Information processing apparatus, control method thereof, storage medium, and computer program product

Abstract

This disclosure relates to an information processing apparatus and its control method, a storage medium, and a computer program product. The information processing apparatus includes: one or more storage devices storing an instruction set; and one or more processors executing the instruction set to: provide an interaction using generative AI to a customer of a service provider; input at least one of the following information: an interaction history with the customer, customer information, and product information, into the generative AI, and cause the generative AI to determine whether the current time is a suitable time to try selling the product; and, if the generative AI determines that the current time is a suitable time to try selling the product, respond to the interaction and offer the product for sale.

Description

Technical Field

[0001] This disclosure relates to an information processing apparatus for printing documents using generative artificial intelligence (AI), its control method, storage medium, and computer program product. Background Technology

[0002] For products with chat functionality, new products and consumables can be sold through this feature. For example, Japanese Patent Application Publication No. 2023-99334 discloses a method in which customers entering a chat room are used as triggers to send promotional information to the chat room.

[0003] However, with this past technology, information for attempting a sale is obtained through chat using generative artificial intelligence (AI) and the like, but entering the chat room is the only trigger used in terms of the timing of the sales activity. In other words, a sales attempt is made regardless of whether the customer has a purchase intention. However, more effective sales activities are possible if the appropriate timing of such sales can be determined based on the content of interactions throughout the response flow during the chat with the customer. Summary of the Invention

[0004] This disclosure enables novel techniques for appropriately determining the timing of a sales attempt based on the content of the interaction.

[0005] One aspect of this disclosure provides an information processing apparatus comprising: one or more storage devices storing an instruction set; and one or more processors executing the instruction set to: provide an interaction using generative AI to a customer of a service device; input at least one of the following information: an interaction history with the customer, customer information, and product information to the generative AI, and cause the generative AI to determine whether the current time is a suitable time to try selling the product; and, if the generative AI determines that the current time is a suitable time to try selling the product, responding to the interaction and proposing the sale of the product.

[0006] Another aspect of this disclosure provides a method for controlling an information processing device, the method comprising: providing an interaction using generative AI to a customer of a service device; inputting at least one of the following information—interaction history of the interaction with the customer, customer information, and product information—to the generative AI, and causing the generative AI to determine whether the current time is a suitable time to try selling the product; and, if the generative AI determines that the current time is a suitable time to try selling the product, responding to the interaction and proposing the sale of the product.

[0007] Another aspect of this disclosure provides a non-transitory computer-readable storage medium storing a computer program for causing a computer to perform steps of a method for controlling an information processing apparatus, the method comprising: providing an interaction using generative AI to a customer of a service apparatus; inputting at least one of the following information—an interaction history with the customer, customer information, and product information—to the generative AI, and causing the generative AI to determine whether the current time is a suitable time to try selling the product; and, if the generative AI determines that the current time is a suitable time to try selling the product, responding to the interaction and offering the product for sale.

[0008] Another aspect of this disclosure provides a computer program product storing a computer program for causing a computer to perform steps of a method for controlling an information processing device, the method comprising: providing an interaction using generative AI to a customer of a service device; inputting at least one of the following information—interaction history of the interaction with the customer, customer information, and product information—to the generative AI, and causing the generative AI to determine whether the current time is a suitable time to try selling the product; and, if the generative AI determines that the current time is a suitable time to try selling the product, responding to the interaction and proposing the sale of the product.

[0009] The features of this disclosure will become apparent from the following description of embodiments with reference to the accompanying drawings. The following description of embodiments is given by way of example. Attached Figure Description

[0010] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the embodiments.

[0011] Figure 1 This is a block diagram illustrating an example of the network configuration of a printing system according to one embodiment.

[0012] Figure 2 This is a block diagram illustrating an example of the hardware construction of a computer and various servers according to one embodiment.

[0013] Figure 3 This is a block diagram illustrating an example of the hardware construction of a printer according to one embodiment.

[0014] Figure 4 This illustrates a UI that indicates a series of interactions according to one embodiment.

[0015] Figures 5A to 5B This is a sequence diagram illustrating the interactions between elements according to one embodiment.

[0016] Figure 6 This is a diagram illustrating an example of a prompt according to one embodiment.

[0017] Figure 7 This illustrates a UI that indicates a series of interactions according to one embodiment.

[0018] Figure 8 This illustrates a UI that indicates a series of interactions according to one embodiment.

[0019] Figure 9 This illustrates a UI that indicates a series of interactions according to one embodiment. Detailed Implementation

[0020] In the following, embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments are not intended to limit the scope of the claims. Several features are described in the embodiments, but not all of these features are necessary, and multiple features can be suitably combined. Furthermore, in the drawings, the same reference numerals are given the same or similar constructions, and redundant descriptions are omitted.

[0021] First Embodiment System Construction The first embodiment of this disclosure will now be described. First, the construction of the printing system according to this embodiment will be described. The printing system according to this embodiment is a printing system targeting an image forming apparatus. In this printing system, a print job is sent from an externally connected PC to the image forming apparatus. When a print job is generated, its print settings are edited on the PC screen as needed.

[0022] Reference Figure 1 This describes an example of the network construction for implementing the printing system of the chat system according to this embodiment. For example... Figure 1 As shown, the printing system 100 includes: a computer 110, which is a terminal device; a printer (printing device) 200, which is configured to print; a control execution server 310; and a network 120, which communicatively connects these components to each other. For example, a wired connection or a wireless communication method such as Wi-Fi can be used for network connection.

[0023] Computer 110 is an example of an information processing device, and is, for example, a device such as a personal computer, a smartphone, a tablet terminal, etc. Printer 200 is an example of an image processing device, an image forming device, etc. Printer 200 is, for example, a single-function printer (SFP) such as an inkjet printer or a copier, or a multi-function peripheral device (MFP) having functions such as printer function, copier function, etc.

