estimation unit 315 performs a second estimation process to estimate third order candidate products that will become order candidates (step S113). Details of the second estimation process will be described later using FIG. 9.

[0083] Subsequently, if the second estimation unit 315 is also unable to estimate an orderable product as the third order candidate product (No in step S115), it returns to step S105 as it has failed to estimate the order candidate product, and starts over from order acceptance.On the other hand, if the second estimation unit 315 is able to estimate an orderable product as the third order candidate product (Yes in step S115), it proceeds to step S117.

[0084] In step S117, the output unit 317 outputs the estimated candidate order items to the user (information processing device 20), and the input / output control unit 201 accepts the candidate order items output (notified) from the server device 30, displays the accepted candidate order items on the order screen on the input display device 24, and updates the order screen.

[0085] Next, if the user selects an order item from the order candidate items (options) on the order screen 401 (Yes in step S119), the input / output control unit 201 transmits the selected order item to the server device 30, thereby sending the order (step S121). On the other hand, if the user does not select an order item and re-enters the order (No in step S119), the process returns to step S105 and starts over from accepting the order.

[0086] Next, the response unit 307 processes the ordered items received by the reception unit 301 as an order, and stores the ordered items processed as an order in the order history memory unit 319 as an order history in association with the user recognized by the recognition unit 303 (step S123).

[0087] Subsequently, if the user inputs an additional order on the order screen 401 (Yes in step S125), the process returns to step S105, and the reception unit 301 receives the additional order from the input / output control unit 201. If an additional order is not placed (No in step S125), the process ends.

[0088] FIG. 8 is a flowchart showing an example of the first estimation process performed in the interactive ordering system 10 of the first embodiment.

[0089] First, the creation unit 311 creates a prompt in response to an input related to a product order from a user accepted by the acceptance unit 301 (step S201). For example, the creation unit 311 searches for product list information stored in the product list information storage unit 309 to generate a context, and creates a prompt that instructs the user to respond to the input accepted by the acceptance unit 301 based on the generated context.

[0090] Next, the first estimation unit 313 inputs the prompt created by the creation unit 311 into the generative model 40 (step S203), and obtains an answer (answer sentence) to the prompt as an output result of the generative model 40 (step S205).

[0091] Next, the first estimation unit 313 estimates the product extracted from the acquired response as the first order candidate product (step S207).

[0092] FIG. 9 is a flowchart showing an example of the second estimation process performed in the interactive ordering system 10 of the first embodiment.

[0093] First, the second estimation unit 315 breaks down the input (character string) received by the receiving unit 301 into words using a technique such as morphological analysis, and performs vector conversion using a known technique such as Word2Vec (step S301).

[0094] Next, the second estimation unit 315 extracts words such as adjectives and nouns that are likely to be related to orders from the vector-converted words (step S303).

[0095] Next, the second estimation unit 315 calculates the similarity between each genre of the product list information stored in the product list information storage unit 309 and the extracted words, and estimates products in a genre where the calculated similarity exceeds a threshold as third order candidate products (step S305).

[0096] As described above, according to the first embodiment, the server device 30 uses the generative model 40 to respond to the user, and can respond in a way that captures the user's intention in the order, even for ambiguous orders. This enables highly accurate responses to user input, which is expected to improve user satisfaction.

[0097] (Second embodiment) In the second embodiment, an example will be described in which a response to a user is made using the user's order history in addition to the generative model 40. Note that in the second embodiment, differences from the first embodiment will be mainly described, and descriptions of similarities to the first embodiment will be omitted.

[0098] 10 is a block diagram showing an example of the functional configuration of a system 1001 according to the second embodiment, and mainly shows an example of the functional configuration of an information processing device 20 and a server device 1030 included in an interactive ordering system 1010 according to the second embodiment. As shown in FIG. 10, the server device 1030 differs from the first embodiment in that a response unit 1307 includes an identification unit 1316, an output unit 1317, and an order history storage unit 1319.

