Information processing device and method

Abstract

We provide technology that dynamically generates prompts to be input into AI in response to changes in the situation. [Solution] An information processing device according to one embodiment includes a detection means for detecting the occurrence of a specified event based on real-time data obtained from a plurality of stationary devices installed in a target space, an access means for accessing a database that records past data collected in the target space, an input means for inputting a prompt including the real-time data and the past data into an artificial intelligence when the occurrence of the event is detected, and an output means for outputting information regarding the processing of a question or task based on the response to the prompt obtained from the artificial intelligence to an output device corresponding to the real-time data.

Description

[Technical Field]

[0001] The present invention relates to a technique for determining whether or not a prompt needs to be generated. [Background technology]

[0002] There is known a technology for generating and outputting prompts to an AI (Artificial Intelligence) system to encourage the system to perform a task. For example, Patent Document 1 discloses an invention that generates prompts to encourage the system to perform a task corresponding to the user's state of mind based on data acquired from multiple sensors. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-159107 Summary of the Invention [Problem to be solved by the invention]

[0004] When prompts to be input into an AI system are generated statically, the prompt generation process relies on predefined, fixed rules, which makes it difficult to fully optimize prompts according to the user's requests or objectives, making the system impractical.

[0005] In response to this, the present invention provides a technique for dynamically generating prompts to be input to AI (Artificial Intelligence) in response to changes in the situation. [Means for solving the problem]

[0006] An information processing device according to one aspect of the present disclosure includes a detection means for detecting the occurrence of a predetermined event based on real-time data obtained from a plurality of stationary devices installed in a target space, an access means for accessing a database that records past data previously collected in the target space, an input means for inputting a prompt including the real-time data and the past data into an artificial intelligence when the occurrence of the event is detected, and an output means for outputting information regarding the processing of a question or task based on the answer to the prompt obtained from the artificial intelligence to an output device corresponding to the real-time data.

[0007] A method according to another aspect of the present disclosure includes the steps of detecting the occurrence of a predetermined event based on real-time data obtained from a plurality of stationary devices installed in a target space, accessing a database that records past data previously collected in the target space, inputting a prompt including the real-time data and the past data into an artificial intelligence when the occurrence of the event is detected, and outputting information regarding the processing of a question or task based on the answer to the prompt obtained from the artificial intelligence to an output device corresponding to the real-time data. [Effects of the Invention]

[0008] According to the present invention, prompts to be input to an AI can be dynamically generated in response to changes in the situation. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an example of a system configuration of an information processing system according to an embodiment. [Figure 2] FIG. 1 is a diagram illustrating an example of the functional configuration of an information processing apparatus according to an embodiment. [Figure 3] FIG. 1 is a diagram illustrating an example of a hardware configuration of an information processing apparatus according to an embodiment. [Figure 4] FIG. 1 is a diagram illustrating a framework of an information processing system according to an embodiment. [Figure 5]FIG. 2 is a sequence diagram illustrating an example of an operation outline of the information processing system according to the embodiment. [Figure 6] FIG. 1 is a diagram illustrating an example of a system configuration of a smart home. [Figure 7] FIG. 4 is a diagram showing an example of a user database. [Figure 8] FIG. 4 is a diagram showing an example of an event database. [Figure 9] FIG. 1 shows an example of output from a smart home (1). [Figure 10] FIG. 4 is a diagram showing an example of an event database. [Figure 11] FIG. 10 is a diagram showing an example of output from a smart home (2). [Figure 12] FIG. 1 is a diagram illustrating an example of the system configuration of a smart store. [Figure 13] FIG. 4 is a diagram showing an example of an event database. [Figure 14] A diagram showing an example of the output of the Smart Store. [Figure 15] FIG. 1 is a diagram illustrating an example of the system configuration of a smart farm. [Figure 16] FIG. 4 is a diagram showing an example of an event database. [Figure 17] A diagram showing an example of smart farm output. DETAILED DESCRIPTION OF THE INVENTION

[0010] 1. Configuration FIG. 1 is a diagram illustrating an example of the system configuration of an information processing system 1 according to an embodiment. In this example, the information processing system 1 is a system for providing information corresponding to a user's situation in a target space. The target space is a limited space, such as a house, a store, or a farm. This information is based on a response generated by an AI. This information is further based on data accumulated in the past (hereinafter referred to as "past data") D1 and data acquired in real time (hereinafter referred to as "real-time data") D2.

[0011] The information processing system 1 includes an information processing device 100, a stationary device 200, a history database 300, an AI 400, and an output device 500. In this example, the components of the system are connected to each other via a network 2. The network 2 is a computer network such as the Internet.

[0012] The information processing device 100 is a device that performs information processing in the information processing system 1. The information processing device 100 is connected to, for example, a stationary device 200, a history database 300, and an AI 400 via a network 2. The information processing device 100 is a device that provides a type of RAG (Retrieval-Augmented Generation). For example, the information processing device 100 cooperates with the AI 400 to provide notifications corresponding to the user's situation based on various types of data.

[0013] The stationary device 200 is a device that acquires various types of information. The stationary device 200 is, for example, at least one of a photographing device such as a camera, an access control device such as an electronic lock, a thermometer, a hygrometer, a pH measuring device, and a sensor device such as a GPS (Global Positioning System) receiver. The stationary device 200 may include multiple types of devices that output different types of information.

[0014] The past database 300 is a database that stores various types of information and is constructed in a server device. For example, information about users (hereinafter referred to as "user information") and information about events that occurred in the past (hereinafter referred to as event information) are recorded in the past database 300. The information stored in the past database 300 will be described later.

[0015] The AI 400 is a type of so-called generative AI and has a large language model (LLM). The AI 400 has functions such as language processing capabilities using natural language, analytical capabilities, and a huge database. The AI 400 generates information related to notifications corresponding to the user's situation based on input from the information processing device 100. Specific processing in the AI 400 will be described later.

[0016] The output device 500 is a device for notifying a user. The output device 500 is, for example, a playback device such as a speaker, a display device such as a monitor or digital signage, commercial equipment such as a sprinkler or air conditioner, a playback device and / or a display device such as a smartphone or tablet.

[0017] 2 is a diagram illustrating a functional configuration of an information processing device 100 according to an embodiment. In this embodiment, the information processing device 100 includes a communication unit 101, a detection unit 102, an access unit 103, a determination unit 104, a generation unit 105, an input unit 106, an acquisition unit 107, an output unit 108, and a storage unit 109.

[0018] The communication means 101 communicates with the stationary device 200, the history database 300, and other devices such as the AI 400. The detection means 102 detects the occurrence of a predetermined event based on real-time data D2 obtained from the stationary device 200. Real-time data is data that can be measured and used immediately. Specifically, if the stationary device 200 is a camera, the real-time data D2 is an image or video currently being captured by the camera.