[0024] The control mechanism execution server 310 is an example of an information processing device, and is, for example, a server composed of a computer such as a personal computer. The control mechanism 311 of the control mechanism execution server 310 uses generative AI 313, etc., to perform at least some of the feature controls of this disclosure. Various types of data, such as user information, sales data, etc., are stored in various databases (DBs) 312. These various types of data include, for example, user personal information, the specifications and status of printer 200, purchase history, and inquiry history.

[0025] Note that, as indicated by 101, in this system, printer 200 can be connected to computer 110 using a Universal Serial Bus (USB) connection or the like. This configuration also allows various databases 331 to be stored and used in a database server 330 separate from the control mechanism execution server 320. This configuration also allows generative AIs 341 to be stored and used in a generative AI execution server 340 separate from the control mechanism execution server 320. In this case, the control mechanism execution server 320 acts as the server that primarily executes the control mechanism 321. Although the following description will describe the form of the control mechanism execution server at control mechanism execution server 310, as indicated by 100, the control mechanism execution server can also be implemented as the control mechanism execution server 320 in the form indicated by 101. The operation of controls 311 and 321 is not limited to being implemented by control mechanism execution servers 310 and 320, and can be distributed to computer 110 or the like and executed as programs created in control mechanism execution servers 310 and 320.

[0026] Hardware Construction Reference Figure 2 An example of the hardware construction of the computer 110 and the server 300, which serves as a control mechanism execution server 310, etc., according to this embodiment is described.

[0027] Computer 110 includes a display unit 111, an operation unit 112, a storage unit 113, a control unit 114, and a network communication unit 115, and these components are communicatively connected to each other. For example, desktop or laptop personal computers, tablet terminals, smartphones, etc., can be used as computer 110.

[0028] The control unit 114 comprises a central processing unit (CPU) 116 and a memory 117, and controls the computer 110 as a whole. The display unit 111 comprises, for example, a display such as an LCD panel, and can display images, etc. The operation unit 112 comprises, for example, a mouse, a keyboard, etc., and can accept input operations performed by the user 10. Note that the operation method is not limited to this, and the configuration may allow, for example, the computer 110 to have a touch panel, and the touch panel serves as both the display unit 111 and the operation unit 112.

[0029] Storage unit 113 is constructed of storage media such as hard disk drives (HDDs) and solid-state drives (SSDs), and stores various types of information, such as various types of programs (software) required for the operation of computer 110. These programs are loaded into memory 117 and executed by CPU 116 as needed. The programs include units, controls, etc., for causing CPU 116 (computer) to execute the printing system 100. Note that the location of the stored programs is not limited to storage unit 113 of computer 110. For example, the programs may be stored in printer 200 or control execution server 310, or they may be distributed among computer 110, printer 200, and control execution server 310. Network communication unit 115 inputs and outputs data to external devices via an external network.

[0030] For server 300, an example of the hardware configuration of server 300 including control mechanism execution servers 310 and 320, DB server 330, generative AI execution server 340, etc. is shown. Server 300 includes display unit 301, operation unit 302, storage unit 303, control unit 304, and network communication unit 305, and these components are communicatively connected to each other.

[0031] The control unit 304 comprises a CPU 306 and a memory 307, and controls the server 300 as a whole. Preferably, the control unit includes a graphics processing unit (GPU) 308. The display unit 301 is, for example, a display such as a liquid crystal panel, and can display images, etc. The operation unit 302 is, for example, a mouse, a keyboard, etc., and can accept input operations performed by the administrator. Note that the operation method is not limited to this, and the configuration may, for example, include a touch panel on the server 300, and the touch panel serves as both the display unit 301 and the operation unit 302. Under the assumption that the server 300 will be operated by connecting to the server 300 from an external machine, the server 300 may be configured to lack at least one of the display unit 301, operation unit 302, etc.

[0032] Storage unit 303 is constructed of storage media such as HDD or SSD, and stores various types of information, such as various programs (software) required for the operation of server 300. These programs are loaded into memory 307 and executed by CPU 306 as needed. Note that the location of the stored programs is not limited to storage unit 303 of server 300. In this way, the hardware configuration of server 300 is the same as that of computer 110.

[0033] Construction of printing device Reference Figure 3 An example describing the hardware construction of a printer according to this embodiment. For example... Figure 3 As shown, printer 200 includes a display unit 201, an operation unit 202, a storage unit 203, a control unit 204, a network communication unit 205, and a printing unit 206, and these components are communicatively connected to each other. Printer 200 can be any device with printing capabilities, such as an SFP (Surface Mount Technology) inkjet printer, a multifunction peripheral device (MFP), etc.

[0034] The control unit 204 consists of a CPU 207 and a memory 208, and controls the printer 200 as a whole. The display unit 201 consists of a monitor such as an LCD panel, and can display images, etc. The operation unit 202 consists of a touch panel, various types of buttons, etc., and can accept input operations performed by the user 10. Note that this configuration allows the computer 110 to have a touch panel, and the touch panel serves as both the display unit 111 and the operation unit 112.

[0035] Storage unit 203 is constructed of storage media such as HDD or SSD, and stores various types of information, such as various programs (software) required for the operation of printer 200. Programs are loaded into memory 208 and executed by CPU 207 as needed. Network communication unit 205 inputs and outputs data to external devices via an external network. Printing unit 206 prints the digital data stored in storage unit 203, memory 208, etc., as images onto a printing medium (sheet) according to instructions from control unit 204. The printing method can be inkjet, toner-based, or another method.

