Processing device, processing program, processing method, and processing system

Abstract

[Problem] To provide a processing device, a processing program, a processing method, and a processing system with which it is possible to acquire advertisement information relating to an advertisement for a commodity or a service related to at least one of input information and answer information. [Solution] Provided is a processing device comprising at least one processor, wherein the at least one processor is configured to execute a process for: acquiring first input information generated on the basis of an operation input from a user in order to input the first input information to a first trained model configured to output first answer information; acquiring the first answer information from the first trained model by inputting the acquired first input information to the first trained model; and acquiring, on the basis of at least one of the first input information and the first answer information, advertisement information relating to an advertisement for a product or service related to at least one of the first input information and the first answer information.

Description

Processing apparatus, processing program, processing method, and processing system

[0001] This disclosure relates to a processing device, processing program, processing method, and processing system used for the use of a trained model.

[0002] Conventionally, systems have been known that generate prompts for input into large-scale language models, etc., for tasks such as document generation, and that can obtain predetermined response information by inputting said prompts into the large-scale language model. For example, Patent Document 1 describes a system comprising: an operation log acquisition unit that acquires operation logs of monitored terminals used for business operations; an operation log classification unit that classifies the operation logs acquired by the operation log acquisition unit by comparing them with business elements that serve as classification indicators for the business operations for each predetermined time frame; and a document creation unit that creates business support documents, including daily reports, based on the classification results classified by the operation log classification unit, wherein the document creation unit has a large-scale language model that probabilistically predicts how likely words and sentences given in prompts are to occur in natural language by combining one or more types of processing in natural language processing such as morphological analysis, syntactic analysis, semantic analysis, contextual analysis, and intent analysis, analyzes the prompts, and predicts and generates documents based on the content of the analyzed prompts. The description includes a "business support document creation device" comprising: a language model unit; a document creation prompt generation unit that outputs a prompt to the large-scale language model unit that instructs the creation of the business support document as a document based on the classification results; a business pattern formation unit that forms a business pattern for a certain period of time of business performed on the monitored terminal based on the classification results in the operation log classification unit; a business change detection unit that detects a qualitative change in the business performed on the monitored terminal based on the amount of change between the business patterns that are repeatedly formed at each of the certain periods; and a warning information output unit that, when a qualitative change in the business is detected, outputs the monitored terminal that caused the qualitative change, along with the content of the qualitative change, as warning information to the administrator terminal.

[0003] Patent No. 7572760

[0004] Therefore, based on the above technologies, an object of the present disclosure is to provide a processing device, a processing program, a processing method, and a processing system capable of acquiring advertisement information related to advertisements of products or services related to at least one of input information and response information according to various embodiments.

[0005] According to one aspect of the present disclosure, there is provided "a processing device including at least one processor, wherein the at least one processor acquires first input information generated based on an operation input from a user in order to input the first input information into a first learned model configured to output first response information, inputs the acquired first input information into the first learned model to acquire the first response information from the first learned model, and acquires advertisement information related to an advertisement of a product or service related to at least one of the first input information and the first response information based on at least one of the first input information and the first response information, and is configured to execute processing for this purpose."

[0006] According to one aspect of the present disclosure, there is provided "a processing program for causing a computer including at least one processor to function such that the at least one processor acquires first input information generated based on an operation input from a user in order to input the first input information into a first learned model configured to output first response information, inputs the acquired first input information into the first learned model to acquire the first response information from the first learned model, and acquires advertisement information related to an advertisement of a product or service related to at least one of the first input information and the first response information based on at least one of the first input information and the first response information, and executes processing for this purpose."

[0007] According to one aspect of the present disclosure, a processing method is provided for a computer having at least one processor, which is performed by the at least one processor and includes the steps of: acquiring first input information generated based on user input for input to a first trained model configured to output first response information; acquiring first response information from the first trained model by inputting the acquired first input information into the first trained model; and acquiring advertising information relating to an advertisement for a product or service related to at least one of the first input information and the first response information, based on at least one of the first input information and the first response information.

[0008] According to one aspect of this disclosure, a processing system is provided that includes "the processing device described above and a terminal device connected to the processing device via a communication network and configured to transmit the first input information to the processing device via the communication network."

[0009] According to this disclosure, it is possible to provide a processing device, a processing program, a processing method, and a processing system that can acquire advertising information relating to advertisements for goods or services related to at least one of the input information and the response information.

[0010] The effects described above are merely illustrative for the sake of explanation and are not limiting. In addition to, or in lieu of, any other effects described herein or that would be obvious to those skilled in the art may be achieved.

[0011] Figure 1 is a block diagram showing the configuration of a processing system 1 according to one embodiment of the present disclosure. Figure 2A is a block diagram showing the configuration of a server device 100 according to one embodiment of the present disclosure. Figure 2B is a block diagram showing the configuration of a terminal device 200 according to one embodiment of the present disclosure. Figure 3A is a diagram conceptually showing a user management table stored in the server device 100 according to one embodiment of the present disclosure. Figure 3B is a diagram conceptually showing a trained model management table stored in the server device 100 according to one embodiment of the present disclosure. Figure 4 is a diagram conceptually showing an example of the trained model selection process in the server device 100 according to one embodiment of the present disclosure. Figure 5A is a diagram showing an example of a processing sequence executed in the processing system 1 according to one embodiment of the present disclosure. Figure 5B is a diagram showing an example of a processing sequence executed in the processing system 1 according to one embodiment of the present disclosure. Figure 5C is a diagram showing an example of a processing sequence executed in the processing system 1 according to one embodiment of the present disclosure. Figure 6A is a diagram showing an example of a processing flow executed in the server device 100 according to one embodiment of the present disclosure. Figure 6B is a diagram showing an example of a processing flow executed in the server device 100 according to one embodiment of the present disclosure. Figure 6C is a diagram showing an example of a processing flow executed in the server device 100 according to an embodiment of this disclosure. Figure 6D is a diagram showing an example of a processing flow executed in the server device 100 according to an embodiment of this disclosure. Figure 6E is a diagram showing an example of processing for obtaining corrected response information executed in the server device 100 according to an embodiment of this disclosure. Figure 7 is a diagram showing an example of a first display screen 10 output by the terminal device 200 in an embodiment of this disclosure. Figure 8 is a diagram showing an example of a first display screen 10 output by the terminal device 200 in an embodiment of this disclosure. Figure 9 is a diagram showing an example of a first display screen 10 output by the terminal device 200 in an embodiment of this disclosure. Figure 10 is a diagram showing an example of a second display screen 40 output by the terminal device 200 in an embodiment of this disclosure. Figure 11 is a diagram showing an example of a second display screen 40 output by the terminal device 200 in an embodiment of this disclosure. Figure 12 is a diagram showing an example of a second display screen 40 output by the terminal device 200 in an embodiment of this disclosure.Figure 13 is a diagram showing an example of a second display screen 40 output by the terminal device 200 in an embodiment of this disclosure. Figure 14 is a diagram showing an example of a second display screen 40 output by the terminal device 200 in an embodiment of this disclosure. Figure 15 is a diagram showing an example of a second display screen 40 output by the terminal device 200 in an embodiment of this disclosure. Figure 16 is a diagram showing an example of a second display screen 40 output by the terminal device 200 in an embodiment of this disclosure. Figure 17 is a diagram showing an example of a first display screen 10 output by the terminal device 200 in an embodiment of this disclosure.

[0012] Various embodiments of the present invention will be described below with reference to the attached drawings. Note that common components in the drawings are denoted by the same reference numerals. Also, please note that components shown in one drawing may be omitted in another drawing for the sake of clarity. Furthermore, please note that the attached drawings are not necessarily drawn to an exact scale.

[0013] The various systems, methods, and apparatus described in this disclosure should not be construed as limiting in any way. In practice, this disclosure is directed to any novel features and aspects of each of the various embodiments disclosed, combinations of these various embodiments, and combinations of some of these various embodiments. The various systems, methods, and apparatus described in this disclosure are not limited to any particular aspect, particular feature, or combination of such particular aspects and particular features, and the things and methods described in this disclosure do not require that one or more particular effects exist or problems are solved. Furthermore, various features or aspects of the various embodiments described in this disclosure, or some of such features or aspects, may be used in combination with each other.

[0014] While the operation of some of the various methods disclosed in this disclosure is described in a specific order for convenience, this method of description should be understood to include the possibility of rearranging the order of the operations unless a specific order is required by the following specific sentences. For example, multiple operations described in order may, in some cases, be rearranged or performed simultaneously. Furthermore, for the sake of simplification, the accompanying drawings do not show various ways in which the various matters and methods described in this disclosure may be used in conjunction with other matters and methods.