[0099] When the receiving unit 301 receives an input regarding a product order from a user, the response unit 1307 analyzes whether the input is an ambiguous order, and if the input is an ambiguous order, further analyzes whether the input is an order that can ambiguously identify the product name of a product included in the product list information stored in the product list information storage unit 309. For example, if the input regarding a product order from a user received by the receiving unit 301 is an order that specifies a product genre, such as "I would like a highball," the response unit 1307 analyzes the input as an order that can ambiguously identify the product. Also, for example, as described in the first embodiment, if the input regarding a product order from a user received by the receiving unit 301 is an abstract order that does not specifically specify a product, such as "I would like something that arrives quickly," the response unit 1307 analyzes the input as not an order that can ambiguously identify the product.

[0100] In the second embodiment, when the order is one in which the product can be ambiguously identified, the identification unit 1316 responds to the user by using the user's order history stored in the order history storage unit 1319. Specifically, the identification unit 1316 refers to the order history of the user recognized by the recognition unit 303, and identifies second order candidate products that will be order candidates from the input accepted by the acceptance unit 301. The identification unit 1316 also refers to the order history of the recognized user, and further identifies the priority of the second order candidate products.

[0101] 11 is a diagram showing an example of a user's order history stored in the order history storage unit 1319 of the second embodiment. In the example shown in FIG. 11, the order history includes, but is not limited to, an ID, order date and time, and ordered items. The ID indicates the ID of the user who placed the order, and corresponds to the ID included in the user information shown in FIG. 4.

[0102] For example, suppose the user ID recognized by the recognition unit 303 is "1," and the input related to the user's product order received by the reception unit 301 is "I'd like a highball." The identification unit 1316 decomposes the input (character string) received by the reception unit 301 into words, performs vector conversion, and leaves words such as adjectives and nouns that are likely to be related to the order, using a method similar to that used by the second estimation unit 315 described in the first embodiment. In the case of "I'd like a highball," the word "highball" remains. Furthermore, the identification unit 1316 references the product list information shown in FIG. 5 to determine whether the ordered products with the ID "1" in the order history shown in FIG. 11 include products whose genre is "highball" or whose product name includes "highball." In the example shown in FIG. 11, "AAA highball" appears twice and "BBB highball" appears once. Therefore, the identification unit 1316 identifies the "AAA Highball" and the "BBB Highball" as second order candidate products, and identifies the "AAA Highball," which has been ordered more frequently, as having a higher priority than the "BBB Highball." Note that the identification unit 1316 may also identify "edamame," which is ordered together with the "AAA Highball," a product in the "highball" category, as a second order candidate product.

[0103] On the other hand, if the order is not ambiguously identifiable for the products, similar to the first embodiment, the creation unit 311 and the first estimation unit 313 respond to the user using the generative model 40. That is, if the identification unit 1316 cannot identify the second order candidate product from the input accepted by the acceptance unit 301 (if the order is not ambiguously identifiable for the products), the first estimation unit 313 estimates the first order candidate product using the generative model 40.

[0104] The output unit 1317 outputs the second order candidate products identified by the identification unit 1316 to the user (information processing device 20). For example, the output unit 1317 outputs the second order candidate products identified by the identification unit 1316 to the user (information processing device 20) according to the priority identified by the identification unit 1316. Note that if the second order candidate products cannot be identified from the input accepted by the acceptance unit 301 (if the order is not ambiguously identifiable), the output unit 1317 outputs the first order candidate products to the user (information processing device 20), and if the second order candidate products can be identified, the output unit 1317 outputs the second order candidate products to the user (information processing device 20).

[0105] 12 is a diagram showing an example of an order screen 401 displayed on the input display device 24 of the second embodiment. The order screen 401 shown in FIG. 12 displays a chat-style exchange between the user and the server device 30, in which a message 511 from the server device 30 saying "Please place your order" is displayed, and a message 513 (order) from the user saying "I'd like a highball" is displayed. In response to this, the server device 30 responds using the user's order history with a message 515 saying "An AAA highball, right?", and displays an option 517 for "AAA highball," which is the second order candidate product with the highest priority, an option 518 for "BBB highball," which is the second order candidate product with the next highest priority, an option 519 for "edamame," and an option 520 requesting an order other than options 517 to 519.