[0036] Chat screen Reference Figure 4 Example of a screen depicting a chat system applying the techniques disclosed herein. Figure 4An example of a user interface (UI) indicating the status of a chat completion is shown. UI 400 is implemented by CPU 116 in computer 110. In this embodiment, such a UI is displayed in display unit 111 of computer 110. UI 400 is an example of a UI displayed when the printer driver of printer 200 displays the print settings screen. However, UI 400 is not limited to this and can be a UI that is displayed and executed in collaboration with a service or app such as Copilot (registered trademark). For example, this configuration can also allow the UI to be displayed in display unit 201 of printer 200. Furthermore, this configuration can allow the UI to be displayed on a touch panel integrating a display unit and an operation unit in computer 110, printer 200, etc. In this process, the interaction in chat window 402 is performed in stages. Additionally, the interaction is not limited to the interaction in chat window 402, and various interactions are performed.

[0037] For example, the printer driver (UI 400) displays print settings 401 and a chat window 402. The content and format displayed are not limited to these, and various display formats can be used. Settings used when printing with printer 200 can be made in print settings 401. For example, paper type, paper size, color, and color correction can be set. These print settings can be set by manipulating objects in print settings 401, and also by interacting with objects in chat window 402 (described later). The details of the settings are merely examples and are not intended to limit this disclosure. Furthermore, print settings 401 are displayed so that the remaining ink levels for each color can be easily identified.

[0038] Chat window 402 includes an input field 403 at its bottom for accepting input from user 10. While an example of text input is described here, voice input can be used alternatively, and in this case, the user's utterances can be converted into text via speech recognition, which can then be input into the generative AI. Input field 403 is the area where user 10 can input natural language (prompts) from the operating unit of computer 110 using, for example, a keyboard. User 10 confirms the content entered in input field 403 by pressing send button 404, and computer 110 sends this content to control execution server 310. In other words, the prompt is the text (natural language) input into the generative AI. The sent content and its response are displayed in chat window 402, as indicated by texts 410 to 415. Figure 4 In the text, texts 410, 412, and 414 indicate natural language text prompts entered and sent by user 10, and texts 411, 413, and 415 indicate response text generated by the system.

[0039] Here, in chat window 402, user 10 has instructed to print on A5 paper using text 410. In response, using text 411, the system has inquired about other conditions required to perform the printing, such as whether the printing is color or black and white, and whether it is single-sided or double-sided. Furthermore, using text 412, user 10 has responded and agreed to the color and single-sided printing proposed by text 411. Subsequently, using text 413, the system has noticed that printing is in progress and is conducting a sales attempt to prompt the user to order the nearly depleted magenta ink. Here, the system has determined that the current time is suitable for attempting a sales attempt based on the interaction history, printing history, etc., and is thus attempting a sales attempt. The determination of the timing of the sales attempt will be described in more detail later. Furthermore, using text 414, user 10 has agreed to the sales in text 413 and has instructed to place an order. Therefore, using text 415, the system places an order and provides notification of the estimated arrival time of the product. In this way, according to this embodiment, the timing of the sales attempt is determined based on the interaction content in the chat window (also called the "chat room") where the printing settings are made, the details of the printing, etc., and the attempt is made. The process will be described in detail later.

[0040] sequence Reference Figures 5A to 5B Description Instructions Figure 4 The sequence of interactions between components. However, Figures 5A to 5B The focus is only on actions that are significantly related to the technology disclosed herein, and not all interactions are included. Figures 5A to 5B As shown in the image. Figure 6 This is a diagram illustrating an example of a prompt according to this embodiment.

[0041] User 10 opens UI 400, for example, for printing purposes. Note that UI 400 can be opened in any way, for example, by pressing a given button. UI 400 is displayed on display unit 111 via CPU 116 controlling display unit 111. As indicated by text 410, in the input field 403 of chat window 402, CPU 116 of computer 110 accepts input from user 10 in natural language (prompts) indicating what user 10 wishes to print. This corresponds to... Figure 5A In step S500, the CPU 116 receives a prompt through the operation unit 112.

[0042] After receiving input from user 10, the CPU 116 of computer 110 sends a prompt to the control mechanism 311 of the control mechanism execution server 310 in step S501. Upon receiving the print instruction, the control mechanism 311 requests subject information from printer 200 in step S502. For example, subject information is information needed to respond to the prompt (such as whether printer 200 is currently in a printable state). In step S503, the printer 200, having been requested for subject information, sends the requested subject information to the control mechanism execution server 310. However, this step is not absolutely necessary and can be omitted. This information can be obtained directly from printer 200 or indirectly. For example, "indirectly obtaining" refers to obtaining information stored in database 312 for so-called cloud services. While this embodiment does not limit the information obtained from printer 200 via control mechanism 311, it is not necessary to obtain all data, and it is conceivable to collect only the necessary data based on the details input by user 10.

[0043] In step S504, the control mechanism 311 uses the generative AI 313 to generate a response to step S501. Here, the generative AI 313 generates a prompt response to confirm the print settings with the user 10. Specifically, after interpreting the prompt received in step S501, the generative AI 313 invokes a print function. This function serves as a printer driver that uses a so-called "function calling" generated by executing this function. Therefore, this function is not absolutely necessary in the technology according to this disclosure.

[0044] like Figure 4 As shown, the user enters the text "Print on A5", as indicated by text 410. The entered text 410 is sent as a prompt to generative AI 313; generative AI 313 interprets text 410 and outputs print settings based on the text. Figure 4 In the example, the output paper size is set to A5, the color mode is set to full color, and single-sided printing is set. Here, for settings other than "A5" specified by the user, default values ​​pre-set in the printer driver can be used, or default values ​​pre-set in the printer designated as the destination for sending print data can be used.