[0019] The access means 103 accesses a database that records past data D1 collected in the target space in the past. The access means 103 accesses the past database 300, for example, to refer to or acquire various pieces of information.

[0020] The determination means 104 uses the AI 400 based on the real-time data D2 to determine whether now is the timing to output a question or task processing to the output device 500 (hereinafter referred to as "dynamic timing"). A question is a sentence used by the information processing system 1 to have a dialogue with a user via voice or text (e.g., collecting information or processing task instructions), particularly a sentence used when first speaking to the user. The determination means 104 determines whether now is dynamic timing based on, for example, a condition regarding a combination of different types of information. The different types of information are, for example, multiple pieces of real-time data D2 acquired from multiple stationary devices 200 or a combination of past data D1 and real-time data D2. The determination means 104 makes this determination using, for example, a machine learning model, and trains this machine learning model using events accumulated in a database as training data. The training data is information regarding the user's state and / or behavior corresponding to events that occurred in the past. Specific processing by the determination means 104 will be described later.

[0021] The generating means 105 generates a prompt P when it is determined that the present is dynamic timing. When the occurrence of an event is detected and the present is determined to be dynamic timing, the input means 106 inputs the prompt P including the past data D1 and the real-time data D2 to the AI 400. For example, the input means 106 inputs the prompt P generated by the generating means 105 to the AI 400. The acquiring means 107 acquires an answer A to the prompt P from the AI 400.

[0022] The output means 108 outputs information relating to the processing of a question or task based on the answer A to the prompt P acquired from the AI 400 to the output device 500 corresponding to the real-time data D2. The output means 108 outputs information relating to the answer A to the output device 500, for example. The storage means 109 stores various data and programs.

[0023] 3 is a diagram showing the hardware configuration of the information processing system 1 according to the embodiment. The information processing device 100 is a computer including a processor 151, a communication IF (Interface) 152, a memory 153, and a storage 154. The processor 151 controls the operations of the other components of the information processing device 100 by executing a predetermined program stored in the storage 154. The communication IF 152 communicates wirelessly with a communication partner device.

[0024] The storage 154 stores a program (hereinafter referred to as a "server program") for causing the computer to function as the information processing device 100 of the information processing system 1. When the processor 151 is executing the server program, the communication IF 152 is an example of the communication means 101, the processor 151 is an example of the detection means 102, the access means 103, the determination means 104, the generation means 105, the input means 106, the acquisition means 107, and the output means 108, and the memory 153 is an example of the storage means 109.

[0025] 4 is a diagram illustrating an example of the framework of the information processing system 1 according to the embodiment. A prompt P is generated based on past data D1 and real-time data D2. The AI 400 generates an answer A based on the past data D1 and the input prompt P. The generated answer A is generated in a format according to the output device 500. The generated answer A is data such as text data A1, image data A2, audio data A3, and program data A4. Details of the generation of the prompt P and the answer A will be described later.

[0026] 2. Operation Fig. 5 is a sequence diagram illustrating an example of an outline of the operation of the information processing system 1 according to the embodiment. The process shown in Fig. 5 is initiated, for example, when the information processing system 1 is started. In the following, hardware such as the information processing device 100 is described as the subject of the process, which means that in this hardware, a hardware element such as a processor 151 executing a program such as a server program executes the process in cooperation with other hardware elements. The same applies to the following explanations of other sequence diagrams. Note that the specific process of each of steps S1 to S7 will be described in 2-1 onwards.

[0027] When the information processing system 1 is started, the information processing device 100 acquires real-time data D2 of the target space from the stationary device 200 (step S1). Next, the information processing device 100 detects the occurrence of a predetermined event based on the acquired real-time data D2 (step S2). Next, the information processing device 100 accesses the past database 300 that records past data D1 collected in the target space in the past (step S3).

[0028] Next, the information processing device 100 determines whether the present timing is dynamic timing (step S4). The information processing device 100 determines whether the present timing is dynamic timing based on, for example, a condition related to a combination of different types of information or a condition related to a combination of past data D1 and real-time data D2. Alternatively, the information processing device 100 determines whether the present timing is dynamic timing based on, for example, a condition related to a combination of past data D1 and real-time data D2. Note that, if past data D1 is not sufficiently stored in the initial stage of introduction of the information processing system 1, the determination of whether the present timing is dynamic timing is not performed. In other words, once a certain amount of past data D1 has been accumulated, the information processing device 100 determines whether the present timing is dynamic timing. The information processing system 1 repeatedly executes the processes of steps S1 to S4 at any timing.

[0029] Next, if it is determined that now is the time to output a question, the information processing device 100 generates a prompt P including the past data D1 and the real-time data D2 (step S5). Next, the information processing device 100 inputs the generated prompt P to the AI 400 (step S6). The AI 400 outputs an answer A to the prompt P. Next, the information processing device 100 obtains a question based on the answer A to the prompt P from the AI 400, and outputs it to the user corresponding to the real-time data D2 (step S7).

[0030] 2-1.Smart Home (1) A specific example of operation when the target space is a "home" will be described below. This home is a so-called smart home in which the information processing system 1 is installed and home appliances and household equipment are connected by communication technology. Note that the user in this example is a person who lives in the home in which the information processing device 100 is installed (hereinafter simply referred to as a "resident").

[0031] FIG. 6 is a diagram illustrating the system configuration of a smart home (1). In this example, the stationary device 200 includes an entry / exit control device that controls people entering and leaving the house. The entry / exit control device is, for example, a device such as an electronic lock, a camera, and a sensor. The electronic lock is, for example, installed at the front door. The camera is, for example, installed at the entrance or living room. In this example, the output device 500 also includes a playback device that suggests to a resident at home what to say to the resident who has returned home. The playback device is, for example, a device such as a speaker. The speaker is, for example, installed in the living room of the house. The speaker is, for example, installed in the living room.

[0032] In the following example, a case will be described in which the entry / exit control devices among the plurality of stationary devices 200 are an electronic lock and a camera, and the playback devices among the output devices 500 are speakers.

[0033] The following is a description based on the sequence diagram in Fig. 5. The information processing device 100 acquires, for example, real-time data D2, information indicating that the electronic lock has been unlocked and information indicating an image captured by the camera (step S1). Existing technologies such as wireless communication and biometric authentication are used for processing the electronic lock. When wireless communication is used for processing the electronic lock, for example, each user's smartphone with an application using NFC (so-called contactless IC) downloaded is used as the key for the electronic lock. Therefore, the owner of the key (smartphone) that unlocked the electronic lock can detect which resident has returned home. The image captured by the camera shows, for example, the resident who was at home and the resident who has returned home (their status).