[0106] FIG. 13 is a flowchart showing an example of an interactive self-ordering process performed in the interactive ordering system 1010 of the second embodiment.

[0107] The processing from steps S401 to S407 in the flowchart shown in FIG. 13 is the same as the processing from steps S101 to S107 in the flowchart shown in FIG.

[0108] In step S408, if the response unit 1307 determines that the order is ambiguous (Yes in step S407), it analyzes whether the order is an order that can ambiguously identify order candidates. If the order is not an order that can ambiguously identify order candidates (No in step S408), the process proceeds to step S409, and if the order is an order that can ambiguously identify order candidates (Yes in step S408), the process proceeds to step S410.

[0109] The process in step S409 is similar to the process in step S109 in the flowchart shown in FIG.

[0110] In step S410, the identifying unit 1316 performs an identifying process to identify second order candidate products that will become order candidates, and the process proceeds to step S411. Details of the identifying process will be described later with reference to FIG.

[0111] The subsequent processing from steps S411 to S425 is the same as the processing from steps S111 to S125 in the flowchart shown in FIG.

[0112] FIG. 14 is a flowchart showing an example of the specification process performed in the interactive ordering system 1010 of the second embodiment.

[0113] First, the identifying unit 1316 breaks down the input (character string) received by the receiving unit 301 into words using a technique such as morphological analysis, and performs vector conversion using a known technique such as Word2Vec (step S501).

[0114] Next, the identifying unit 1316 extracts words such as adjectives and nouns that are likely to be related to orders from the vector-converted words (step S503).

[0115] Next, the identification unit 1316 refers to the product list information shown in FIG. 5, extracts products that have the extracted word as a genre or part of the product name from the user's ordered products in the order history shown in FIG. 11, and identifies them as second order candidate products (step S505).

[0116] Next, the specifying unit 1316 sets priorities for the specified second order candidate products according to the number of times the user has ordered them (step S507).

[0117] As described above, in the second embodiment, the server device 30 can respond to the user by using both the generative model 40 and the user's order history. Therefore, according to the second embodiment, a response that takes into account the user's preferences can be made to an ambiguous order, and a highly accurate response to user input is possible, which is expected to further improve user satisfaction.

[0118] (Variation 1) In the above second embodiment, an example was described in which a response is made using the user's order history when the order is ambiguous and the product can be ambiguously identified, but the opportunities for responding using the order history are not limited to this.

[0119] For example, similar to the first embodiment, in the case of an ambiguous order, the first estimation unit 313 may respond to the user using the generative model 40, and if the first order candidate product estimated by the first estimation unit 313 is not an acceptable product for order, the identification unit 1316 may perform the identification process using the order history described above. That is, the identification unit 1316 may identify a second order candidate product when the first estimation unit 313 is unable to estimate an acceptable product for order as the first order candidate product. In this case, the output unit 1317 outputs the first order candidate product to the user (information processing device 20) if the first order candidate product estimated by the first estimation unit 313 is an acceptable product for order, and outputs the second order candidate product to the user (information processing device 20) if the first order candidate product is not an acceptable product for order.

[0120] If the specifying unit 1316 is unable to specify the second order candidate product, the second estimating unit 315 may further estimate a third order candidate product.

[0121] Furthermore, in the case of an ambiguous order, the user may refer to the order history and place the order voluntarily, rather than using the user's order history. For example, an order using the user's order history may be provided to the user as a favorites function, and when the user uses the favorites function, a second selection of order candidate products may be displayed on the order screen, allowing the user to order the selected products. In this case, products that are ordered less frequently may be treated as products that the user dislikes, and may not be displayed, or may be displayed as products with a low priority.

[0122] (Variation 2) In the above second embodiment, an example has been described in which the identification unit 1316 identifies the second order candidate product using the recognized user's own order history, but the identification unit 1316 may also identify the second order candidate product using the order history of other users.