[0045] In step S505, the control mechanism 311 uses the information about the printer 200 obtained in step S503 to update various databases 312, and requests and obtains user information from the updated databases 312. Here, user information includes product usage history such as inquiry history and printing history, purchase history such as purchase history, and sales text for current promotions to the user. In this embodiment, the sales text is generated by a system that dynamically generates the sales text based on the printing speed and the target ink level for consumables that may run out, obtained as various databases are updated. More specifically, in this system, a preset message stating "At the current printing speed, ink (Y) will run out in X weeks. Do you want to place an order?" is provided, and the values ​​of X and Y are dynamically adjusted according to the user's specific situation by updating various databases 312. However, the sales text generation method is not limited to updating in response to consumables used in this way. For example, the user's purchase history, session history, etc., can be provided to the generative AI 313 along with product information, and sales text can be generated if a suitable product is selected. Alternatively, you can send the sales text for the new product to all users.

[0046] In step S506, the control mechanism executes the generative AI 313 of server 310 to determine whether the sales text should be inserted at that time. Here, such as... Figure 6The indicated prompt has been input into generative AI 313 and has received an "inappropriate" output. Therefore, it is determined not to attempt a sale, and the response generated in step S504 is provided as is. In this way, according to this embodiment, a prompt including at least one of customer information, conversation history (interaction history), and product information is input into generative AI 313, and it is determined whether the current time is an appropriate time to attempt a sale. In addition to past interactions with the customer, the interaction history preferably includes the immediately preceding interaction (utterance). This allows it to determine the conversation flow to date and the state of the immediately preceding conversation. Although an example of using all these information items to determine the timing of an attempt to sell has been described herein, the technology is not intended to be limited thereto, and at least one of these information items can be used for determination. For example, if the determination is based solely on customer information, it can be made using at least one of the customer's personal information, purchase history, repair history, information about the product currently in use, and response history included in the customer information. The timing for attempting a sale can also be determined by inputting only product information into the generative AI. For example, situations can be anticipated such as: the time to purchase a replacement is approaching, the warranty period has expired, or sales of new features, optional units, etc., are about to begin. In such cases, product model, version update information, etc., can be input into the generative AI for determination. This information is stored in various DBs 312 in association with the user, product, etc. Although the basis for determining the generative AI 313 is not included in the output in this embodiment, if Figure 6 The prompt shown also includes a prompt indicating the reason for the output, such as: "Considering the flow of the session, the sales attempt is determined to be inappropriate at the current time." In this way, by using methods such as... Figure 6 Using the prompts shown as input, the generative AI 313 can output whether the sales attempt at that time was “appropriate” or “inappropriate”. However, the input prompts, output format, etc. are not limited to this, and for example, the input prompts can be changed according to the purpose, and text such as “appropriate” or “inappropriate” can be converted into numerical values ​​such as 0 or 1, making the output easier to process.

[0047] This will refer to Figure 6Example of a prompt used to determine the timing of a sales attempt. First, text 600 clarifies the details of the determination. At this point, it will be indicated that several conditions exist. Details about the sales attempt are specified using text 601. Although a sales text that is actually assumed is described here, it can be redesigned later, and thus the information collected in step S505 can be described instead of using complete sentences. Text 602 analyzes the emotions of user 10 and makes the specification taking these emotions into account. Although for simplicity, this is an analysis of emotions immediately preceding the previous one, it can also be specified in other ways, such as by considering the entire conversation history of the day. Text 603 is used to specify the determination taking into account the flow of the conversation. Because generative AI is a system capable of achieving natural conversations, the specification is made by utilizing this characteristic. Here, texts 600 and 601 are examples of “determined details,” and texts 602 and 603 are examples of “determined conditions.”

[0048] Using text 604, previous conversation history is provided as reference information. This allows the generative AI 313 to increase the analysis for user 10 and the material for future determination regarding whether to attempt a sale, and thus make a more appropriate determination. Here, although the history of the conversation between user 10 and the generative AI 313 is provided as reference information, the reference information is not limited to this, and other examples of sales successes, failures, etc., with the user can also be provided as reference information. The results of sentiment analysis can be provided together with the conversation history in the reference information. At this time, the reference information can also be extracted only for the product to be sold. In addition, in addition to the system's conversation history, procedure manuals, determination criteria, etc., written by sales personnel, etc., can also be provided as reference information. In this embodiment, for example, if attempting to sell a product that is not new but that user 10 may not be aware of, product information can be used as reference information, and whether user 10 might like the product can be considered as a perspective for determining whether to attempt a sale.

[0049] The above prompts are merely examples, and the text used to assess the timing of a sales attempt is not limited to these. For example, conditions such as, "You are a talented salesperson. Pay close attention to the customer's emotions and decide whether to attempt a sale while considering the conversation flow," can be added. This can be used as a prompt input or as a setting for system prompts, etc. Furthermore, it is not necessary to complete all the prompts shown here in a single input, and prompt input and processing can be performed multiple times. Additionally, a mechanism for determining the validity of the determination can be added by inputting the determined result back into generative AI 313.

[0050] The description will now return to Figures 5A to 5BIn step S507, the control mechanism 311 of the server 310 stores the prompt received from user 10 in step S501 and the response generated in step S504 as a response history (interaction history) in various databases 312. In step S508, the control mechanism 311 sends the prompt response generated in step S504 to computer 110. In step S509, computer 110 displays the received prompt. This is text 411.

[0051] In step S510, the CPU 116 of the computer 110 confirms the details of the text 411 from the user 10 and accepts, for example, a print instruction indicated by text 412. In step S511, the CPU 116 sends the prompt entered in step S510 to the control mechanism execution server 310. In step S512, the control mechanism 311 sends an instruction to the printer 200 to execute printing. Such an instruction can be sent from the control mechanism execution server 310 to the printer 200 via a network, or via the computer 110 to the printer 200 connected to the computer 110 via USB, etc.