[0034] Next, the information processing device 100 detects the occurrence of an event that a resident has returned home, for example, based on information indicating that the electronic lock has been unlocked (step S2). For example, when the electronic lock is unlocked, the information processing device 100 detects that the resident has returned home. Alternatively, the information processing device 100 may detect the occurrence of an event that a resident has returned home, for example, based on information indicating that the electronic lock has been unlocked and an image captured by a camera. For example, when a user is shown in an image captured at the time the electronic lock was unlocked (for example, within 10 seconds after the unlocking), the information processing device 100 may detect that the resident has returned home.

[0035] Next, the information processing device 100 accesses, for example, the history database 300 (step S3). The history database 300 includes, for example, history related to the house. The history database 300 also stores multiple databases. The history database 300 includes, for example, a user database 301 and an event database 302a.

[0036] FIG. 7 is a diagram showing an example of the user database 301. The user database 301 stores information about the attributes of each resident. The attributes include, for example, a user ID, a name, an image (photograph), a date of birth, a gender, a relationship (within the family), and information about a location. The user database 301 includes a plurality of records. Each record stores information about each user. The user database 301 includes, for example, a "user ID," a "name," a "photograph (photograph)," a "date of birth," a "gender," a "relationship (within the family)," and a "location." In the example of FIG. 8a, the top record of the user database 301 includes "user ID: U1," "name: Yamada Taro," "image: (photograph of an adult male)," "date of birth: 1981 / 01 / 01," "gender: male," "relationship (within the family): father," and "location: out." This shows that the resident with user ID U1 has the name Yamada Taro, a photo of his face, his date of birth January 1, 1981, his gender is male, his relationship to the family is father, and that he is currently out. Note that user database 301 may include other records, but to simplify the drawing, other descriptions are omitted here. The information stored in user database 301 (particularly name, image, date of birth, gender, and relationship) is, for example, input in advance by the user himself.

[0037] FIG. 8 is a diagram illustrating an example of the event database 302a. The event database 302a stores events that have occurred in the past. In this example, an event is, for example, a resident's behavior, such as a conversation, going out, or returning home. The event database 302a includes multiple records. In this example, each record stores information about a resident's past events. The event database 302a includes, for example, an "event ID," "event content," "date and time," "user ID," "status," "image," and "details." The event ID is a serial code assigned to each event. In this example, the event content is a resident's behavior that represents the content of each event. The status is the state of the object to be detected (in this example, the resident). The (resident's) status is, for example, having emotions such as joy, anger, sadness, fear, fatigue, or peace. The details are details about the event content. In this example, if the event content is "conversation," for example, the details of the conversation are stored.

[0038] In the example of FIG. 8, the top record of the event database 302a includes "Event ID: EV1," "Event Content: Return Home," "Date and Time: 202x / 04 / 01 18:00," "User ID: U3," "Status: Angry Expression," "Image: (Image of Angry Expression)," and "Details: -." This indicates that for the event with event ID EV1, a resident with user ID U3 (i.e., the eldest son) returned home with an angry expression at 18:00 on April 1, 202x. The second-to-top record of the event database 302 includes "Event ID: EV2," "Event Content: Conversation," "Date and Time: 202x / 04 / 01 18:01," "User ID: U2," "Status: Kind Expression, Question," "Image: (Image of Kind Expression)," and "Details: Welcome home. You seem angry, but are you okay?" This indicates that in the event with event ID EV2, a resident with user ID U2 (i.e., the mother) spoke with a gentle expression at 18:01 on April 1, 202x. Note that the event database 302a may include other records, but these are not described here to simplify the drawing. Information stored in the event database 302a is automatically input by the information processing device 100, for example.

[0039] Returning to FIG. 5, the explanation will be resumed. The information processing device 100 determines whether the current timing is dynamic or not (step S4). The conditions for determining whether the current timing is dynamic or not include a condition item that the entry / exit control device detects that a certain resident has returned home. Specifically, the condition item is that when there is at least one resident in the house and the electronic lock detects that one of the resident has returned home, the state of the resident who has returned home, as photographed by the camera, is in a predetermined state. The predetermined state is, for example, a negative expression such as an angry expression, a sad expression, or a dissatisfied expression.

[0040] As an example, assume that a resident (mother) with user ID U2 is at home and a resident (eldest son) with user ID U3 is detected to have returned home with an angry expression at 18:00 on May 1, 202x (event ID: EVx1). The information processing device 100 determines whether the timing is dynamic based on, for example, the state of the resident (eldest son) who has returned home, “angry expression,” shown in an image captured by a camera, and an event (event ID: EV1-2) associated with “eldest son: angry expression” in the event database 302a. In “EV2” in the event database 302, the resident (mother) with user ID U2 has a conversation with the resident (eldest son) with user ID U3 who has an angry expression other than “Welcome home.” Therefore, the information processing device 100 determines that the resident (eldest son) with user ID U3 who has returned home is in a predetermined state (i.e., it is determined that the current timing is dynamic). Note that if it is determined that the current timing is not dynamic, the information processing device 100 terminates the processing.

[0041] If it is determined that the timing is dynamic, the information processing device 100 generates a prompt P for suggesting something to say to the resident who has returned home (step S5). The information processing device 100 generates the prompt P based on, for example, the past data D1 and the real-time data D2.

[0042] FIG. 9 is a diagram showing an example of an output from the smart home (1). The prompt P is composed of, for example, first input data P1 generated based on real-time data D2, second input data P2 generated based on past data D1, and output instructions P3. The first input data is information about the current situation. For example, information about [DATE], [ENVIRONMENT], [PERSON], and [ATTRIBUTE] is input to the first input data P1. [ENVIRONMENT] is, for example, meteorological information such as weather, temperature, and humidity. In this example, [PERSON] is, for example, a detected resident. In this example, [ATTRIBUTE] is the resident's status. Note that the first input data P1 may be provided with other items [ ] (for example, [PLACE], [IMAGE], etc.), and information corresponding to those items may be input; however, for simplicity of the drawing, other descriptions are omitted here.

[0043] The second input data is information relating to past situations. For example, information relating to [INPUT] is input to the second input data P2. The output instruction is information for instructing the AI 400 to perform processing. For example, information relating to [INSTRUCT] is input to the output instruction P3.

[0044] The information processing device 100 generates the first input data P1 portion of the prompt P based on, for example, real-time data D2 acquired by an electronic lock and a camera. The information processing device 100, for example, inputs "date and time" information from the real-time data D2 into [DATE]. The information processing device 100 may, for example, acquire weather information to be input into [ENVIRONMENT] from the stationary device 200, or may acquire weather information for the current location of the information processing device 100 from an external source via the network 2. The information processing device 100, for example, refers to the user database 301 and inputs information corresponding to the "user ID" from the real-time data D2 into [PERSON]. The information processing device 100, for example, inputs "status" information from the real-time data D2 into [ATTRIBUTE].

[0045] [DATE] May 1, 202x 18:00 [ENVIRONMENT] Sunny (weather), 23℃ (temperature), 56% (humidity) [PERSON] Ichiro Yamada, eldest son [ATTRIBUTE] Angry expression ...