[0123] For example, the identification unit 1316 may refer to the user information shown in FIG. 4, extract users with attributes that are the same as or similar to those of the recognized user, and identify ordered items included in the order history of the extracted users as second order candidate items.

[0124] For example, in the second embodiment, the case where the ID of the user recognized by the recognition unit 303 is "1" is taken as an example. However, since the user with ID "1" is a male (see FIG. 4), the order history of the user with ID "3", who is also a male, may be referenced (see FIG. 11) to identify "fried chicken", which was ordered together with the "AAA highball", as the second order candidate item.

[0125] (Variation 3) In the above first embodiment, an example was described in which the second estimation unit 315 estimates the third order candidate product by using the similarity with each genre of the product list information, but this is not limited to this, and the estimation may also be performed by using the similarity with the product name.

[0126] For example, in the case of an input such as "I would like a highball," the second estimation unit 315 breaks down the input (character string) into words, performs vector conversion, and leaves "highball" as a word such as an adjective or noun that is likely to be related to the order. The second estimation unit 315 may calculate the similarity between the product names in the product list information stored in the product list information storage unit 309 and the leftover words, and may infer that "AAA highball," "BBB highball," "CCC highball," and "DDD highball" are third candidate order products for which the calculated similarity exceeds a threshold.

[0127] (Variation 4) In the above-described embodiments and modifications, the server device is an example of an interactive ordering device, but an information processing device can also be an example of an interactive ordering device.

[0128] 15 is a block diagram showing an example of the functional configuration of a system 2001 of Modification 4, and mainly shows an example of the functional configuration of an information processing device 2020 and a server device 2030 included in an interactive ordering system 2010 of Modification 4. As shown in FIG. 15, the information processing device 2020 may include a reception unit 2201, a recognition unit 2203, a user information storage unit 2205, a response unit 2207, a product list information storage unit 2209, an output unit 2217, an order history storage unit 2219, a creation unit 2211, a first estimation unit 2213, and a second estimation unit 2215, which correspond to the respective functional units described in the server device 30 of the first embodiment.

[0129] In this case, the processing of each functional unit described in the server device 30 of the first embodiment is performed by the information processing device 2020, and the main control unit 2301 of the server device 2030 performs processing to output the ordered items notified by the information processing device 2020 to the store terminal 50.

[0130] The configuration of the interactive ordering system 2010 in the fourth modification can be adopted, for example, when the functions of the interactive ordering device are realized as an application.

[0131] In addition, the functional units that realize the interactive ordering device described in each of the above embodiments and variants may be distributed and arranged in a server device and an information processing device, and the interactive ordering device may be realized as an interactive ordering system.

[0132] Furthermore, the generative model 40 described in each of the above embodiments and modifications may be included in a server device or an information processing device. In this case, the generative model 40 may be realized as, for example, an AI chat service in which small language models (SLMs) are fine-tuned.

[0133] (program) The programs executed by the interactive ordering devices of the above embodiments and variants are provided as installable or executable files stored on computer-readable storage media such as CD-ROMs, CD-Rs, memory cards, DVDs, and flexible disks (FDs).

[0134] Furthermore, the programs executed by the interactive ordering devices of the above embodiments and modifications may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Furthermore, the programs executed by the interactive ordering devices of the above embodiments and modifications may be provided or distributed via a network such as the Internet. Furthermore, the programs executed by the interactive ordering devices of the above embodiments and modifications may be provided by being pre-installed in ROM or the like.

[0135] The programs executed by the interactive ordering devices of the above-mentioned embodiments and modifications have a modular structure for implementing the above-mentioned units on a computer. In terms of actual hardware, for example, the CPU reads the learning program from the HDD onto the RAM and executes it, thereby implementing the above-mentioned units on a computer.

[0136] As described above, according to the above-described embodiments and modifications, highly accurate responses to inputs from the user are possible.