[0052] In step S513, the control unit 311 sends the prompt received in step S511 to the generative AI 313. After receiving the prompt, the generative AI 313 generates a current response based on the prompt received in step S501 and its response to the prompt. In step S514, the control unit 311 requests and obtains user information from various databases 312 in the same manner as in step S505. Here, user information includes product usage history (such as inquiry history, printing history, etc.), purchase behavior history (such as purchase history), and sales text for the current promotion to the user. It also includes the response history from S501 to S508 stored in step S507. At this time, in addition to requesting user information from various databases 312, processing for storing the print instructions from step S512 in various databases 312 can also be performed.

[0053] In step S515, the control mechanism executes the generative AI 313 of server 310 to determine the timing of a sales attempt in the same manner as in step S506. Except for adding information in step S514, the determination process uses the same procedure and will not be described here. Here, generative AI 313 determines that the current time is suitable for attempting a sale. In other words, for Figure 6The prompt in step S506, in response to the input and response details immediately preceding the prompt in texts 602 and 603, results in generative AI 313 outputting "Appropriate" in step S515 in a form corresponding to step S515. This is considered because generative AI 313 has determined that the session has ended and user 10 appears to be in a calm state, and has concluded that a sales attempt can be made.

[0054] In step S516, in response to the determination of the attempted sale in step S515, control unit 311 instructs control unit 310 to execute generative AI 313 of server 310 to regenerate the prompt response. Specifically, generative AI 313 performs the following processing: the pure response generated in step S513 in response to the prompt received from user 10 in step S511 is naturally connected with the attempted sale determined in step S515 as natural language. In this embodiment, generative AI 313 generates natural language text indicating that printing will be performed and that, at the current usage frequency of printer 200, a certain amount of ink will be used up within two weeks, and prompts the user to purchase more ink. Here, the text "Printing will now take place. At the current printing speed, ink (M) will be used up within two weeks. Do you want to place an order?" is generated.

[0055] In step S517, the control unit 311 stores the instruction to attempt a sale made in step S516, the prompt from user 10, and its response, etc., in various databases 312. In step S518, the control unit 311 sends the prompt response generated in step S516 to the computer 110. In step S519, the CPU 116 of the computer 110 displays the received prompt response (text 413). In step S520, the CPU 116 receives from user 10 an indication that they have agreed to the details of the sale proposed in step S519 and have requested an order. For example, this is a prompt such as that indicated by text 414.

[0056] In step S521, CPU 116 sends the prompt input in step S520 to control mechanism 311. In step S522, control mechanism 311 sends the prompt to generative AI 313 and issues instructions to generate a response to the prompt. In step S523, control mechanism 311 performs processing for ordering ink. Although in this embodiment control mechanism 311 performs order processing by rewriting data from various DBs 312, the processing is not limited to this and can be tailored to a separate ordering system.

[0057] In step S524, the control mechanism 311 stores the interactions to date in various databases 312. In this embodiment, ink is sold to serve as a basis for determining the timing of future sales attempts, and the fact of successful sales is also stored. This increases the amount of material to be considered when determining the timing in the future, which makes it possible to improve the accuracy of the determination. However, to reduce the load on the system, some or all of the information may not be stored.

[0058] In step S525, the control mechanism 311, in response to the prompt made in step S521, sends the response generated in step S522 to the computer 110. In step S526, the CPU 116 displays the response to the prompt received from the control mechanism execution server 310 in step S525. This is text 415.

[0059] As described above, the information processing apparatus (control mechanism execution server) according to this embodiment provides customers of the service provider with an interaction using generative AI. The information processing apparatus inputs at least one of the following information: the interaction history with the customer, customer information, and product information, into the generative AI, and enables the generative AI to determine whether the current time is an appropriate time to attempt a product sale. Furthermore, when the generative AI determines that the current time is an appropriate time to attempt a product sale, the information processing apparatus responds to the interaction and proposes a product sale. In this way, according to this embodiment, a novel mechanism can be provided for appropriately determining the timing of an attempt to sell within the flow of a chat response.

[0060] Second Embodiment The second embodiment of this disclosure will be described below. This embodiment will use a construction equivalent to that described in the first embodiment to describe an embodiment with different display and input methods in the chat window 402. Since the difference from the first embodiment lies in the display and input methods, the discussion of determining the sales timing will be omitted.

[0061] Chat window Reference Figure 7 The description indicates a chat window in the interactive UI according to this embodiment. This embodiment uses different display and input methods to perform the same processing as described in the first embodiment. In other words, chat window 700 is equivalent to chat window 402 in terms of interactive content.

[0062] Input field 701 corresponds to input field 403, and send button 703 corresponds to send button 404. Similarly, text 710 corresponds to text 410, text 711 corresponds to text 411, text 712 corresponds to text 412, text 713 corresponds to text 413, text 714 corresponds to text 414, and text 715 corresponds to text 415.

[0063] The difference between chat window 700 and chat window 402 lies in the display methods of voice input 702, text 713, and text 714. As indicated by voice input 702, a voice input button can be displayed in the input bar 701, etc., and voice input can be accepted. Furthermore, as indicated by text 713, in addition to text, button 704 can be displayed as an object to be operated on by user 10, facilitating user understanding during promotions, etc. This also makes it easier to determine the action desired by user 10 ("order" or "do not order"). In other words, by confirming button 704, user 10 can easily understand that they are being asked to determine whether to order ink (M) and can easily respond to this.