[0046] Furthermore, the information processing device 100 generates the second input data P2 portion by extracting, for example, the "date and time," "user ID," and "status" from the event database 302. Specifically, the following information is generated from the record for "EV2-3" in the event database 302. Note that the information input to [INPUT] may be generated from multiple records, but other descriptions are omitted here to simplify the drawing. [INPUT] April 1, 202x, 18:00, home, eldest son, angry expression April 1, 202x, 18:01, Mother, Kind expression, Welcome back. You seem angry, but is it okay? ...

[0047] Furthermore, the information processing device 100 generates the portion of the output instruction P3 based on, for example, a pre-stored fixed phrase, "The above event has occurred. How should we respond?" Note that the information input to the output instruction P3 is not limited to this.

[0048] [INSTRUCT] The above event has occurred. How should I respond?

[0049] Alternatively, the information processing device 100 may generate the output instruction P3 based on, for example, the historical data D1 and the real-time data D2. Specifically, the information processing device 100 stores in advance a fixed phrase such as "The above event has occurred. Please recommend 'yy' because of 'xx'." The information processing device 100 inputs, for example, the wording of [ATTRIBUTE] in the first input data P1 into the fixed phrase "xx," and inputs the wording of "event content" corresponding to the event content of [ATTRIBUTE] in the event database 302a into the fixed phrase "yy." In this example, [ATTRIBUTE] is "angry expression," and the "event content" corresponding to the event content "going home" of the "angry expression" (i.e., the next event) is "conversation." Therefore, the information processing device 100 generates the sentence "The above event has occurred. Please recommend 'conversation' because of 'angry expression.'"

[0050] The information processing device 100 inputs the generated prompt P to the AI 400 (step S6). The information processing device 100 outputs a question based on the answer A to the prompt P acquired from the AI 400 to the resident corresponding to the real-time data D2 (step S7). The AI 400 generates voice data A3 of the answer A, for example, "Ichiro looks a little irritated. If you say to him, 'Are you okay?' or 'Is something wrong?', he might calm down a little." based on the input prompt P. The information processing device 100 acquires, for example, the generated voice data A3. Thereafter, the information processing device 100 outputs the acquired answer A to the output device 500. The information processing device 100 plays the acquired voice data A3 from, for example, a speaker installed in the living room. Alternatively, the output device 500 may be, for example, a smartphone owned by a resident at home.

[0051] 2-2.Smart Home (2) Another specific example of operation when the target space is a "home" will be described below. This home is a so-called smart home in which the information processing system 1 is installed, similar to the smart home (1), and home appliances and household equipment are connected by communication technology. Note that the user in this example is a person (hereinafter simply referred to as a "resident") who lives in the home in which the information processing device 100 is installed.

[0052] The system configuration of the smart home is similar to that illustrated in FIG. 6. In this example, the multiple stationary devices 200 include an entry / exit control device that controls people entering and leaving the house, and an imaging device that captures images of people or objects in the house. The entry / exit control device is, for example, an electronic lock, a camera, a sensor, or other device. The imaging device is, for example, a camera, or other device. The camera is installed, for example, on a balcony, at the entrance, or in the living room. In this example, the output device 500 also includes a playback device that speaks words according to predetermined items to people in the house or who have returned home. The playback device is, for example, a speaker, or other device. The speaker is installed, for example, in the living room of the house.

[0053] In the following example, a case will be described in which, among the multiple stationary devices, the entry / exit control device is an electronic lock, the image capturing device is a camera, and the output device 500 is a speaker.

[0054] The following description will be given based on the sequence diagram of Fig. 5. The information processing device 100 acquires, for example, information indicating an image captured by a camera as real-time data D2 (step S1). The image captured by the camera shows, for example, whether laundry is hung out to dry and the weather.

[0055] Next, the information processing device 100 detects the occurrence of an event that laundry has been hung out or that laundry has been brought in, for example, based on an image captured by a camera (step S2). For example, if laundry is shown in an image captured by a camera, the information processing device 100 detects that the resident has hung out laundry. Alternatively, for example, if laundry that was shown in an image captured by a camera is no longer shown, the information processing device 100 detects that the resident has brought in laundry.

[0056] Next, the information processing device 100 accesses, for example, the past database 300 (step S3). The past database 300 includes, for example, the behavioral history of the residents living in the house. The past database 300 also stores multiple databases. The past database 300 includes, for example, a user database 301 and an event database 302b. The user database 301 is the same as the one exemplified in FIG. 7.

[0057] FIG. 10 is a diagram illustrating an example of the event database 302b. The event database 302b stores events that have occurred in the past. In this example, an event is, for example, a resident's action such as hanging out laundry and taking in laundry. The event database 302b includes multiple records. In this example, each record stores information about a past event of each resident. The event database 302b includes, for example, an "event ID," "event content," "date and time," "user ID," "status," "image," and "details." The event ID is a serial code assigned to each event. In this example, the event content is a resident's action that represents the content of each event. In this example, the status is the status of the detected object (in this example, laundry). The (laundry) status is, for example, whether there is laundry or not. The details are details about the event content. In this example, the details include, for example, the weather (for example, sunny, cloudy, and rainy) at the time the event occurred. The information about the weather may be acquired from the stationary device 200, for example, or information about the weather at the current location of the information processing device 100 may be acquired from outside via the network 2.

[0058] In the example of FIG. 10 , the top record of the event database 302b includes "Event ID: EV1," "Event Content: Hanging Out Laundry," "Date and Time: 202x / 04 / 01 10:00," "User ID: U2," "Status: Laundry Present," "Image: (Image of Laundry Hanging Out)," and "Details: Sunny." This indicates that for the event with event ID EV1, a resident with user ID U2 (i.e., the mother) hung out laundry at 10:00 on April 1, 202x, and the weather at that time was sunny. Furthermore, the second-to-top record of the event database 302b includes "Event ID: EV2," "Event Content: Bringing In Laundry," "Date and Time: 202x / 04 / 01 16:00," "User ID: U2," "Status: No Laundry," "Image: - (Image of nothing displayed)," and "Details: Sunny." This indicates that for an event with an event ID of EV2, a resident with a user ID of U2 (i.e., the mother) brought in laundry at 16:00 on April 1, 202x, and the weather was sunny at the time. Note that the event database 302b may include other records, but these are omitted here to simplify the drawing. Information stored in the event database 302b is, for example, automatically input by the information processing device 100.

[0059] Returning to FIG. 5, the explanation will be resumed. The information processing device 100 determines whether or not the present timing is a dynamic timing (step S4). For example, the conditions for the information processing device 100 to determine whether or not the present timing is a dynamic timing include a condition that a predetermined item is included in the past data D1. Specifically, the condition is that laundry is shown in an image captured by a camera when predetermined weather is detected. The predetermined weather is, for example, weather that is unfavorable for laundry, such as rain or strong wind.