[0015] Any operating theories, scientific principles, or other theoretical descriptions presented in connection with the apparatus or method of this disclosure are provided for the purpose of better understanding and are not intended to limit the technical scope. The apparatus and method in the appended claims are not limited to apparatus and method that operate in the manner described by such operating theories.

[0016] Any of the various methods disclosed herein can be implemented using a plurality of computer-executable instructions stored on one or more computer-readable media, and can be executed on a computer. The one or more media may be non-transient computer-readable storage media such as, for example, at least one optical media disk, a plurality of volatile memory components, or a plurality of non-volatile memory components. The plurality of volatile memory components include, for example, DRAM or SRAM. The plurality of non-volatile memory components include, for example, hard drives and solid-state drives (SSDs). Furthermore, the computer includes any computer available on the market, including, for example, smartphones and other mobile devices having computing hardware.

[0017] Any of the multiple computer-executable instructions for implementing the technology disclosed herein may be stored in one or more computer-readable media (e.g., non-temporary computer-readable storage media) along with any data generated and used during implementations of the various embodiments disclosed herein. Such multiple computer-executable instructions may, for example, be part of a separate software application, or part of a software application accessed or downloaded via a web browser or other software application (such as a remote computing application). Such software may be executed, for example, on a single local computer (as a process run on any suitable computer available on the market), or in a network environment (e.g., the Internet, a wide area network, a local area network, a client-server network (such as a cloud computing network), or other such network) using one or more network computers.

[0018] For clarity, only specific selected aspects of various software-based implementations are described. Other details that are well known in the art are omitted. For example, the technology disclosed in this disclosure is not limited to any particular computer language or program. For example, the technology disclosed in this disclosure may be executed by software written in C, C++, Java®, or any other suitable programming language. Similarly, the technology disclosed in this disclosure is not limited to any particular computer or type of hardware. Specific details of suitable computers and hardware are well known and do not need to be described in detail in this disclosure.

[0019] Furthermore, any of the various embodiments of such software (including, for example, a set of computer-executable instructions for causing a computer to perform any of the various methods disclosed herein) may be uploaded, downloaded, or accessed remotely by preferred means of communication. Such preferred means of communication include, for example, the Internet, the World Wide Web, intranets, software applications, cables (including fiber optic cables), magnetic communications, electromagnetic communications (including RF communications, microwave communications, and infrared communications), electronic communications, or other such means of communication.

[0020] 1. Schematic diagram 1 of the processing system 1 is a block diagram showing the configuration of the processing system 1 according to one embodiment of the present disclosure. According to Figure 1, the processing system 1 includes at least a server device 100 and a terminal device 200, and each device is connected to communicate via a wired or wireless network. The server device 100, for example, acquires first input information from the terminal device 200 while outputting first response information and advertising information to the terminal device 200. The terminal device 200, for example, is configured to be held by the user and receives first response information and advertising information from the server device 100 while transmitting first input information to the server device 100.

[0021] Such a processing system 1 is used to acquire advertising information relating to advertisements for products or services related to at least one of the input information and the response information. Such a processing system 1 is also used to obtain the output of arbitrary response information from a trained machine learning model. Specifically, the processing system 1 acquires first input information generated based on user input in order to input it into a first trained model configured to output first response information. The processing system 1 also acquires first response information from the first trained model by inputting the acquired first input information into the first trained model. Furthermore, the processing system 1 acquires advertising information relating to advertisements for products or services related to at least one of the first input information and the first response information, based on at least one of the first input information and the first response information.

[0022] Therefore, the processing system 1 can acquire advertising information related to products or services associated with at least one of the input information and the response information. Thus, it can increase the user's willingness to purchase.

[0023] In this disclosure, "processing device" means a server device 100, a terminal device 200, or a combination thereof. That is, the following description will focus on the case where the server device 100 functions as a processing device, but the terminal device 200 can also function as a processing device in the same way. Furthermore, in this disclosure, the storage and processing performed by the processing device may be distributed to other terminal devices or other server devices. In other words, the processing device is not limited to those consisting of a single enclosure, but includes the server device 100, the terminal device 200, other server devices, other terminal devices, or a combination thereof.

[0024] Furthermore, in this disclosure, the "trained model" can be any model that outputs response information when arbitrary input information is input. Such a trained model is a model that has undergone arbitrary machine learning. As an example, a trained model can be generated by preparing multiple combinations of training data and the correct labels assigned to said training data, and inputting these combinations into a learner to machine learn the correct patterns. Examples of such learning models include neural networks, convolutional neural networks, multilayer Herceptons (MLP), LSTM (Long Short Term Memory), GRU (Gated Recurrent Unit), GNN (Graph Neural Network), Transformers, and other neural network-based methods; gradient boosting decision trees (GBDT) such as LightGBM (Light Gradient Boosting Machine), XGBoost, and CatBoost; and various learning models using ridge regression, logistic regression, support vector regression (SVR), nearest neighbor, decision trees, regression trees, random forests, or combinations thereof.

[0025] Furthermore, pre-trained models include generative pre-trained models such as Large Language Models (LLMs), which are generated by deep learning using large amounts of text data. Examples of such pre-trained models include BERT (Bidirectional Encoder Representations from Transformers), GPT (Generative Pre-trained Transformer), Gemini, DALL-E, Midjourney, or combinations thereof.

[0026] As illustrated above, the trained model will output response information when arbitrary input information is provided. The input information can be in various formats, including text, images (including both video and still images), audio, documents, PDFs, links, or combinations thereof. Similarly, the output response information can be in various formats, including text, images (including both video and still images), audio, documents, PDFs, links, or combinations thereof. For the sake of explanation, the following will describe the case where the input information is in text format and the response information is also in text format, but it is certainly not limited to this case.

[0027] 2. Configuration diagram 2A of the server device 100 is a block diagram showing the configuration of a server device 100 according to one embodiment of the present disclosure. According to Figure 2A, the server device 100 includes a processor 111, a memory 112, and a communication interface 113. Each of these components is electrically connected to the others via control lines and data lines. The server device 100 does not need to have all of the components shown in Figure 2A; it is possible to omit some components or add other components. For example, it is possible to use an external memory connected via communication as memory, a database (DB) device, or other server devices. It is also possible to distribute and execute some processing with processing devices including other server devices. In other words, the server device 100 is not limited to a single device, but also includes cases where it is distributed across multiple devices depending on the handling of information and the processing load.

[0028] The processor 111 functions as a control unit that controls other components of the processing system 1 based on a processing program stored in the memory 112. Based on the processing program stored in the memory 112, the processor 111 executes processes related to the generation of input information to be input to the trained model. Specifically, it executes processes such as "a process to acquire first input information generated based on user operation input via the communication interface 113," "a process to acquire first response information from the first trained model by inputting the acquired first input information to the first trained model via the communication interface 113," and "a process to acquire advertising information related to advertisements for products or services associated with at least one of the first input information and the first response information, based on at least one of the first input information and the first response information," based on the processing program stored in the memory 112. The processor 111 is mainly composed of one or more CPUs, but may be combined with a GPU or FPGA as appropriate.

[0029] Memory 112 is composed of RAM, ROM, non-volatile memory, HDD, SSD, etc., and functions as a storage unit that stores various information. Memory 112 stores instruction commands for various controls of the processing system 1 according to this embodiment as processing programs. Specifically, memory 112 stores processing programs for execution by the processor 111, such as "a process to acquire first input information generated based on user operation input via the communication interface 113", "a process to acquire first response information from the first trained model by inputting the acquired first input information to the first trained model via the communication interface 113", and "a process to acquire advertising information related to advertisements for products or services associated with at least one of the first input information and the first response information, based on at least one of the first input information and the first response information". In addition to these programs, memory 112 also stores various information stored in a user management table or a trained model management table, etc. Note that this information does not need to be constantly stored in memory 112 within the server device 100, but may be stored in a remotely installed database (DB) device. In that case, the database device is also included in memory 112. This information may also be recorded in the memory 212 of the terminal device 200.

[0030] The communication interface 113 functions as a notification unit for sending and receiving various information with a terminal device 200 connected via a wired or wireless network. Examples of the communication interface 113 include wired communication connectors such as USB and SCSI, wireless communication transceivers such as wireless LAN, Bluetooth®, LTE, and infrared, and various connection terminals for printed circuit boards and flexible circuit boards. For example, the communication interface 113 receives first input information from the terminal device 200 or transmits first response information or advertising information to the terminal device 200.