[0064] As indicated by text 714, the result of operating button 704 can be displayed as text. Specifically, user 10 presses the "Order" button among buttons 704 displayed in text 713. To indicate that the button has been pressed, the text 714 indicates that the "Order" button is being pressed. The display format is not limited to this, and other display and notification methods can be designed and displayed. Alternatively, when displaying the operation result by changing the display color of button 704 itself, the operation result can be displayed as the text indicated by text 714.

[0065] As described above, the information processing apparatus according to this embodiment presents a product sales attempt along with operation objects ("Order" button and "Don't Order" button) that accept the customer's response to the sales attempt. In this way, this disclosure is not limited to the construction of the first embodiment described above, and can be applied to various other display formats and input methods. The construction of this embodiment can also be implemented in combination with the third and fourth embodiments (described below).

[0066] Third Embodiment The third embodiment of this disclosure will be described below. This embodiment will use a construction equivalent to that described in the first embodiment to describe an embodiment where the date of the inquiry and the date of the attempted sale are different dates. In this embodiment, the printer 200 is disconnected from the network and only chats corresponding to the inquiry are conducted.

[0067] Chat window Reference Figure 8The UI describes a chat window that indicates the interaction according to this embodiment. Chat window 800 and chat window 820 correspond to chat window 402. Chat window 820 is an interaction from a date after the date indicated by chat window 800. Input fields 801 and 821 correspond to input field 403. Send button 802 and send button 822 correspond to send button 404. The method for displaying the chat window is not limited to the method according to the first embodiment described above, and for example, the UI associated with the driver and the UI for querying the chat can be displayed independently. Chat window 800 can be displayed using computer 110, and chat window 820 can be displayed using a different computer.

[0068] Chat window 800 is an example of an interaction on a day when printer 200 is not in good working order. As indicated by text 810, computer 110 may assume an inquiry will be made and display a message to user 10. However, this is not a necessary configuration and does not need to be displayed as in the first embodiment described above. As indicated by text 811, user 10 inputs an instruction that printer 200 is not printing correctly. It is estimated that user 10 feels negative or angry due to the fact that printing did not go as expected. As indicated by text 811, such emotions may appear in the input from user 10. In this embodiment, the user's emotions are detected in such an interaction and then used as material for determining the timing of an attempt to sell.

[0069] Subsequent texts 812 to 814 are responses to text 811 and subsequent interactions. Because this embodiment focuses on an example of a system aiming to achieve results over multiple days, the input (such as that indicated by text 811) is not intended to limit this disclosure, but other inputs may be used alternatively. Therefore, the interaction from text 812 to text 814 is not limited to this embodiment, and other interactions are possible.

[0070] As indicated by text 812, the system responds to input from user 10 by presenting a proposed solution. Such an inquiry response can be achieved by creating a generative AI 313 specifically for such inquiries, or by using mechanisms such as Retrieval Enhanced Generation (RAG). The system determines that attempting a sale at this time is inappropriate and presents only the proposed solution to the inquiry in text 811. More specifically, when text 810 is published, in step S501, the prompt “The ink is not flowing correctly!!” in text 811 is sent from computer 110 to the control execution server 310. In step S504, generative AI 313 generates an answer based on the prompt. Generative AI 313 uses the response and sales text received in step S505 to determine whether the timing is suitable for attempting a sale. In other words, in Figure 6In the prompt, generative AI 313 determines when to attempt a sale when the prompt indicating that the input sales text has been updated is displayed, using the immediate preceding prompt as text 811 and the response details as text 812. Here, as a result, generative AI 313 outputs "Inappropriate". This is a result of determining that the conversation flow is a complaint and a response to the complaint, and that attempting a sale during this process is inappropriate.

[0071] As indicated by text 813, user 10 implements the proposed solution presented in text 812. User 10 is angry that the printout is not as expected and expresses negative emotions in text 813. The system responds to text 813 as indicated by text 814. The system also determines whether to attempt a sale in response to this text. In the process of this inquiry, the system has already determined, based on the details of the session and the assumed state of user 10, that it is best not to attempt a sale at this time. More specifically, as in the case of text 812, when in Figure 6 When the prompt in the input field is text 813 immediately preceding the prompt and text 814 as the response details, generative AI 313 outputs "Inappropriate" as a determination of when to attempt a sale. This is to determine that the conversation flow is a result of a complaint and a response to the complaint, and that attempting a sale during this process is inappropriate.

[0072] Chat window 820 assumes that the same user made an inquiry in chat window 800 on a date after the date of the inquiry made in chat window 800. Text 830 corresponds to text 810. As indicated by text 831, user 10, for example, inquires about the functionality of printer 200 in a calm state.

[0073] The system responds to the query in text 831, as indicated in text 832. Similar to text 812, once a response to user 10's input text 831 has been created, it determines whether a sale should be attempted, and determines that it is preferable not to attempt a sale at this time. As a result, as indicated in text 832, only text 831 is responded to.

[0074] As indicated by text 833, the response from the system has resolved the problem faced by user 10. As indicated by text 834, after accepting the input of text 833, the system also attempted a sale in addition to the response in text 833. Although the system detected negative emotions when user 10 entered texts 811 and 813, it estimated that the user was experiencing positive emotions when entering text 833. Therefore, when responding to text 834, the system determined that the timing was suitable for attempting a sale and added a sale attempt in addition to the response to text 833. As indicated by text 835, the sale was attempted at exactly the right time, and user 10 became interested in the new product and confirmed the proposed website and introductory video.

[0075] As described above, as input to generative AI, the information processing apparatus according to this embodiment includes conditions for determining the timing of attempting to sell a product based on the customer's emotions inferred from the immediate preceding interaction. In this way, according to this embodiment, by considering the customer's emotions when determining the timing of a sale, a sale can be attempted at a more effective time. Furthermore, according to this embodiment, even if the product itself (printer) is isolated from the network, interaction with the customer can be achieved using a chat window displayed on the display unit of the computer 110 connected via the network, and thus the same effects as described in the first and second embodiments can be achieved.