[0060] As an example, assume that a resident (mother) with user ID U2 is at home, laundry is hanging out to dry, and it is detected that rain has started falling at 1:00 PM on May 1, 202x (event ID: EVx1). The information processing device 100 determines whether the timing is dynamic based on, for example, the laundry status "laundry present" shown in the image captured by the camera and events (event ID: EV6-7) associated with "event content: rainfall" and "status: laundry present" in the event database 302b. For example, in "EV7" in the event database 302b, the resident (mother) with user ID U2 immediately brought in her laundry after it started raining, so the information processing device 100 determines that the current timing is dynamic. Note that if it is determined that the current timing is not dynamic, the information processing device 100 ends the process.

[0061] If it is determined that the timing is dynamic, the information processing device 100 generates a prompt P for suggesting that a person at home say something according to a predetermined item (step S5). The information processing device 100 generates the prompt P based on, for example, the past data D1 and the real-time data D2.

[0062] FIG. 11 is a diagram showing an example of an output from a smart home (2). Similar to other examples, the prompt P is composed of, for example, first input data P1 generated based on real-time data D2, second input data P2 generated based on historical data D1, and output instructions P3 generated based on the historical data D1 and real-time data D2. The first input data is information about the current situation. For example, information about [DATE], [ENVIRONMENT], [PERSON], [ATTRIBUTE], [PLACE], [IMAGE], etc. is input to the first input data P1. [ENVIRONMENT] is, for example, meteorological information such as weather, temperature, and humidity. In this example, [PERSON] is, for example, a detected resident. In this example, [ATTRIBUTE] is, for example, the state of laundry. Note that the first input data P1 may have other items [ ] (for example, [PLACE], [IMAGE], etc.), and information corresponding to those items may be input, but to simplify the drawing, other descriptions are omitted here.

[0063] The second input data is information relating to past situations. For example, information relating to [INPUT] is input to the second input data P2. The output instruction is information for instructing the AI 400 to perform processing. For example, information relating to [INSTRUCT] is input to the output instruction P3.

[0064] The information processing device 100 generates the first input data P1 portion of the prompt P based on, for example, real-time data D2 acquired by a camera. The information processing device 100, for example, inputs "date and time" information of the real-time data D2 into [DATE]. The information processing device 100 may, for example, acquire weather information to be input into [ENVIRONMENT] from the stationary device 200, or may acquire weather information for the current location of the information processing device 100 from an external source via the network 2. The information processing device 100, for example, refers to the user database 301 and inputs information corresponding to the "user ID" of the real-time data D2 into [PERSON]. The information processing device 100, for example, inputs "status" information of the real-time data D2 into [ATTRIBUTE]. The information processing device 100, for example, inputs "status" information of the real-time data D2 into [ATTRIBUTE].

[0065] [DATE] May 1, 202x 13:00 [ENVIRONMENT] Rain (weather), 23℃ (temperature), 56% (humidity) [PERSON] Hanako Yamada, mother [ATTRIBUTE] Laundry available ...

[0066] Furthermore, the information processing device 100 extracts, for example, the "date and time," "user ID," and "status" from the event database 302b to generate the second input data P2. Specifically, the following information is generated from the record of "EV6-7" in the event database 302b: [INPUT] April 3, 202x, 12:00, Rain, Laundry April 3, 202x, 12:05, Laundry collection, Mother, No laundry ...

[0067] Furthermore, the information processing device 100 generates the portion of the output instruction P3 based on, for example, a pre-stored fixed phrase, "The above event has occurred. How should we respond?" Note that the information input to the output instruction P3 is not limited to this. [INSTRUCT] The above event has occurred. How should I respond?

[0068] The information processing device 100 inputs the generated prompt P to the AI 400 (step S6). The information processing device 100 outputs a question based on the answer A to the prompt P acquired from the AI 400 to the resident corresponding to the real-time data D2 (step S7). The AI 400 generates voice data A3 of the answer A, for example, "It's started to rain. The laundry is hanging outside, so it would be a good idea to bring it in early. Let's put it away before it gets wet," based on the input prompt P. The information processing device 100 acquires, for example, the generated voice data A3. Thereafter, the information processing device 100 outputs the acquired answer A to the output device 500. The information processing device 100 plays the acquired voice data A3 from a speaker installed in the living room, for example. Alternatively, the output device 500 may be a smartphone owned by the resident at home.

[0069] 2-3.Smart Store A specific example of operation when the target space is a "store" will be described below. This store is a so-called smart store in which the information processing system 1 has been installed and commercial electrical appliances and commercial facilities are connected by communication technology. Note that the user in this example is a store clerk (hereinafter simply referred to as "store clerk") working in the store in which the information processing device 100 is installed.

[0070] FIG. 12 is a diagram illustrating an example of the system configuration of a smart store. In this example, multiple stationary devices 200 include a photographing device that photographs customers in the store and a checkout device that stores information about products purchased by customers (a so-called POS (Point Of Sale) system). The photographing device is, for example, a device such as a camera. The camera is, for example, installed inside the store or near the entrance / exit of the store. The checkout device is, for example, a device such as a cash register. The cash register is, for example, installed at the checkout counter inside the store. In this example, the output device 500 also includes a display device that recommends specific products to customers. The display device is, for example, a device such as a monitor or digital signage. The monitor or digital signage is, for example, installed inside the store or near the entrance / exit of the store.

[0071] In the following example, a case will be described in which, among the multiple stationary devices 200, the image capturing device is a camera, the accounting device is a cash register, and the display device among the output devices 500 is a monitor.

[0072] The following description will be given based on the sequence diagram of Fig. 5. The information processing device 100 acquires, for example, information indicating an image captured by a camera as real-time data D2 (step S1). The image captured by the camera shows, for example, (the state of) a customer who has visited the store.

[0073] Next, the information processing device 100 detects the occurrence of an event, that is, a customer visiting a store, based on an image captured by a camera, for example (step S2).

[0074] Next, the information processing device 100 accesses, for example, the history database 300 (step S3). The history database 300 includes, for example, history related to the store. The history database 300 also stores a plurality of databases. The history database 300 includes, for example, an event database 302c.

[0075] FIG. 13 is a diagram illustrating an example of the event database 302c. The event database 302c stores events that have occurred in the past. In this example, an event is, for example, a customer's behavior such as purchasing or returning a product. The event database 302c includes multiple records. Each record stores information about a customer's past events. The event database 302c includes, for example, an "event ID," "event content," "date and time," "attributes (gender, age, body shape)," "image," and "details." The event ID is a serial code assigned to each event. In this example, the event content is customer behavior that represents the content of each event. In this example, the attributes are personality traits that represent the characteristics of the customer. Examples of the attributes include gender (male or female), age (pre-teen, teens, 20s, 30s, 40s, etc.), and body shape (thin, normal, or obese). The details are details about the event content. In this example, if the event content is "purchase," for example, the "details" stores the purchased product or type of product. In this example, it is assumed that the product to be sold (purchased) is rice balls, and the type of product to be stored in detail is rice ball fillings.