[0031] 3. Diagram 2B of the configuration of the terminal device 200 is a block diagram showing the configuration of a terminal device 200 according to one embodiment of the present disclosure. According to Figure 2B, the terminal device 200 includes a processor 211, a memory 212, an input interface 213, an output interface 214, and a communication interface 215. Each of these components is electrically connected to the others via control lines and data lines. Note that the terminal device 200 does not need to have all of the components shown in Figure 2B; it is possible to omit some components or add other components. The terminal device 200 can be any device that can communicate with the server device 100 via a wired or wireless network, and smartphones, tablet devices, laptop PCs, desktop PCs, etc., can be used as terminal devices 200. Note that if there are multiple users as described above, multiple terminal devices 200 will be used for each user, but each terminal device 200 may be a different type of terminal device. Furthermore, it is not necessarily required that there be one terminal device 200 for each user; multiple users may use one terminal device 200, or one user may use multiple terminal devices 200.

[0032] The processor 211 functions as a control unit that controls other components of the processing system 1 based on a processing program stored in the memory 212. Based on the processing program stored in the memory 212, the processor 211 executes processes related to the generation of input information. Specifically, it executes processes such as "receiving user operation input via the input interface 213 and generating first input information," "receiving first response information obtained by inputting the first input information into the first trained model via the communication interface 215 and outputting the first response information via the output interface 214," and "receiving advertising information related to an advertisement for a product or service related to at least one of the first input information and the first response information and outputting the advertising information via the output interface 214," based on the processing program stored in the memory 212. The processor 111 is mainly composed of one or more CPUs, but a GPU or FPGA may be combined as appropriate.

[0033] Memory 212 is composed of RAM, ROM, non-volatile memory, HDD, SSD, etc., and functions as a storage unit. Memory 212 stores instruction commands for various controls of the processing system 1 according to this embodiment as programs. Specifically, memory 212 stores processing programs for the processor 211 to execute, such as "a process that receives user operation input via the input interface 213 and generates first input information," "a process that receives first response information obtained when the first input information is input to the first trained model via the communication interface 215 and outputs the first response information via the output interface 214," and "a process that receives advertising information relating to an advertisement for a product or service related to at least one of the first input information and the first response information and outputs the advertising information via the output interface 214."

[0034] The input interface 213 functions as an input unit that receives user input to the terminal device 200. Examples of the input interface 213 include physical key buttons and a touch panel having an input coordinate system corresponding to the display coordinate system of the display. In the case of a touch panel, icons are displayed on the display, and the user makes a selection for each icon by making an input via the touch panel. The method for detecting the user's input via the touch panel can be any method, such as capacitive or resistive. The input interface 213 does not always need to be physically provided on the terminal device 200 and may be connected as needed via a wired or wireless network. Therefore, in addition to the above, a mouse or keyboard can also be used as the input interface 213. The input interface 213 may also include a microphone that acquires information in audio format.

[0035] The output interface 214 functions as an output unit for outputting various information, such as the first display screen or the second display screen. An example of the output interface 214 is a display composed of a liquid crystal panel, an organic EL display, or a plasma display. However, the terminal device 200 itself does not necessarily need to be equipped with a display. For example, an interface for connecting to a display that can be connected to the terminal device 200 via a wired or wireless network can also function as the output interface 214 for outputting display data to the display.

[0036] The communication interface 215 functions as a communication unit for sending and receiving various types of information with the server device 100, which is connected via a wired or wireless network. Examples of the communication interface 215 include wired communication connectors such as USB and SCSI, wireless communication transceivers such as wireless LAN, Bluetooth®, LTE, and infrared, and various connection terminals for printed circuit boards and flexible circuit boards. For example, the communication interface 215 receives first response information or advertising information from the server device 100, or transmits first input information to the server device 100.

[0037] 4. Various types of information used in processing in processing system 1 (1) User management table Figure 3A is a conceptual diagram showing a user management table stored in a server device 100 according to one embodiment of the present disclosure. According to Figure 3A, the user management table stores user name information, attribute information, first model designation information, second model designation information, and third model designation information, etc., associated with user ID information.

[0038] "User ID information" is unique information for each user, used to identify each user. Examples of such user ID information include various pieces of information such as arbitrary identification information assigned by the server device 100, email addresses, SNS accounts, telephone numbers, or combinations thereof. User ID information is generated by the server device 100, for example, each time a request for new registration as a user of a service provided by the processing system 1 via the terminal device 200 is received.

[0039] "Username information" is information that indicates the name of each user. Such username information can include various types of information, such as the user's name, nickname, SNS account name, or a combination of these. Username information is, for example, information specified by the user when registering as a new user on the aforementioned service.

[0040] "Attribute information" refers to information that indicates the attributes of each user. Examples of such attribute information include various types of information such as authentication information, gender, age, occupation, address, telephone number, email address, SNS account, workplace, membership rank of the above service, past usage history of the above service, search and browsing history of websites, etc., or combinations thereof. In this embodiment, as shown in Figure 4, one or more trained models are selected and input information is entered, and attribute information can be used to select the trained model. Attribute information may be, for example, information specified by the user when registering as a new user to the above service, or information collected by the server device 100.

[0041] "First model designation information" is information for identifying the trained model to which the first input information is input from among one or more trained models. "Second model designation information" is information for identifying the trained model to which the second input information is input from among one or more trained models. "Third model designation information" is information for identifying the trained model to which the third input information is input from among one or more trained models. Hereafter in this specification, when "model designation information" is used, "model designation information" includes "first model designation information," "second model designation information," and "third model designation information." Hereafter in this specification, when "input information" is used, "input information" includes "first input information," "second input information," and "third input information." Hereafter in this specification, when "response information" is used, "response information" includes "first response information," "second response information," and "third response information." Furthermore, in this specification, when the term "trained model" is used, it includes the "first trained model," the "second trained model," and the "third trained model." Examples of such model specification information include various pieces of information, such as model ID information to identify each trained model, a name indicating each trained model, information indicating the storage location of each trained model, or a combination thereof. The model specification information stores information that identifies which of the one or more trained models, including trained model A, trained model B, and trained model C, the input information will be input to. This model specification information is generated when the processor 211 of the terminal device 200 receives user input via the input interface 213 and selects the desired trained model from among the one or more trained models. In other words, in this embodiment, the user can also select the trained model to which the input will be received. Note that information identifying only one trained model may be stored, or information identifying multiple trained models may be stored.Also, when information for identifying a plurality of learned models is stored, the model specification information may store information for specifying a method for generating response information, such as whether to output, as response information, a plurality of pieces of output information output from each learned model as they are, or whether to perform ensemble processing on the plurality of pieces of output information output from each learned model and output the result as response information.

[0042] Each piece of information stored in the user management table is typically stored in the memory 112 of the server device 100. However, naturally, it may be stored in a database device or other server devices. Also, each piece of information shown in FIG. 3A is an example of the information stored in the user management table, and naturally, other information may be stored.

[0043] (3) Model Management Table FIG. 3B is a diagram conceptually showing a learned model management table stored in the server device 100 according to an embodiment of the present disclosure. According to FIG. 3B, condition information is stored in association with model ID information.

[0044] "Model ID information" is information unique to each learned model to which input information is input, and is information for identifying each learned model. In the present embodiment, as shown in FIG. 4, it is possible to use a plurality of learned models. Such model ID information is information assigned by the server device 100 each time each learned model is newly registered as an available model for each learned model.

[0045] "Conditional information" refers to the conditions under which input information is entered and response information is obtained for each trained model identified by the model ID information; in other words, it indicates the conditions for using each trained model. Here, each trained model differs in the information used for machine learning, the machine learning method, performance, specifications, or the learner used for machine learning. For example, each trained model may differ in whether the input information is used as training data, whether confidential information such as personal information can be entered, or whether it can be used by minors. Therefore, it is possible to set conditions that can be determined based on the user's attribute information as conditional information. Examples of such conditional information include whether the user's age is above a certain age, whether the user's address is in a certain country, whether the user's workplace allows the use of trained models, whether the service's lowest rank is above a certain rank, or a combination of these, and so on.

[0046] Each piece of information stored in the model management table is typically stored in the memory 112 of the server device 100, but of course, it may also be stored in a database device or other server devices. Furthermore, the information shown in Figure 3B is just an example of the information stored in the model management table, and of course, other information may also be stored.