[0137] The above-described embodiments and modifications merely illustrate examples of specific embodiments of the present disclosure, and the technical scope of the present disclosure should not be construed as being limited by these. Therefore, the present disclosure can be implemented in various forms without departing from its gist or main features. For example, the above-described embodiments and modifications may be appropriately combined in their respective constituent units. Furthermore, for example, some components may be deleted from all components in the above-described embodiments and modifications.

[0138] The present disclosure includes the following aspects.

[0139] (1) a reception unit that receives input regarding product orders from users; a generator for generating a prompt in response to the received input; a first estimation unit that inputs the generated prompt into a generative model and estimates a first order candidate product that will be an order candidate; an output unit that outputs the estimated first order candidate products to the user; An interactive ordering device comprising:

[0140] (2) the prompt includes a context in which the creation unit retrieved product list information including a list of products; The interactive ordering device according to (1) above.

[0141] (3) the first estimation unit confirms whether the estimated first order candidate product is a product that can be accepted as an order; The output unit outputs the estimated first order candidate product to the user if the product is an orderable product. An interactive ordering device according to (1) or (2) above.

[0142] (4) a recognition unit that recognizes the user; and a specification unit that specifies second order candidate products that are order candidates from the received input by referring to the order history of the recognized user, the output unit outputs the identified second order candidate product to the user. The interactive ordering device according to (1) above.

[0143] (5) The identification unit further identifies a priority of the second order candidate product by referring to the order history of the recognized user; the output unit outputs the identified second order candidate products to the user in accordance with the identified priorities. The interactive ordering device according to (4) above.

[0144] (6) When the second order candidate product cannot be identified from the received input, the output unit outputs the first order candidate product to the user, and when the second order candidate product can be identified, the output unit outputs the second order candidate product to the user. The interactive ordering device according to (4) above.

[0145] (7) The first estimation unit estimates the first order candidate product when the identification unit cannot identify the second order candidate product from the received input. The interactive ordering device according to (6) above.

[0146] (8) The output unit outputs the first order candidate product to the user when the estimated first order candidate product is a product that can be accepted as an order, and outputs the second order candidate product to the user when the estimated first order candidate product is not a product that can be accepted as an order. The interactive ordering device according to (4) above.

[0147] (9) The identification unit identifies the second order candidate product when the first estimation unit is unable to estimate a product that can be accepted as an order as the first order candidate product. The interactive ordering device according to (8) above.

[0148] (10) A second estimation unit is further provided for estimating, when the estimated first order candidate product is not a product that can be accepted as an order, a product that has a similarity to the received input that exceeds a threshold value from product list information including a list of products that can be accepted as an order, as a third order candidate product; the output unit outputs the estimated third order candidate product to the user when the estimated first order candidate product is not an acceptable product for order. The interactive ordering device according to (1) above.

[0149] (11) The receiving unit further receives an input specifying an order target product to be ordered from among the order candidate products output to the user, the output unit outputs the accepted ordered product to an order receiving terminal. The interactive ordering device according to any one of (1), (2), and (4) to (10) above.

[0150] (12) An interactive ordering system comprising a server device and an information processing device, a reception unit that receives input regarding product orders from users; a generator for generating a prompt in response to the received input; an estimation unit that inputs the generated prompt into a generative model and estimates order candidate products; an output unit that outputs the estimated order candidate products to the user; An interactive ordering system comprising:

[0151] (13) a receiving step in which the receiving unit receives an input regarding an order for a product from a user; a creating step in which a creating unit creates a prompt in response to the received input; an estimation step in which an estimation unit inputs the generated prompt into a generative model to estimate order candidate products; an output step in which an output unit outputs the estimated order candidate products to the user; Interactive ordering methods including.