[0076] In the example of FIG. 13, the top record of the event database 302c includes "Event ID: EV1," "Event Content: Purchase," "Date and Time: 202x / 04 / 01 12:00," "Attributes: Gender: Male, Age: 30s, Body Type: Thin," "Image (Image of a Thin Person)," and "Details: Salmon." This indicates that for the event with event ID EV1, a thin male customer in his 30s purchased salmon (rice balls) at 12:00 on April 1, 202x. Note that the event database 302c may include other records, but these are omitted here for simplicity.

[0077] The information stored in the event database 302c is automatically input by, for example, the information processing device 100. Specifically, the information processing device 100 acquires information about products read by the cash register. The information about products read by the cash register includes, for example, product information used in the POS system. Note that existing technology is used for processing the POS system. Furthermore, the information processing device 100 detects the occurrence of an event that a customer has purchased a product, for example, based on information indicating that the cash register has read a product. For example, when the cash register reads a product, the information processing device 100 detects that the customer has purchased a product. Alternatively, the information processing device 100 may detect the occurrence of an event that a customer has purchased a product, for example, based on an image captured by a camera. For example, when a customer and a product are shown in an image captured near a cash register, the information processing device 100 may detect that the customer has purchased a product.

[0078] Returning to FIG. 5, the explanation will be resumed. The information processing device 100 determines whether now is dynamic timing (step S4). For example, the conditions for the information processing device 100 to determine whether now is dynamic timing include a condition item related to a combination of different types of information or a condition item related to a combination of past data D1 and real-time data D2. Specifically, the condition item is that the customer photographed by the camera has a predetermined attribute. The predetermined attribute is, for example, an attribute common to the customer who visited the store and the customer stored in the past data D1.

[0079] As an example, assume that a thin male customer in his 30s visits the store at 12:00 on May 1, 202x (event ID: EVx1). The information processing device 100 determines whether the timing is dynamic based on, for example, the customer's attributes "gender: male, age: 30s, body type: thin" shown in the image captured by the camera and events (event IDs: EV1, EV7, and EV11) related to the customer's attributes "gender: male, age: 30s, body type: thin" in the event database 302c. In the example of "EV1" in the event database 302c, the customer's attributes are "gender: male, age: 30s, body type: thin," so the information processing device 100 determines that the customer has the predetermined attributes (i.e., it determines that the current timing is dynamic). Note that if it is determined that the current timing is not dynamic, the information processing device 100 ends the processing.

[0080] If it is determined that the timing is dynamic, the information processing device 100 generates a prompt P for recommending a specific product to the customer (step S5). The information processing device 100 generates the prompt P based on, for example, past data D1 and real-time data D2. Alternatively, the output device 500 may be a smartphone owned by a store clerk. Specifically, the store clerk may check the recommended product displayed on the smartphone and verbally recommend it to the customer.

[0081] FIG. 14 is a diagram showing an example of an output from a smart store. As with the other examples, the prompt P is composed of, for example, first input data P1 generated based on real-time data D2, second input data P2 generated based on past data D1, and output instructions P3. The first input data is information about the current situation. For example, information about [DATE], [ENVIRONMENT], and [ATTRIBUTE] is input into the first input data P1. In this example, [ATTRIBUTE] is a customer attribute. Note that the first input data P1 may be provided with other items [ ] (e.g., [PLACE], [IMAGE], etc.), and information corresponding to those items may be input, but other descriptions are omitted here to simplify the drawing.

[0082] The second input data is information relating to past situations. For example, information relating to [INPUT] is input to the second input data P2. The output instruction is information for instructing the AI 400 to perform processing. For example, information relating to [INSTRUCT] is input to the output instruction P3.

[0083] The information processing device 100 generates the first input data P1 portion of the prompt P based on, for example, real-time data D2 acquired by a camera. The information processing device 100, for example, inputs "date and time" information from the real-time data D2 into [DATE]. The information processing device 100 may, for example, acquire information to be input into [ENVIRONMENT] from the stationary device 200, or may acquire weather information for the current location of the information processing device 100 from an external source via the network 2. The information processing device 100, for example, inputs "attribute" information from the real-time data D2 into [ATTRIBUTE].

[0084] [DATE] May 1, 202x 12:00 [ENVIRONMENT] Sunny (weather), 23℃ (temperature), 56% (humidity) [ATTRIBUTE] Male, 30s, thin ...

[0085] Furthermore, the information processing device 100 extracts, for example, the "date and time" and "attributes" from the event database 302c to generate the second input data P2. Specifically, the following information is generated from the record for "EV1" in the event database 302c. The same applies to other records. Note that the information entered in [INPUT] may be generated from multiple records, but other details are omitted here to simplify the drawing. [INPUT] April 1, 202x, 12:00, Male, 30s, Thin, Salmon ...

[0086] Furthermore, the information processing device 100 generates the portion of the output instruction P3 based on, for example, a pre-stored fixed phrase, "The above event has occurred. How should we respond?" Note that the information input to the output instruction P3 is not limited to this.

[0087] [INSTRUCT] The above event has occurred. How should I respond?

[0088] The information processing device 100 inputs the generated prompt P to the AI 400 (step S6). The information processing device 100 outputs processing based on the answer A to the prompt P acquired from the AI 400 to the output device 500 corresponding to the real-time data D2 (step S7). For example, the AI 400 generates image data A2 of the answer A for causing the monitor to recommend salmon (rice balls) based on the input prompt P. The information processing device 100, for example, acquires the generated image data A2. Thereafter, the information processing device 100 outputs the acquired answer A to the output device 500. For example, the information processing device 100 displays the acquired image data A2 on a monitor installed in the store.

[0089] 2-4.Smart Farm Below, a specific example of operation when the target space is a "farm" will be described. This farm is a so-called smart farm in which the information processing system 1 has been installed and agricultural electrical appliances and agricultural equipment are connected via communication technology. Note that the user in this example is a breeder of crops (hereinafter simply referred to as "breeder") in the smart farm in which the information processing device 100 is installed.

[0090] FIG. 15 is a diagram illustrating the system configuration of a smart farm. In this example, multiple stationary devices 200 include environmental sensor devices that detect the farm environment and imaging devices that capture images of crops. The environmental sensor devices are, for example, devices such as a thermometer, a hygrometer, and a pH meter. The thermometer, the hygrometer, and the pH meter are installed, for example, in a greenhouse. The imaging devices are, for example, cameras. The cameras are installed, for example, in the greenhouse. In this example, the output device 500 also includes a cultivation device that takes action according to the state of the crops. The cultivation devices are, for example, devices such as sprinklers and air conditioning. The air conditioning and sprinklers are installed, for example, in the greenhouse.