[0076] Fourth embodiment The fourth embodiment of this disclosure will be described below. This embodiment will be referred to... Figure 9 This technology can be applied not only to inquiries about printers, but also to inquiries about other products. The construction according to this embodiment is similar to the construction described above in the third embodiment, and differs only in terms of the product under discussion.

[0077] Chat window Reference Figure 9 The description indicates a chat window in the interactive UI according to this embodiment. Chat windows 900 and 920 are chat UIs for conducting chat according to this disclosure. Input fields 901 and 921 are UIs through which user 10 inputs information. User 10 can send the text entered therein by pressing send buttons 902 and 922. Text 910 and text 930 are displayed when chat windows 900 and 920 are open, but this is not a necessary configuration.

[0078] Chat window 900 indicates that a chat should be processed for inquiries about cameras, lenses, etc., rather than inquiries about printers. As indicated by text 911, user 10 has inquired about methods for updating the firmware of a camera lens. The system presents a response to the inquiry in text 911, as indicated by text 912. The system determines that a sale will not be attempted here, and only uses the response to the inquiry presented in text 911 to confirm whether user 10 is satisfied with the generated response. Specifically, when text 911 is published, in step S501, a prompt “I want to update the lens firmware” is sent from computer 110 to control execution server 310. In step S504, generative AI 313 generates an answer based on the prompt. Generative AI 313 uses the response and sales text received in step S505 to determine whether the timing is suitable for attempting a sale. In other words, in Figure 6In the prompt, generative AI 313 takes the preceding prompt as text 911 and the response details as text 912. In text 912, it inputs a prompt that the sales text has been updated and determines the timing of the sale. Here, the result is that generative AI 313 outputs "Inappropriate." This is because it determines that the session has not ended and similar sessions will continue thereafter, and attempting to sell during this process is inappropriate.

[0079] As indicated by text 913, user 10 uses text 912 to indicate their satisfaction with the response. As indicated by text 914, upon receiving texts 911 and 913, the system determines that it is best to attempt a sale at the appropriate time to respond to text 913, and responds to text 913 to attempt a sale. At this point, the same processing as when displaying text 912 is performed, but generative AI 313 determines that the response to text 913 is an interruption in the session and determines that the sale attempt thereafter is appropriate. Therefore, in step S516, generative AI 313 uses the response to text 913 and the sale text to update the response details and performs the response as indicated by text 914. As indicated by text 915, user 10, who is attempted to sell at the appropriate time, may be interested in the lens being sold.

[0080] Chat window 920 indicates that the chat should handle inquiries about smartphone accessories, not printers. As indicated by text 931, user 10 has inquired about their concerns regarding upgrading their smartphone. The system presents a response to the inquiry in text 931, as indicated by text 932. The system here determines not to attempt a sale and only presents the response to the inquiry in text 931 to confirm the response and the smartphone being used. Specifically, the same processing is performed when displaying text 912; however, since the response to text 931 ends with a question, generative AI 313 determines that continuing the conversation thereafter is not appropriate to attempt a sale at the current time and only responds to text 931.

[0081] As indicated by text 933, user 10 inputs an indication that they have changed their smartphone model. As indicated by text 934, based on the details of text 933, the system responds with an indication that the smartphone accessory being used by user 10 cannot be used with the smartphone being used by user 10. At this point, the system determines that the current time is suitable for attempting a sale based on user 10's smartphone, and attempts to sell. Specifically, the same processing as when displaying text 912 is performed, but generative AI 313 determines the response to text 933 and determines that the current time is suitable for attempting to sell a new product. Therefore, in step S516, generative AI 313 uses the response to text 933 and the sales text to update the response content and makes a response as indicated by text 934. As indicated by text 935, user 10, who needs a product that matches their smartphone, inputs an indication that they wish to make a purchase. The system indicates that the order has been accepted, as indicated by text 936.

[0082] The first to third embodiments described above describe the case where the device is an image forming apparatus (printer) that provides a service. For example, in the case of an image forming apparatus, interaction takes place in a chat window on a print settings screen, and the product to be sold is a consumable used in the image forming apparatus or a new image forming apparatus product. However, as described in this embodiment, as a service provider, this technology can be applied to products other than printers, such as cameras or smartphones. For example, as mentioned above, in the case of a camera, this technology can be applied in interactions responding to inquiries about updating lens firmware. In the case of a smartphone, this technology can be applied in interactions responding to inquiries about smartphone connectivity.

[0083] in conclusion As described above, according to the embodiments, during a chat interaction with user 10, the optimal time to attempt a sale can be determined, and a sale can be attempted at that time. Note that this disclosure is not intended to be limited to the foregoing embodiments, and many variations and modifications can be made within its scope. This disclosure can also be implemented as a process performed by providing a program implementing one or more functions of the foregoing embodiments to a system or device via a network or storage medium, and then causing one or more processors of the system or device's computer to read and execute the program.

[0084] In this embodiment, a chat window 402 is displayed in the display unit 111 of the computer 110. The computer 110 interacts with the control mechanism execution server 310 and interacts with the user 10 in the chat window 402. However, this configuration is not limited to this. For example, the chat window can be provided as a chat for asking questions, and the UI can be displayed in a browser such as Microsoft Edge or Google Chrome. The chat window can also be displayed in the UI of a generative application such as ChatGPT, or in the UI used when printing using an application such as Microsoft Office. The chat window can also be displayed in the UI of a chat application such as Microsoft Teams or Copilot. In this case, this disclosure can be implemented such that the control mechanisms 311, 321, etc., are created as plugins of the UI and perform various processes. In addition, although the first to third embodiments describe formats applied to printers, this disclosure is not intended to be limited to printers and can be applied to products other than printers, as in the fourth embodiment described above.