[0047] Figure 4 is a conceptual diagram showing an example of the selection process for a trained model in a server device 100 according to one embodiment of the present disclosure. Specifically, Figure 4 is a diagram showing the connection relationship between the server device 100 and each processing unit that stores one or more trained models into which input information is input. According to Figure 4, any number of processing units 300n are connected to the server device 100, such as a processing unit 300A that can process based on trained model A, a processing unit 300B that can process based on trained model B, and a processing unit 300C that can process based on trained model C. When input information is generated, the server device 100 transmits the input information and processing request to the processing unit that stores the trained model identified by the model specification information in Figure 3A, and obtains response information from the processing unit.

[0048] In FIG. 4, each process by the learned models including the learned models A to C has been described for the case where it is executed by each processing device. However, naturally, the server device 100 may read each learned model from the memory and execute it by the server device 100. Further, the learned model to be executed does not have to be one for one input information. A plurality of learned models may be executed for one input information, and a plurality of answer information or a plurality of output information may be subjected to ensemble processing to obtain the answer information.

[0049] In addition, for each learned model, any learned model such as a learned model obtained by machine learning the combination of the learning data and the correct label exemplified above, a generative learned model, etc. can be used.

[0050] 5. Processing Sequence Executed by Processing System 1 FIG. 5A is a diagram showing an example of a processing sequence executed in the processing system 1 according to an embodiment of the present disclosure. Specifically, FIG. 5A is a diagram showing a series of processing flows related to the output of the first answer information.

[0051] According to FIG. 5A, the processor 211 of the terminal device 200 receives a user's operation input via the input interface 213 and starts an application program for using the service provided via the processing system 1 (S11). Then, the processor 211 receives a user's operation input via the input interface 213 and inputs user ID information registered in advance as a user who can use the service and authentication information for authenticating the user (S12). When the user ID information and the authentication information are input, the processor 211 transmits the user ID information etc. (T11) to the server device 100 via the communication interface 215.

[0052] When the processor 111 of the server device 100 receives user ID information and authentication information from the terminal device 200 via the communication interface 113, it authenticates whether the user is a legitimate user of the service by referring to the user ID information and attribute information in the user management table (S13). If the processor 111 is authenticated as a legitimate user, it reads user information from the user management table based on the user ID information, including attribute information, model specification information, or username information of the user. The processor 111 then transmits the read user information (T12) to the terminal device 200 that sent the user ID information via the communication interface 113.

[0053] When the processor 211 of the terminal device 200 receives user information via the communication interface 215, it generates a first display screen 10 for receiving user input and generating first input information in order to input it into a first trained model configured to output first response information. The processor 211 then outputs the generated first display screen 10 to the display via the output interface 214 (S14). The processor 111 may also transmit information for displaying the first display screen 10 (first display screen information) along with user information (T12) via the communication interface 113, and the processor 211 of the terminal device 200 may receive this first display screen information via the communication interface 215.

[0054] Here, Figure 7 shows an example of a first display screen 10 output by the terminal device 200 in an embodiment of the present disclosure. Specifically, Figure 7 shows an example of a first display screen 10 output in S14 of Figure 5A. According to Figure 7, the first display screen 10 includes a first trained model selection area 12, a display area 14, a file addition area 16, a text input area 18, a voice input button 22, and a send button 24. The first display screen 10 may output a user ID.

[0055] The first pre-trained model selection region 12 outputs the first pre-trained model selected by receiving user input via the input interface 213. The user can change the first pre-trained model by selecting the first pre-trained model selection region 12. In Figure 7, the first pre-trained model selection region 12 outputs "Gemini". In other words, in this example, "Gemini" is selected as the first pre-trained model. That is, in this example, the first model specification information identifies "Gemini".

[0056] Display area 14 outputs the first input information, the first response information, etc. As will be described later, the first input information and the first response information may be output in chat format in display area 14. That is, in display area 14, the first input information may be output to the right of the first display screen 10, and the first response information may be output to the left of the first display screen 10. In this example, since no first input information has been entered, the initial output in display area 14 is "Hello. Please ask me anything."

[0057] The file addition area 16, text input area 18, voice input button 22, and send button 24 are used to generate the first input information. In this example, the file addition area 16 displays "+". When the user selects the file addition area 16, the processor 211 of the terminal device 200 uses various files such as image files, audio files, document files, PDF files, and link files to generate the first input information. In other words, by selecting the file addition area 16, the user can use various files such as image files, audio files, document files, PDF files, and link files to generate the first input information. The text input area 18 outputs any string entered by the user through the input interface 213. When the user operates the text input area 18, the processor 211 of the terminal device 200 uses the string in text format to generate the first input information. In this example, the voice input button 22 displays a microphone icon. When the user selects the voice input button 22, the processor 211 of the terminal device 200 can accept voice input via the input interface 213. Therefore, when the user selects the voice input button 22, the processor 211 of the terminal device 200 uses the voice-formatted information to generate the first input information. After the file addition area 16, text input area 18, and voice input button 22 are operated, the send button 24 is selected, causing the processor 211 of the terminal device 200 to generate the first input information. Note that at least one of the file addition area 16, text input area 18, and voice input button 22 may be operated or selected to generate the first input information, and multiple of the file addition area 16, text input area 18, and voice input button 22 may be operated or selected.

[0058] Returning to Figure 5A, when the first display screen 10 illustrated in Figure 7 is output, the processor 211 of the terminal device 200 generates first input information based on the user's operation input received via the input interface 213 (S15).

[0059] Here, Figure 8 shows an example of the first display screen 10 output by the terminal device 200 in the embodiment of this disclosure. Specifically, Figure 8 shows an example of the first display screen 10 output in S15 of Figure 5A. In this example, the first display screen 10 differs from Figure 7 in that the first input information 28 displays "Please tell me your recommendations for recently released teas." The configuration other than the display of "Please tell me your recommendations for recently released teas." may be the same as in Figure 7. The first input information 28 is generated, for example, when the user operates the text input area 18 and selects the send button 24.

[0060] Returning to Figure 5A, once the first input information is generated, the processor 211 of the terminal device 200 sends a first response generation request, including the first input information (T13), to the server device 100 via the communication interface 215. In this example, the processor 211 of the terminal device 200 generates the first input information, but the processor 111 of the server device 100 may also generate the first input information. When the processor 111 of the server device 100 generates the first input information, the processor 111 of the server device 100 generates the first input information based on information regarding user operations and selections in the file addition area 16, text input area 18, voice input button 22, etc.

[0061] When the processor 111 of the server device 100 receives a first response generation request, including first input information, from the terminal device 200 via the communication interface 113, the processor 111 of the server device 100 executes the first response generation process by inputting the first input information generated in the first trained model selected based on at least one of the first model specification information in the user management table and the condition information in the model management table (S16). In this example, since the first model specification information specifies "Gemini", the first response generation process is executed by inputting the first input information generated in the first trained model selected based on the first model specification information in the user management table. Through this first response generation process, the processor 111 obtains the first response information for the first input information from the first trained model. Details of the first response generation process will be explained in Figure 6A.

[0062] When the processor 111 of the server device 100 obtains the first response information, it determines whether to output a notification suggesting obtaining second response information, which is different from the first response information, from a second trained model, which is different from the first trained model, based on at least one of the first input information, the first response information, and the user's attribute information (S17). The processor 111 of the server device 100 may also determine whether to output a notification when it receives a first response generation request, including the first input information, from the terminal device 200 via the communication interface 113. Since the notification is output only when it is determined to be "OK", the user can easily obtain not only the first response information but also the second response information, making it possible to obtain response information output by a trained machine learning model more efficiently.