[0091] In the following example, a case will be described in which, among the plurality of stationary devices 200, the environmental sensor devices are thermometers, the photographing devices are cameras, and the growing devices among the output devices 500 are sprinklers and air conditioners.

[0092] The following description will be given based on the sequence diagram of Fig. 5. The information processing device 100 acquires, for example, information indicating an image captured by a camera as real-time data D2 (step S1). The image captured by the camera shows, for example, (the state of) a crop being grown.

[0093] Next, the information processing device 100 detects the occurrence of an event that the crop is infested by insects or disease, for example, based on an image captured by a camera (step S2). For example, if the captured image shows the crop being infested by insects or disease, the information processing device 100 detects that the crop is infested by insects or disease.

[0094] Next, the information processing device 100 accesses, for example, the history database 300 (step S3). The history database 300 includes, for example, history related to the farm. The history database 300 also stores a plurality of databases. The history database 300 includes, for example, an event database 302d.

[0095] FIG. 16 is a diagram illustrating an example of the event database 302d. The event database 302d stores events that have occurred in the past. In this example, an event refers to the state of a crop or the response to the crop, such as an abnormal occurrence, response, prevention, or death. The event database 302d includes multiple records. Each record stores information about each past event. The event database 302d includes, for example, an "event ID," "event content," "date and time," "attributes," "image," and "details." The event ID is a serial code assigned to each event. In this example, the event content refers to the state of the crop or the response to the crop, which represents the content of each event. The state refers to the state of the detected object (in this example, the crop). The crop state refers to, for example, an abnormal occurrence or its prevention. The response to the crop refers to, for example, the response to the crop, such as spraying pesticides, spraying fertilizer, adjusting temperature, and adjusting humidity. In this example, the attribute refers to the type of abnormal occurrence, such as insect damage or disease. The details refer to details about the event content. In detail, for example, if the event content is "an abnormality occurs," the cause is stored, or if the event content is "a response," the implementation content is stored.

[0096] In the example of FIG. 16, the top record of the event database 302d includes "Event ID: EV1," "Event content: Abnormal occurrence," "Date and time: 202x / 04 / 01 10:00," "Attribute: Insect damage," "Image: (Image of crops and caterpillars)," and "Details: xx insect." This indicates that for the event with event ID EV1, an abnormal insect damage caused by xx insect occurred in the crops at 10:00 on April 1, 202x. Furthermore, the second record from the top of the event database 302d includes "Event ID: EV2," "Event content: Action," "Date and time: 202x / 04 / 01 11:00," "Attribute: -," "Image: (Image of crops)," and "Details: Pesticide x sprayed (1L / m 2 )" in the event with event ID EV2. This means that the breeder will inject 1L / m of drug x at 11:00 on April 1, 202x. 2 This indicates that the crops were sprayed at a concentration of

[0097] Furthermore, the third record from the top of the event database 302d includes "Event ID: EV3," "Event Content: Pest Control," "Date and Time: 202x / 04 / 02 10:00," "Attributes: -," "Image: (Leaf Image)," and "Details: -." This indicates that for the event with event ID EV3, an abnormality that had occurred in the crop at 11:00 on April 1, 202x was controlled. Note that the event database 302d may include other records, but these are omitted here for simplicity. Information stored in the event database 302d is, for example, input manually by the breeder or automatically by the information processing device 100.

[0098] Returning to FIG. 5, the explanation will be resumed. The information processing device 100 determines whether or not the present timing is dynamic (step S4). For example, the conditions for the information processing device 100 to determine whether or not the present timing is dynamic include a condition related to a combination of different types of information or a condition related to a combination of past data D1 and real-time data D2. Specifically, the condition is that the customer photographed by the camera has a predetermined attribute. The predetermined attribute includes, for example, a condition that the state of the crop photographed by the camera is detected to be a predetermined state. The predetermined state is, for example, a state in which an abnormality such as insect damage or disease has occurred.

[0099] As an example, assume that an abnormality in the occurrence of insect damage caused by xx insects is detected in a crop at 10:00 on May 1, 202x (event ID: EVx1). The information processing device 100 determines whether the timing is dynamic based on, for example, the crop condition (attribute: insect damage) and (details: xx insect) shown in an image captured by a camera, and events (event IDs: EV1-3 and EV10-12) related to the "attribute: insect damage" and "details: xx insect" in the event database 302d. In "EV2" in the event database 302d, the grower has performed "details: spraying pesticide x" to combat the insect damage, so the information processing device 100 determines that the crop is in a predetermined state (i.e., it is determined that the current timing is dynamic). Note that if it is determined that the timing is not dynamic, the information processing device 100 terminates processing.

[0100] If it is determined that the timing is dynamic, the information processing device 100 generates a prompt P for taking measures according to the state of the crop (step S5). The information processing device 100 generates the prompt P based on, for example, the past data D1 and the real-time data D2.

[0101] FIG. 17 is a diagram showing an example of an output from a smart farm. As in other examples, the prompt P is composed of, for example, first input data P1 generated based on real-time data D2, second input data P2 generated based on past data D1, and output instructions P3. In this example, information regarding, for example, [DATE], [ENVIRONMENT], and [ATTRIBUTE] is input to the first input data P1. [ENVIRONMENT] is, for example, meteorological information such as weather, temperature, and humidity. In this example, [ATTRIBUTE] is the condition of the crops. Note that the first input data P1 may be provided with other items [ ] (for example, [PLACE], [IMAGE], etc.), and information corresponding to those items may be input; however, other descriptions are omitted here to simplify the drawing.

[0102] For example, information relating to [INPUT] is input to the second input data P2. For example, information relating to [INSTRUCT] is input to the output instruction P3.

[0103] The information processing device 100 generates the first input data P1 portion of the prompt P based on, for example, real-time data D2 acquired by a camera. The information processing device 100, for example, inputs "date and time" information from the real-time data D2 into [DATE]. The information processing device 100 may, for example, acquire information to be input into [ENVIRONMENT] from the stationary device 200, or may acquire weather information for the current location of the information processing device 100 from an external source via the network 2. The information processing device 100, for example, inputs "attribute" and "details" information from the real-time data D2 into [ATTRIBUTE].

[0104] [DATE] May 1, 202x 10:00 [ENVIRONMENT] Sunny (weather), 23℃ (temperature), 56% (humidity) [ATTRIBUTE] insect damage, xx insects ...

[0105] Furthermore, the information processing device 100 generates the second input data P2 by extracting, for example, the "date and time" and "details" from the event database 302d. Specifically, the following information is generated from the "EV2" record in the event database 302d. Note that the information entered in [INPUT] may be generated from multiple records, but other details are omitted here to simplify the drawing. [INPUT] April 1, 202x 10:00, insect damage, xx insects April 1, 202x, 11:00, Spraying of chemicals (1L / m 2 ) April 2, 202x, 10:00, Pest Control ...