[0085] Other embodiments The embodiments of this disclosure can also be implemented by a computer that reads and executes computer-executable instructions (e.g., one or more programs) recorded on a storage medium (which may also be more fully referred to as a "non-transitory computer-readable storage medium") to perform one or more functions of the above embodiments, and / or includes one or more circuits (e.g., application-specific integrated circuits (ASICs)) for performing one or more functions of the above embodiments. The embodiments of this disclosure can also be implemented by a method performed by a computer of a system or device, for example, reading and executing computer-executable instructions from a storage medium to perform one or more functions of the above embodiments, and / or controlling one or more circuits to perform one or more functions of the above embodiments. The computer may include one or more processors (e.g., a central processing unit (CPU), a microprocessor unit (MPU)) and may include a network of separate computers or separate processors to read and execute computer-executable instructions. The computer-executable instructions may be provided to the computer, for example, from a network or storage medium. Storage media may include, for example, hard disks, random access memory (RAM), read-only memory (ROM), storage devices for distributed computing systems, optical discs (such as optical discs (CD), digital versatile optical discs (DVD), or Blu-ray discs (BD)). TM One or more of the following: flash memory devices, memory cards, etc.

[0086] Other embodiments The embodiments of the present invention can also be implemented by the following method: providing software (including computer program products of computer programs) that performs the functions of the above embodiments to a system or device via a network or various storage media, and the computer (central processing unit (CPU) or microprocessor unit (MPU) of the system or device) reads out and executes the computer program.

[0087] Although this disclosure has been described with reference to embodiments, it should be understood that this disclosure is not limited to the disclosed embodiments. The scope of the appended claims should be given the broadest interpretation to cover all such variations and equivalent structures and functions.

Claims

1. An information processing apparatus, comprising: One or more storage devices, and their storage instruction sets; as well as One or more processors that execute the instruction set to: Provide customers of the service-providing devices with interactions using generative AI; The generative AI is input with at least one of the following information: interaction history with the customer, customer information, and product information. The generative AI then determines whether the current time is an appropriate time to attempt to sell the product. If the generative AI determines that the current time is a suitable time to attempt to sell the product, it responds to the interaction and proposes the sale of the product.

2. The information processing device according to claim 1, in, The interaction history includes not only past interactions with the customer but also interactions immediately preceding the interaction with the customer.

3. The information processing device according to claim 2, in, The one or more processors execute instructions in the one or more storage devices to store the interaction history of the customer in a database.

4. The information processing device according to claim 2, in, The interaction is an interaction in an inquiry about the device providing the service.

5. The information processing apparatus according to claim 4, in, In addition to at least one of the interaction history with the customer, the customer information, and the product information, the input to the generative AI also includes the determined details and the conditions used for the determination.

6. The information processing apparatus according to claim 5, in, The inputs to the generative AI include conditions for determining that the time prior to the completion of the query is not a suitable time to attempt to sell the product.

7. The information processing apparatus according to claim 5, in, The inputs to the generative AI include conditions for determining the appropriate time to try the product based on the customer's emotions inferred from the immediately preceding interaction.

8. The information processing apparatus according to claim 1, in, The one or more processors execute instructions in the one or more storage devices to present a sales attempt for the product together with an operational object that accepts the customer's response to the attempt.

9. The information processing apparatus according to claim 8, in, The customer information includes at least one of the customer's personal information, purchase history, repair history, information about the product currently in use, and response history.

10. The information processing apparatus according to claim 8, in, The one or more processors execute instructions in the one or more storage devices to order the product in response to obtaining customer consent for the proposed sale of the product.

11. The information processing apparatus according to claim 8, in, The one or more processors execute instructions in the one or more storage devices to interact with the customer via the display unit of the device providing the service or via the display unit of the computer used by the customer, the device and the customer being communicatively connected via a network.

12. The information processing apparatus according to claim 10, in, The apparatus providing the service is an image forming apparatus, and Interact in the chat window of the print settings screen.

13. The information processing apparatus according to claim 12, in, The products undergoing the sales attempt are consumables used in the image forming apparatus or new image forming apparatus products.

14. The information processing apparatus according to claim 10, in, The device that provides the service is a camera or a smartphone.

15. A method for controlling an information processing device, the method comprising: Provide customers of the service-providing devices with interactions using generative AI; The generative AI is input into the interaction history of the interaction with the customer, customer information, and product information, and the generative AI determines whether the current time is a suitable time to try selling the product. as well as If the generative AI determines that the current time is a suitable time to attempt to sell the product, it responds to the interaction and proposes the sale of the product.

16. A non-transitory computer-readable storage medium storing a computer program for causing a computer to perform steps of a method for controlling an information processing apparatus, the method comprising: Provide customers of the service-providing devices with interactions using generative AI; The generative AI is input into the interaction history of the interaction with the customer, customer information, and product information, and the generative AI determines whether the current time is a suitable time to try selling the product. as well as If the generative AI determines that the current time is a suitable time to attempt to sell the product, it responds to the interaction and proposes the sale of the product.

17. A computer program product storing a computer program for causing a computer to perform steps of a method for controlling an information processing apparatus, the method comprising: Provide customers of the service-providing devices with interactions using generative AI; The generative AI is input into the interaction history of the interaction with the customer, customer information, and product information, and the generative AI determines whether the current time is a suitable time to try selling the product. as well as If the generative AI determines that the current time is a suitable time to attempt to sell the product, it responds to the interaction and proposes the sale of the product.

Images