[0063] The difference between the first and second pre-trained models may be, for example, a difference in the training models used. For example, the first model specification information might identify "Gemini," while the second model specification information might identify "GPT-4." The difference between the first and second pre-trained models may also be, for example, a difference in the training methods used. A difference in training methods might mean, for example, a difference in the period over which the information used for training is acquired. For example, the first pre-trained model is trained on information acquired up to August 2023, and the second pre-trained model is trained on information acquired up to November 2024. A difference in training methods might mean, for example, a difference in the period over which the information used for training is acquired. A difference in training methods might mean, for example, a difference in the datasets used for training. For example, the first pre-trained model might use Common Crawl Corpus as its dataset, while the second pre-trained model might use BookCorpus as its dataset. Furthermore, a difference between the first and second trained models may be that the first trained model does not use a search process that retrieves relevant information from a database for a specific question or request and generates an answer based on that information, while the second trained model does use such a search process. Known methods such as exact match, partial match, and similarity match can be used as the search process. Another difference between the first and second trained models may be that the first trained model does not use the grounding described later, while the second trained model does. A further difference between the first and second trained models may be that the first trained model does not use the fine-tuning described later, while the second trained model does. A further difference between the first and second trained models may be that the evaluation models are different. A further difference between the first and second trained models may be that the first trained model is available for free, while the second trained model is available for a fee. Finally, a further difference between the first and second trained models may be that, for example, they can handle different data formats (text format, image format, etc.). Furthermore, the difference between the first and second trained models may simply be in their display methods.For example, the difference between the first and second trained models may be that the response information output by the first trained model does not include links to web pages, while the response information output by the second trained model does include links to web pages.

[0064] As an example, determining whether or not to output a notification based on the first input information means that if the first input information contains a specific string, the notification output will be determined as "OK". The specific string is a string that suggests the user wants to obtain the latest answer information, such as "latest", "new", "state-of-the-art", "cutting edge", "new arrival", "new publication", "new release", etc. Depending on the training status of the configured first trained model, the latest information may not be included in the first answer information. For example, if the first trained model has learned information up to August 2023, information from September 2023 onwards will not be included in the first answer information. Therefore, if the first input information contains a string that suggests the user wants to obtain the latest answer information, the processor 111 of the server device 100 outputs a notification so that the latest information can be obtained from the second trained model.

[0065] The processor 111 of the server device 100 may determine whether or not to output a notification based on the history of the first input information. For example, the determination of whether or not to output a notification may be based not only on the first input information generated at the S15 stage, but also on the first input information generated before S15. In this case, even if it is not inferred from the most recently generated first input information that the user wishes to obtain the latest response information, the determination of whether or not to output a notification can be made based on the history of the first input information. As an example, determining whether or not to output a notification based on the history of the first input information means that if a specific string is included in the history of the first input information, the notification output will be determined to be "OK".

[0066] As an example, determining whether or not to output a notification based on the first response information means that if it is presumed that the information contained in the first response information is insufficient as response information, then the notification will be output "OK". For example, if the first response information contains the information "We recommend the tea that was released in July 2023", and the current date and time is December 2024, then the date and time contained in the first response information and the current date and time differ by more than one year, so it is presumed that the information contained in the first response information is insufficient as response information. In other words, it is acceptable to determine whether or not to output a notification based on the date and time information contained in the first response information. Alternatively, it is also acceptable to determine whether or not to output a notification based on the first input information and the first response information. For example, it is acceptable to determine whether or not to output a notification based on the date and time information contained in the first input information and the date and time information contained in the first response information.

[0067] The processor 111 of the server device 100 may determine whether or not to output a notification based on the first input information and the first model designation information. If it is determined that the first response information alone is insufficient based on the first input information and the first model designation information, the processor 111 of the server device 100 may determine "OK" to output the notification. For example, if the first input information includes a string that suggests the user wants to obtain the latest response information, and the first trained model specified in the first model designation information has not yet learned the new information, the processor 111 of the server device 100 may determine "OK" to output the notification. If it is determined that the first trained model specified in the first model designation information has already learned the new information, the processor 111 of the server device 100 may determine "No" to output the notification. For example, if the first input information includes a string that suggests the user wants to obtain the latest response information, and the first trained model specified in the first model designation information has already learned relatively new information, the processor 111 of the server device 100 may determine "No" to output the notification. To determine whether new information has already been learned, for example, the system may determine whether information from a predetermined period (specifically, three months ago, etc.) has been learned. Similarly, the processor 111 of the server device 100 may determine whether or not to output a notification based on the first response information and the first model designation information. Similarly, the processor 111 of the server device 100 may determine whether or not to output a notification based on the first input information, the first response information, and the first model designation information.

[0068] Furthermore, determining whether or not to output a notification based on user attribute information means that if the user's attribute information matches predetermined conditions, the system will determine "yes" to outputting a notification. For example, if the user's attribute information includes a specific professional occupation such as doctor, lawyer, patent attorney, or administrative scrivener, it is inferred that the user wishes to obtain the latest response information. Therefore, if it is inferred that the user wishes to obtain the latest response information based on the user's attribute information, the processor 111 of the server device 100 will output a notification so that the latest information can be obtained. In addition, if the user's attribute information includes a specific professional occupation such as doctor, lawyer, patent attorney, or administrative scrivener, it is particularly inferred that the user wishes to obtain the latest response information in the field of that profession (for example, the medical field if the user is a doctor). Therefore, the processor 111 of the server device 100 may, as an example, determine whether or not to output a notification based on the user's attribute information and the first input information.

[0069] When the processor 111 determines whether or not to output a notification, it transmits the acquired first response information (T14) to the terminal device 200 that sent the first response generation request via the communication interface 113. At this time, if the processor 111 determines that the notification output is "OK", it transmits a notification (T15) along with the first response information (T14). Then, the processor 211 outputs the first display screen 10, which displays the received first response information, to the display via the output interface 214 (S18).

[0070] Next, Figure 5B will be described. Figure 5B is a diagram showing an example of a processing sequence executed in the processing system 1 according to the embodiment of this disclosure. Specifically, Figure 5B is a diagram showing a series of processing flows related to the output of the second response information and advertising information. Figure 5B is also a diagram showing a series of processing flows when "OK" is determined in the output of the notification.

[0071] If the processor 111 determines that the notification output is "OK", it sends a notification (T15) along with the first response information (T14). Then, the processor 211 outputs the first display screen 10 showing the received notification to the display via the output interface 214 (S19). At this time, the processor 211 outputs the first display screen 10 showing the received first response information and notification to the display via the output interface 214. Although S18 and S19 are shown separately in Figures 5A and 5B, S18 and S19 may be performed simultaneously.

[0072] Here, Figure 9 shows an example of the first display screen 10 output by the terminal device 200 in an embodiment of the present disclosure. Specifically, Figure 9 shows an example of the first display screen 10 output at S19 in Figure 5B. Specifically, in this example, the first display screen 10 differs from Figure 8 in that the first response information 32 displays "This is a recently released tea. We recommend product A, which was released in October 2023," and the notification 34 displays "The latest information is R." Other configurations may be the same as in Figure 8. When the area of ​​the notification 34 that displays R (hereinafter referred to as the R area) is selected by the user's operation, the processor 211 of the terminal device 200 generates the second display screen 40, which will be described later. The notification 34 may also be displayed on a banner.

[0073] Returning to Figure 5B, when the first response information 32 and notification 34 illustrated in Figure 9 are output, the processor 211 of the terminal device 200 selects the R region of the notification 34 based on the user's operation input received via the input interface 213 (S20). Also in S20, in order to generate second input information, the processor 211 of the terminal device 200 generates notification selection information on the first display screen 10 indicating that the R region of the notification 34 has been selected based on the user's operation input.

[0074] When the processor 211 of the terminal device 200 acquires notification selection information, it generates a second display screen 40 for receiving user input and generating second input information in order to input it into a second trained model configured to output second response information. The processor 211 then outputs the generated second display screen 40 to the display via the output interface 214 (S21). The processor 111 may also transmit information for displaying the second display screen 40 (second display screen information) via the communication interface 113 when it receives notification selection information, and the processor 211 of the terminal device 200 may receive this second display screen information via the communication interface 215.

[0075] Here, Figure 10 shows an example of a second display screen 40 output by the terminal device 200 in an embodiment of the present disclosure. Specifically, Figure 10 shows an example of a second display screen 40 output in S21 of Figure 5B. According to Figure 10, the second display screen 40 includes a display area 54, a text input area 42, a file addition area 44, a voice input button 46, and a send button 48. The text input area 42, file addition area 44, voice input button 46, and send button 48 in Figure 10 correspond to the text input area 18, file addition area 16, voice input button 22, and send button 24 in Figure 7, respectively, and their explanation is omitted. The second display screen 40 may include a second trained model selection area corresponding to the first trained model selection area 12 in Figure 7. The user may be able to change the second trained model by selecting the second trained model selection area.

[0076] The display area 54 outputs the second input information, the second response information, etc. As will be described later, the second input information and the second response information may be output in chat format in the display area 54. That is, in the display area 54, the second input information may be output to the right of the second display screen 40, and the second response information may be output to the left of the second display screen 40. In this example, since no second input information has been entered, nothing is displayed in the display area 54.