[0106] Furthermore, the information processing device 100 generates the portion of the output instruction P3 based on, for example, a pre-stored fixed phrase, "The above event has occurred. How should we respond? Please generate a command to be output to the corresponding device." Note that the information input to the output instruction P3 is not limited to this.

[0107] [INSTRUCT] The above event has occurred. How should we respond? Generate a command to output to the corresponding device.

[0108] The information processing device 100 inputs the generated prompt P to the AI 400 (step S6). The information processing device 100 outputs a process based on the answer A to the prompt P acquired from the AI 400 to the output device 500 corresponding to the real-time data D2 (step S7). For example, based on the input prompt P, the AI 400 may inject 1 L / m of chemical x into the sprinkler. 2The information processing device 100 generates program data A4 of answer A for spraying chemical x at a concentration of 1 L / m. The information processing device 100, for example, acquires the generated program data A4. Thereafter, the information processing device 100 outputs the acquired answer A to the output device 500. The information processing device 100, for example, uses the acquired program data A4 to generate program data A4 for spraying chemical x at a concentration of 1 L / m to a sprinkler installed in a greenhouse. 2 Alternatively, the output device 500 may be a smartphone owned by the grower. Specifically, the grower may check the measures corresponding to the state of the crops displayed on the smartphone and carry them out themselves.

[0109] As described above, the information processing device 100 can more accurately determine whether or not dynamic timing exists based on a combination of the past data D1 and the real-time data D2 (i.e., dynamically generate prompts to be input to the AI in response to changes in the situation). Furthermore, the information processing device 100 can reduce the burden on the user by supporting decision-making based on advanced data analysis. Furthermore, the information processing device 100 can reduce the burden on the user and the system by suppressing unnecessary output such as notifications. Furthermore, the information processing device 100 can output optimal output according to the situation, rather than a fixed output in response to the occurrence of an event.

[0110] 3. Variations The present invention is not limited to the above-described embodiment, and various modifications are possible. Two or more of the features described in the following modifications may be applied in combination.

[0111] Furthermore, the operation examples described in the embodiments are merely examples, and part of the matters described in one operation example may be applied in combination with part of the matters described in another operation example.

[0112] The correspondence between the functions and hardware in the information processing device 100 is not limited to that exemplified in the embodiment. For example, a plurality of physical devices may cooperate to have the functions of the information processing device 100. Furthermore, some of the functions of the information processing device 100 exemplified in the embodiment may be omitted.

[0113] The hardware configuration of the information processing device 100 is not limited to that exemplified in the embodiment. For example, the information processing device 100 may be a server on a computer network. This server may be a physical server or a virtual server (so-called cloud).

[0114] The program executed by processor 151 may be provided in a state recorded on a computer-readable recording medium such as a DVD-ROM, or may be provided by downloading via a network such as the Internet. In short, the information processing system according to the present invention may include the steps of: detecting the occurrence of a predetermined event based on real-time data obtained from a plurality of stationary devices installed in a target space; accessing a database that records past data collected in the target space; when the occurrence of the event is detected, inputting a prompt including the real-time data and the past data into an artificial intelligence; and outputting, to an output device corresponding to the real-time data, information regarding the processing of a question or task based on the answer to the prompt obtained from the artificial intelligence. [Explanation of symbols]

[0115] 100...information processing device, 101...communication means, 102...detection means, 103...access means, 104...determination means, 105...generation means, 106...input means, 107...acquisition means, 108...output means, 109...storage means, 200...stationary device, 300...past database, 400...AI, 500...output device

Claims

1. a detection means for detecting the occurrence of a predetermined event based on real-time data obtained from a plurality of stationary devices installed in a target space; access means for accessing a database that records past data collected in the target space; an input means for inputting a prompt including the real-time data and the past data to an artificial intelligence when the occurrence of the event is detected; an output means for outputting information about a question sentence or task processing based on an answer to the prompt acquired from the artificial intelligence to an output device corresponding to the real-time data; a determination means for determining whether it is time to output information about the query or task processing to the output device based on the real-time data; a generating means for generating the prompt when it is determined that it is time to output information about the question sentence or task processing; An information processing device having the above.

2. the plurality of stationary devices include a plurality of types of devices that output different types of information, The determining means determines whether it is the timing based on a condition related to a combination of the different types of information. The information processing device according to claim 1 .

3. The determining means determines whether the timing is right based on a condition related to a combination of the past data and the real-time data. The information processing device according to claim 1 .

4. the target space is a house, the plurality of stationary devices includes an access control device that controls entry and exit of people into the house, the condition includes a condition that the entry / exit control device detects that a person has returned to the house, the historical data includes history associated with the home; The generating means generates the prompt so as to include a history of a person detected as having returned home from the past data.

4. The information processing device according to claim 2 or 3.

5. the plurality of stationary devices include an imaging device that images a person, the condition includes a condition that when the entry and exit control device detects that one of the residents has returned home while at least one resident of the house is present in the house, the state of the person who has returned home, photographed by the photographing device, is a predetermined state; The generating means generates a prompt for suggesting to a person already at home to say something to the person who has just returned home. The information processing device according to claim 4 .

6. the target space is a house, the plurality of stationary devices include an access control device that controls entry and exit of people into the house and an image capture device that captures images of people or objects, the historical data includes behavioral history of the resident of the house; the condition includes a condition that a predetermined item is included in the past data, The generating means generates a prompt for suggesting that the person at home or the person who has returned home say something according to the predetermined item.

4. The information processing device according to claim 2 or 3.

7. The target space is a store, the plurality of stationary devices include a photographing device that photographs customers in the store, the condition includes a condition that the customer photographed by the photographing device has a predetermined attribute; The generating means generates a prompt for recommending a predetermined product to the customer.

4. The information processing device according to claim 2 or 3.

8. the target space is a farm, the plurality of stationary devices include an environmental sensor device that detects an environment of the farm and an image capturing device that captures images of crops, the condition includes a condition that the state of the crop photographed by the photographing device is a predetermined state; The generating means generates a prompt to cause a user to take action according to the state of the crop.

4. The information processing device according to claim 2 or 3.

9. detecting the occurrence of a predetermined event based on real-time data obtained from a plurality of stationary devices installed in a target space; accessing a database that records past data collected in the target space; inputting a prompt including the real-time data and the historical data into an artificial intelligence when the occurrence of the event is detected; outputting information about a query or task processing based on the response to the prompt obtained from the artificial intelligence to an output device corresponding to the real-time data; determining whether it is time to output information about the query or task processing to the output device based on the real-time data; generating the prompt when it is determined that it is time to output information about the question or task processing; A method having the following.

Images