[0152] (14) a reception unit that receives input regarding product orders from users; a generator for generating a prompt in response to the received input; an estimation unit that inputs the generated prompt into a generative model and estimates order candidate products; an output unit that outputs the estimated order candidate products to the user; A program that allows a computer to function. [Explanation of symbols]

[0153] 1, 1001, 2001 systems 2 Network 10, 1010, 2010 Interactive ordering system 20, 2020 Information processing equipment 30, 1030, 2030 Server equipment 40 Generative Models 50 store terminals 201 Input / Output Control Unit 301, 2201 Reception 303, 2203 recognition section 305, 2205 User information storage unit 307, 1307, 2207 Response Section 309, 2209 Product list information storage section 311, 2211 Creation Department 313, 2213 1st estimation part 315, 2215 2nd estimation part 317, 1317, 2217 output section 319, 1319, 2219 Order history memory section 1316 Specific part

Claims

1. a reception unit that receives input regarding product orders from users; a generator for generating a prompt in response to the received input; a first estimation unit that inputs the generated prompt into a generative model and estimates first order candidate products that will be order candidates; an output unit that outputs the estimated first order candidate products to the user; An interactive ordering device comprising:

2. The prompt includes a context in which the creation unit retrieved product list information including a list of products.

10. The interactive ordering device of claim 1.

3. the first estimation unit confirms whether the estimated first order candidate product is a product that can be accepted as an order; the output unit outputs the estimated first order candidate product to the user if the product is an orderable product.

3. An interactive ordering device according to claim 1 or 2.

4. a recognition unit that recognizes the user; and a specifying unit that specifies second order candidate products as order candidates from the received input by referring to the order history of the recognized user, the output unit outputs the identified second order candidate product to the user.

10. The interactive ordering device of claim 1.

5. the specifying unit further specifies a priority of the second order candidate product by referring to the order history of the recognized user; the output unit outputs the identified second order candidate products to the user in accordance with the identified priorities.

5. The interactive ordering device of claim 4.

6. the output unit outputs the first order candidate product to the user when the second order candidate product cannot be identified from the received input, and outputs the second order candidate product to the user when the second order candidate product can be identified from the received input.

5. The interactive ordering device of claim 4.

7. the first estimation unit estimates the first order candidate product when the identification unit cannot identify the second order candidate product from the received input; 7. The interactive ordering device of claim 6.

8. the output unit outputs the first order candidate product to the user when the estimated first order candidate product is a product that can be accepted as an order, and outputs the second order candidate product to the user when the estimated first order candidate product is not a product that can be accepted as an order.

5. The interactive ordering device of claim 4.

9. the specifying unit specifies the second order candidate product when the first estimation unit is unable to estimate a product that can be accepted as an order as the first order candidate product.

9. The interactive ordering device of claim 8.

10. a second estimation unit that, when the estimated first order candidate product is not a product that can be accepted as an order, estimates a product that has a similarity to the received input that exceeds a threshold value from product list information including a list of products that can be accepted as an order, as a third order candidate product; the output unit outputs the estimated third order candidate product to the user when the estimated first order candidate product is not an acceptable product for order.

10. The interactive ordering device of claim 1.

11. the receiving unit further receives an input specifying an order target product to be ordered from among the order candidate products output to the user; the output unit outputs the accepted ordered product to an order receiving terminal. The interactive ordering device according to any one of claims 1, 2, and 4 to 10.

12. An interactive ordering system comprising a server device and an information processing device, a reception unit that receives input regarding product orders from users; a generator for generating a prompt in response to the received input; an estimation unit that inputs the generated prompt into a generative model and estimates order candidate products; an output unit that outputs the estimated order candidate products to the user; An interactive ordering system comprising:

13. a receiving step in which a receiving unit receives an input regarding an order for a product from a user; a creating step in which a creating unit creates a prompt in response to the received input; an estimation step in which an estimation unit inputs the generated prompt into a generative model to estimate order candidate products; an output step in which an output unit outputs the estimated order candidate products to the user; Interactive ordering methods including.

14. a reception unit that receives input regarding product orders from users; a generator for generating a prompt in response to the received input; an estimation unit that inputs the generated prompt into a generative model and estimates order candidate products; an output unit that outputs the estimated order candidate products to the user; A program that allows a computer to function.

Images