[0077] Returning to Figure 5B, when the second display screen 40 illustrated in Figure 10 is output, the processor 211 of the terminal device 200 generates second input information based on the user's operation input received via the input interface 213 (S22).

[0078] Here, Figure 11 shows an example of a second display screen 40 output by the terminal device 200 in an embodiment of the present disclosure. Specifically, Figure 11 shows an example of a second display screen 40 output in S22 of Figure 5B. In this example, the second display screen 40 differs from Figure 10 in that the second input information 52 displays "Please recommend a recently released tea." Other than the display of "Please recommend a recently released tea.", the configuration may be the same as in Figure 10. The second input information 52 is generated, for example, when the user inputs "Please recommend a recently released tea." into the text input area 42 and selects the send button 48. Note that the first input information "Please recommend a recently released tea." may be input into the text input area 42 at stage S21, in which case the second input information 52 may be generated simply by the user selecting the send button 48. Alternatively, the second input information may be generated with the same content as the first input information at the stage when the second display screen 40 is output.

[0079] Returning to Figure 5B, once the second input information is generated, the processor 211 of the terminal device 200 sends a second response generation request, including the second input information (T16), to the server device 100 via the communication interface 215. In this example, the processor 211 of the terminal device 200 generates the second input information, but the processor 111 of the server device 100 may also generate the second input information. When the processor 111 of the server device 100 generates the second input information, the processor 111 of the server device 100 generates the second input information based on information regarding user operations and selections in the text input area 42, file addition area 44, voice input button 46, etc.

[0080] When the processor 111 of the server device 100 receives a request to generate a second answer from the terminal device 200 via the communication interface 113, including second input information, the processor 111 of the server device 100 executes the second answer generation process by inputting the second input information generated for the second trained model selected based on at least one of the second model specification information in the user management table and the condition information in the model management table (S23). In this example, the second model specification information identifies a trained model (e.g., "Gemini") for which the information acquisition period used for training is recent. In this example, since the second model specification information identifies a trained model for which the information acquisition period used for training is recent, the second answer generation process shown in Figure 6B can be executed by inputting the input information generated for the second trained model selected based on the second model specification information in the user management table. Through this second answer generation process, the processor 111 obtains the second answer information for the second input information from the second trained model. Therefore, the user can obtain second answer information that is different from the first answer information. For example, if the second model designation information identifies a newly trained model with a recent acquisition period for the information used in training, the second response generation process allows the user to obtain response information that includes relatively recent data.

[0081] In this example, the second model specification information may identify a trained model (e.g., "Gemini") that uses web search processing (e.g., grounding). In this example, since the second model specification information identifies a trained model that uses grounding via web search, the second response generation process shown in Figure 6B can be executed by inputting the generated input information into the second trained model selected based on the second model specification information in the user management table.

[0082] Grounding refers to the process of connecting a model to verifiable information sources. Grounding can help improve the accuracy, reliability, and usefulness of the output of a trained model. Grounding may be available for specific trained models such as "Gemini". By using grounding in the second trained model, the response information output by the second trained model will include relatively recent information. Therefore, because the second model specification information identifies a trained model that uses grounding via web search, the user can obtain response information containing relatively recent information through the second response generation process.

[0083] When the processor 111 of the server device 100 obtains the second response information, it obtains advertising information relating to advertisements for products or services related to at least one of the second input information and the second response information, based on at least one of the second input information and the second response information (S24).

[0084] Here, Figure 6C is a diagram showing an example of a processing flow executed in the server device 100 according to the embodiment of this disclosure. Specifically, Figure 6C is a diagram showing an example of a processing flow executed in S24 of Figure 5B. In S24, first the processor 111 accesses the advertising server device (S131). Next, the processor 111 searches for advertising information stored in the advertising server device that matches predetermined conditions based on the second input information or the second response information (S132). Matching predetermined conditions may mean, for example, that the information contained in the second input information or the second response information partially matches the information contained in the advertising information. For example, if the second input information or the second response information contains information about "tea", the information about "tea" may be extracted as advertising information. For example, if the second input information or the second response information contains information about "tea", the information about products belonging to the same category as "tea" (including, for example, drinking water, coffee, soft drinks, etc.) may be extracted as advertising information. The processor 111 may also search for advertising information stored in the advertising server device that matches predetermined conditions, based on information regarding the history of the second input information or information regarding the history of the second response information. The processor 111 then obtains advertising information that matches the conditions from the advertising server device (S133).

[0085] When the processor 111 acquires advertising information, it transmits the acquired second response information (T17) and advertising information (T18) to the terminal device 200 that sent the response generation request via the communication interface 113.

[0086] Then, the processor 211 outputs the second display screen 40, which displays the received second response information and advertising information, to the display via the output interface 214 (S25, S26).

[0087] Here, Figure 12 shows an example of a second display screen 40 output by the terminal device 200 in an embodiment of the present disclosure. Specifically, Figure 12 shows an example of a second display screen 40 output in S25 (and S26) of Figure 5B. In this example, the second display screen 40 differs from Figure 11 in that the second input information 52 and the second response information 56 are displayed in the display area 54. Other configurations may be the same as in Figure 11.

[0088] In this example, the second display screen 40 shows the second response information 56 as, "This is a recently released tea. We recommend product B, which was released in December 2024. We also recommend products C and D." In this example, the second response information 56 includes information about multiple products (product B, product C, and product D). In other words, the second trained model may output second response information that includes information about multiple products when second input information is input. Alternatively, the second trained model may obtain output information from multiple trained models and then ensemble process this output information to obtain response information, thereby outputting second response information that includes information about multiple products. In this example, since second input information about products is input to the second trained model, second response information about multiple products is output. However, for example, if second input information about services is input to the second trained model, second response information about multiple services may be output. Since the second response information 56 includes information about multiple products or services, the user can obtain response information about various products or services. The second input information provided to the second trained model may be modified by the processor 111 so that the second response information includes information about multiple products or services. For example, the processor 111 may change the second input information from "Please recommend some recently released teas." to "Please recommend several recently released teas."

[0089] In this example, the second display screen 40 displays various information about products B, C, and D as second response information 56. This information includes product images, manufacturers, release dates, and features. In this example, the second display screen 40 displays various information about products B, C, and D in a table format for comparison as second response information 56. Since various information is displayed on the second display screen 40 as second response information 56, the user can compare various products or services. The second input information input to the second trained model may be modified by the processor 111 so that the second response information includes various information. For example, the processor 111 may change the second input information from "Please recommend some recently released teas." to "Please recommend several recently released teas. And please display them in a table format comparing their product images, manufacturers, release dates, and features."

[0090] Figure 13 is a diagram showing an example of a second display screen 40 output by the terminal device 200 in an embodiment of the present disclosure. Specifically, Figure 13 is a diagram showing an example of a second display screen 40 that is displayed after the second response information 56 is output. The second display screen 40 in Figure 13 may be displayed by scrolling the second display screen 40 in Figure 12 by user operation.

[0091] In this example, the advertising information 64 includes information about nutritional drinks. For example, if the second input information includes information about tea, the processor 111 of the server device 100 obtains advertising information from the advertising server that includes products related to the second input information (e.g., beverages, etc.). Because advertising information including products related to the second input information is obtained, it becomes easier to obtain advertising information that the user desires. Also, if the second response information includes information about tea, the processor 111 of the server device 100 obtains advertising information from the advertising server that includes products related to the second response information (e.g., beverages, etc.). The advertising information 64 may include not only information in text format, but also information in image and video format. The advertising information 64 may also be displayed on a banner.

[0092] Figure 14 shows an example of a second display screen 40 output by the terminal device 200 in an embodiment of the present disclosure. Specifically, Figure 14 shows an example of a second display screen 40 when, after the second response information 56 and advertising information 64 have been output, the second input information 58 has been further input by the user. In other words, after the processing in S26 of Figure 5B, the processing from S22 to S26 may be executed again. Note that the output of advertising information is omitted in Figure 14.

[0093] In this example, the second input information 58 is generated when the user enters "I want product B" into the text input area 18 and selects the send button 48. When the processor 111 of the server device 100 receives a second response generation request, including the second input information, from the terminal device 200 via the communication interface 113, the processor 111 of the server device 100 executes the second response generation process by inputting the generated second input information into the second trained model, which is selected based on at least one of the second model specification information in the user management table and the condition information in the model management table.

[0094] In this example, the second display screen 40 shows "Product B. It can be purchased at sales site H, sales site I, and sales site J" as the second answer information 62. In this example, the second answer information 62 includes information about multiple sales sites (sales site H, sales site I, and sales site J). In other words, the second trained model may output second answer information that includes information about multiple sales sites when second input information is input. The second trained model may also output second answer information based on information about the history of second output information or information about the history of second answer information.

[0095] In this example, the second display screen 40 displays various information about product B for each sales site as second response information 62. This information includes the price at each sales site, the shipping cost at each sales site, the delivery date at each sales site, and the link at each sales site. In this example, the second display screen 40 displays various information about product B for each sales site in a table format for comparison as second response information 62. Since the second display screen 40 displays various information for each sales site as second response information 62, the user can obtain response information about various sales sites. The second input information input to the second trained model may be modified by the processor 111 so that the second response information includes information about multiple sales sites. For example, the processor 111 may change the second input information from "I want product B" to "I want product B, please display the price, shipping cost, delivery date, and link in a table format for multiple sales sites."

[0096] Figure 15 shows an example of a second display screen 40 output by the terminal device 200 in an embodiment of the present disclosure. Specifically, Figure 15 shows an example of a second display screen 40 output in S26 of Figure 5B. In this example, the second display screen 40 differs from Figure 13 in that advertising information 68 is displayed instead of advertising information 64. Other configurations may be the same as in Figure 13.

[0097] The advertising information 68 may be generated as third response information by inputting information generated based on the second response information or second input information into a third trained model. This third trained model may be different from or the same as the first trained model or second trained model. The information input to the trained model that outputs advertising information (third input information) may be different from the second response information or second input information. For example, if the second input information is "Please tell me your recommendation for a recently released tea," the information input to the trained model that outputs advertising information may be, as an example, "Please output an advertisement (recommendation) recommending the most recommended product (in the case of Figure 15, tea B)." In other words, the processor 111 performs an advertising information generation process (Figure 6D) to obtain advertising information 68 from the third trained model by inputting the acquired (generated) third input information into the third trained model. Since advertising information recommending products is displayed along with the response information, the user's desire to purchase can be increased.

[0098] Figure 5C is a diagram showing an example of a processing sequence executed in the processing system 1 according to the embodiment of this disclosure. Specifically, Figure 5C is a diagram showing a series of processing flows related to the output of the second response information. Figure 5C is also a diagram showing a series of processing flows when the output of the notification is determined to be "acceptable". The processing up to S24 in Figure 5C may be the same as the processing up to S24 in Figure 5B. In this example, the processing shown in Figure 6C is executed at S24.

[0099] The processor 111 acquires corrected response information based on the advertising information and the second response information (S27). Here, Figure 6E is a diagram showing an example of the process for acquiring corrected response information executed in the server device 100 according to the embodiment of this disclosure. The processor 111 may change the second response information to corrected response information based on the advertising information acquired from the advertising server. For example, if the advertising information acquired from the advertising server includes product D, the corrected response information may be changed to recommend product D preferentially. For example, the corrected response information may be changed to change the order of recommendations based on predetermined numerical values ​​(recommended values) for each product included in the advertising information acquired from the advertising server. For example, if the advertising information acquired from the advertising server includes products from a specific manufacturer (sponsor, etc.), the corrected response information may be changed to recommend that product preferentially. By changing the second response information to corrected response information based on the advertising information, the user's willingness to purchase can be further increased.

[0100] Then, the processor 211 outputs a second display screen 40 showing the received corrected response information to the display via the output interface 214 (S28).

[0101] Figure 16 shows an example of a second display screen 40 output by the terminal device 200 in an embodiment of the present disclosure. Specifically, Figure 16 shows an example of a second display screen 40 output in S28 of Figure 5C. In this example, the second display screen 40 differs from Figure 12 in that corrected response information 66 is displayed instead of second response information 56. Other configurations may be the same as in Figure 12.

[0102] 6. The processing flow diagram 6A executed by the server device 100 is a diagram showing an example of a processing flow executed in the server device 100 according to the embodiment of this disclosure. Specifically, it is a diagram showing the processing flow of the first response generation process executed in S16 of Figure 5A. This processing flow is mainly performed by the processor 111 of the server device 100 reading and executing a program stored in the memory 112.

[0103] According to Figure 6A, the processor 111 starts the processing flow when it receives the first input information, as explained in S16 of Figure 5A (S111).

[0104] First, the processor 111 refers to the user management table based on the user ID information. Then, the processor 111 reads the first model designation information associated with the user ID information (S112). The processor 111 determines whether or not model information specifying one or more trained models is set as the first model designation information (S113). If the first model designation information is set, the processor 111 proceeds to the process in S115 without executing the process in S114.

[0105] On the other hand, if, for example, the first model specification information reads that any pre-trained model is acceptable, but none of the first model specification information is set due to a setting error, the processor 111 refers to the condition information in the model management table. Then, the processor 111 refers to the user management table based on the user ID information and reads the attribute information associated with the user ID information. The processor 111 compares the condition information and attribute information set as conditions for the pre-trained models identified by each model ID information to identify the available pre-trained models. For example, if the condition information for each pre-trained model is set as "the user is 18 years of age or older", the processor 111 refers to the user's attribute information to determine whether the condition is met. Then, the processor 111 selects at least one pre-trained model that satisfies all the conditions (S114). If there are multiple pre-trained models that satisfy the conditions, it is also possible to further narrow down the pre-trained models to be selected by using other attribute information, for example, by selecting a pre-trained model that has been used frequently based on the user's past usage history. For example, if the user's attribute information includes a specific professional occupation such as doctor, lawyer, patent attorney, or administrative scrivener, the processor 111 may select a pre-trained model that uses grounding.

[0106] The processor 111 sets at least one trained model selected by either the model specification information in S112 or the condition information in S114 as the first trained model that takes the first input information received in S111 as input (S115).

[0107] The processor 111 inputs the first input information received in S111 to the configured first trained model (S116). Specifically, as shown in Figure 4, if the configured first trained model is stored in a processing unit connected via a communication network, the processor 111 transmits the first input information to the processing unit via the communication interface 113. Also, if the configured first trained model is stored in memory 112 or the like, the processor 111 inputs the first input information to the first trained model.

[0108] Then, the processor 111 executes the first trained model to which the first input information has been input, and obtains the first response information from the first trained model (S117). If multiple trained models are set in S117, the processor 111 may obtain response information from each of the multiple trained models and transmit it directly to the terminal device 200. Alternatively, the processor 111 may obtain output information from each of the multiple trained models, perform ensemble processing on this output information to obtain the response information, and transmit it directly to the terminal device 200.

[0109] The processing flow is now terminated.

[0110] Figure 6B is a diagram showing an example of a processing flow executed in the server device 100 according to the embodiment of this disclosure. Specifically, it is a diagram showing the processing flow of the second answer generation process executed in S23 of Figure 5B. This processing flow is mainly performed by the processor 111 of the server device 100 reading and executing a program stored in the memory 112. This processing flow is the same as that in Figure 6A, so its explanation is omitted.

[0111] Figure 6D is a diagram showing an example of a processing flow executed in the server device 100 according to the embodiment of this disclosure. Specifically, it is a diagram showing the processing flow of the advertising information generation process executed in S24 of Figure 5B. This processing flow is mainly performed by the processor 111 of the server device 100 reading and executing a program stored in the memory 112. This processing flow is the same as that in Figure 6A, so its explanation is omitted.

[0112] In this way, by selecting a pre-trained model to input information based on at least one of the model specification information and condition information, users can obtain response information more efficiently.

[0113] In this embodiment, we can provide a processing device, a processing program, a processing method, and a processing system capable of generating advertising information relating to advertisements for products or services associated with at least one of the input information and the response information.

[0114] 7. Modifications (A) Input and Output Information Figures 1 to 16 mainly describe the case where first, second, and third input information are generated in text format, and first, second, and third response information are obtained in text format based on the first, second, and third input information. However, as described above, the first, second, and third input information can be in various formats other than text, such as image (including both video and still images), audio, document, PDF, link, or a combination thereof. Similarly, the first, second, and third response information can be obtained in various formats other than text, such as image (including both video and still images), audio, document, PDF, link, or a combination thereof.

[0115] For example, processor 111 generates first, second, and third input information, each containing the message, "Please obtain the latest image of product K." Then, processor 111 obtains the image of product K by inputting this first, second, and third input information into the first, second, and third trained models, respectively.

[0116] Another example is when processor 111 generates first, second, and third input information containing the message, "Please retrieve the latest video of product L." Then, processor 111 retrieves the video of product L by inputting this first, second, and third input information into the first, second, and third trained models, respectively.

[0117] (B) Regarding the processing unit, Figures 1 to 16 describe the case where the server device 100 functions as the processing unit. However, it is also possible for the terminal device 200 to function as the processing unit instead of the server device 100, or in combination with the server device 100. For example, at least one of the processes S13, S16, S17, S23, S24, and S27 in Figures 5A, 5B, and 5C is performed by the processor 211 of the terminal device 200 executing a program stored in the memory 212. In addition, the processing related to the execution of each trained model may be performed by other processing units connected via a communication network.

[0118] (C) Regarding the response generation process, Figures 1 to 16 describe an example in which a first response generation process and a second response generation process are performed, and the response generation process is performed twice. However, the response generation process may be performed only once. For example, advertising information 64 may be generated during the first response generation process. Figure 17 is a diagram showing an example of a first display screen 10 output by the terminal device 200 in an embodiment of this disclosure. In this example, the first display screen 10 differs from Figure 9 in that advertising information 64 is displayed instead of the notification 34. Other configurations may be the same as in Figure 9. The advertising information 64 in Figure 17 may be obtained from an advertising server. The advertising information 64 in Figure 17 may be obtained from a trained model. In Figure 17, the first display screen 10 does not display product information in tabular format as the first response information 32. In Figure 17, the first display screen 10 may display various product information in tabular format as the first response information 32.

[0119] Alternatively, the first response generation process may be omitted, and the second response generation process may be performed instead. In other words, the process may start from S21 in Figures 5B and 5C.

[0120] (D) Regarding the trained models, Figures 1 to 16 mainly describe an example where the first trained model is a model that does not use grounding and the second trained model is a model that uses grounding, but the examples of the first and second trained models are not limited to these. For example, the first trained model may be a model that uses grounding. Any other model disclosed in this specification can be used as the first and second trained models.

[0121] 8. Addendum The configurations of each embodiment described above are illustrated below. (1) A processing apparatus comprising at least one processor, wherein the at least one processor is configured to acquire first input information generated based on user input for input to a first trained model configured to output first response information, acquire the first response information from the first trained model by inputting the acquired first input information to the first trained model, and acquire advertising information relating to an advertisement for a product or service related to at least one of the first input information and the first response information.

[0122] (2) The processing apparatus according to (1) above, wherein at least one processor is configured to output second answer information, to acquire second input information generated based on user input for input to a second trained model different from the first trained model, to input the acquired second input information to the second trained model to acquire the second answer information from the second trained model, and to perform a process for determining whether to output a notification suggesting to acquire first answer information different from the second answer information from the first trained model based on at least one of the second input information, the second answer information, and the user's attribute information.

[0123] (3) The processing apparatus according to (2) above, wherein at least one processor is configured to perform a process for obtaining the first input information when the notification is selected based on the user's operation input, and inputting the obtained first input information into the first trained model, thereby obtaining the first response information from the first trained model into which the first input information has been input.

[0124] (4) The processing apparatus according to any one of (1) to (3) above, wherein at least one processor is configured to perform processing for obtaining the advertising information from the advertising server device.

[0125] (5) The processing apparatus according to (4) above, wherein at least one processor is configured to perform processing for obtaining corrected response information generated based on the advertising information and the first response information.

[0126] (6) The processing apparatus according to any one of (1) to (5) above, wherein at least one processor is configured to perform a process for obtaining third input information generated based on at least one of the first input information and the first response information in order to input it into a third trained model configured to output the advertising information, and inputting the obtained third input information into the third trained model to obtain the advertising information from the third trained model.

[0127] (7) The first trained model is a processing device according to any one of (1) to (6) above, which uses a web search process.

[0128] (8) The processing device according to any one of (1) to (7) above, wherein the first input information includes any of the following: text format, image format, audio format, document format, PDF format and link format.

[0129] (9) The processing apparatus described in any of (1) to (8) above, wherein the first response information includes information relating to multiple products or services.

[0130] (10) The processing device described in any of (1) to (9) above, wherein the first response information includes information regarding the price and delivery date of the product or service for each sales site.

[0131] (11) The processing device according to any one of (1) to (10) above, wherein the first input information includes information relating to the history of the user's first input information.

[0132] (12) The processing device according to any one of (1) to (11) above, wherein the first response information includes information relating to the history of the user's first response information.

[0133] (13) A processing program that enables a computer having at least one processor to perform the following processes: to acquire first input information generated based on user input for input to a first trained model configured to output first response information; to acquire the first response information from the first trained model by inputting the acquired first input information to the first trained model; and to acquire advertising information relating to an advertisement for a product or service related to at least one of the first input information and the first response information.

[0134] (14) A processing method performed by at least one processor in a computer having at least one processor, the processing method comprising: acquiring first input information generated based on user input for input to a first trained model configured to output first response information; acquiring first response information from the first trained model by inputting the acquired first input information into the first trained model; and acquiring advertising information relating to an advertisement for a product or service related to at least one of the first input information and the first response information, based on at least one of the first input information and the first response information.

[0135] (15) A processing system comprising: a processing device described in any of (1) to (12) above; and a terminal device connected to the processing device via a communication network and configured to transmit the first input information to the processing device via the communication network.

[0136] The embodiments and variations of this disclosure are presented as examples only and are not intended to limit the scope of this disclosure. These embodiments and variations can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of this disclosure. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents.

[0137] 1 Processing system 100 Server device 200 Terminal device

Claims

1. A processing device comprising at least one processor, wherein the at least one processor is configured to perform the following processes: acquire first input information generated based on user input for input to a first trained model configured to output first response information; acquire first response information from the first trained model by inputting the acquired first input information to the first trained model; and acquire advertising information relating to an advertisement for a product or service related to at least one of the first input information and the first response information.

2. The apparatus according to claim 1, wherein at least one processor is configured to output second answer information, to acquire second input information generated based on user input for input to a second trained model different from the first trained model, to input the acquired second input information to the second trained model to acquire the second answer information from the second trained model, and to perform a process for determining whether to output a notification suggesting to acquire first answer information different from the second answer information from the first trained model, based on at least one of the second input information, the second answer information, and the user's attribute information.

3. The apparatus according to claim 2, wherein at least one processor is configured to perform a process for obtaining the first input information when the notification is selected based on the user's operation input, and inputting the obtained first input information into the first trained model, thereby obtaining the first response information from the first trained model into which the first input information has been input.

4. The processing apparatus according to claim 1, wherein at least one processor is configured to perform processing for obtaining the advertising information from the advertising server device.

5. The apparatus according to claim 4, wherein the at least one processor is configured to perform processing for obtaining modified response information generated based on the advertising information and the first response information.

6. The apparatus according to claim 1, wherein the at least one processor is configured to perform a process for obtaining third input information generated based on at least one of the first input information and the first response information, for input to a third trained model configured to output the advertising information, and for inputting the obtained third input information to the third trained model to obtain the advertising information from the third trained model.

7. The processing apparatus according to claim 1, wherein the first trained model uses a search process via web search.

8. The processing apparatus according to claim 1, wherein the first input information includes any of the following: text format, image format, audio format, document format, PDF format, and link format.

9. The processing apparatus according to claim 1, wherein the first response information includes information relating to multiple products or services.

10. The processing apparatus according to claim 1, wherein the first response information includes information regarding the price and delivery date of products or services for each sales site.

11. The processing apparatus according to claim 1, wherein the first input information includes information relating to the history of the user's first input information.

12. The processing apparatus according to claim 1, wherein the first response information includes information relating to the history of the user's first response information.

13. A processing program that causes a computer having at least one processor to function to perform the following processes: to acquire first input information generated based on user input for input to a first trained model configured to output first response information; to acquire the first response information from the first trained model by inputting the acquired first input information to the first trained model; and to acquire advertising information relating to an advertisement for a product or service related to at least one of the first input information and the first response information.

14. A processing method performed by at least one processor in a computer having at least one processor, the processing method comprising: acquiring first input information generated based on user input for input to a first trained model configured to output first response information; acquiring first response information from the first trained model by inputting the acquired first input information into the first trained model; and acquiring advertising information relating to an advertisement for a product or service related to at least one of the first input information and the first response information, based on at least one of the first input information and the first response information.

15. A processing system comprising: a processing device according to claim 1; and a terminal device connected to the processing device via a communication network and configured to transmit the first input information to the processing device via the communication network.

Images