Information provision method and information provision device

Abstract

This provides an information provision method that enables the easy generation of optimal landscape plans by utilizing generation AI. [Solution] The information provision method performed by the information provision device 3 includes: a question reception process that receives question information D1, which is construction target information relating to the construction target of the landscaping work; a prompt generation process that generates prompt information D3 that instructs the generative artificial intelligence model 20 to output answer information D4, which is plan information indicating a plan for carrying out landscaping work on the construction target, based on the question information D1; and an answer provision process that provides the answer information D4 output from the generative artificial intelligence model 20 by inputting the prompt information D3 to the generative artificial intelligence model 20.

Description

Technical Field

[0001] The present invention relates to an information providing method and an information providing apparatus.

Background Art

[0002] In recent years, generative artificial intelligence models called generative AI and generative AI have been utilized for various purposes and applications (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the system disclosed in Patent Document 1, an efficient reminder is set using generative AI. However, the utilization of generative AI is required in more diverse fields without being limited to the efficiency improvement of general desk work such as the system disclosed in Patent Document 1.

[0005] In view of the above problems, the present invention has been made, and by utilizing generative AI, it is possible to comprehensively support a series of operations from the planning to the construction of a garden and further to the maintenance management after completion, and to provide an information providing method and an information providing apparatus capable of realizing an optimal garden.

Means for Solving the Problems

[0006] In order to achieve the above object, an information providing method according to an aspect of the present invention is an information providing method for providing information by a computer, A prompt generation process generates prompt information that instructs a generative artificial intelligence model to output answer information based on the aforementioned question information, The process involves inputting the prompt information into the generative artificial intelligence model and then providing the response information output from the generative artificial intelligence model. The aforementioned question information is, This is information regarding the scope of work for landscaping projects. The aforementioned response information is, This is planning information showing the plan for carrying out the aforementioned landscaping on the aforementioned construction site. [Effects of the Invention]

[0007] According to one aspect of the present invention, by utilizing generated AI, it is possible to comprehensively support a series of tasks, from planning and construction of a landscape to maintenance and management after completion, thereby realizing an optimal landscape.

[0008] Other issues, configurations, and effects will be clarified in the embodiments for carrying out the invention described later. [Brief explanation of the drawing]

[0009] [Figure 1] This is an overall diagram showing an example of Information Provision System 1. [Figure 2] This is a block diagram showing an example of an information provision device 3. [Figure 3] This is a hardware configuration diagram showing an example of a computer 900 that makes up each device. [Figure 4] This is a flowchart illustrating an example of the operation (one stage) of Information Provision System 1. [Figure 5] This is a flowchart illustrating an example of the operation (two stages) of Information Provision System 1. [Figure 6] This is a data structure diagram showing an example of the reference information management database 50. [Figure 7] This figure shows an example of construction target information and planning information. [Modes for carrying out the invention]

[0010] <Common Embodiment: Generative AI> Before providing a detailed description of the present invention, a general explanation of the common behavior of generation AI, its basic concepts, configuration, and operation will be given with reference to the drawings. This will deepen understanding of the specific embodiments for carrying out the present invention described later. Any parts omitted from the explanation will be based on prior art. Below, the scope necessary for explaining how to achieve the objectives of the present invention will be schematically shown, and the scope necessary for explaining the relevant parts of the present invention will be mainly explained, with any parts omitted from the explanation being based on prior art.

[0011] (System Overview) Figure 1 is an overall diagram showing an example of the information provision system 1. The information provision system 1 comprises, as its main components, a generative AI device 2, an information provision device 3 that operates in cooperation with the generative AI device 2, a user terminal device 4 used by user U, and a database device 5 where various types of information are stored. Each device 2 to 5 is, for example, composed of a general-purpose or dedicated computer (see Figure 3 below) and connected to a wired or wireless network 6, enabling the mutual transmission and reception of various types of data. Note that the number of each device 2 to 5 and the connection configuration of the network 6 are not limited to the example in Figure 1 and may be changed as appropriate.

[0012] The generative AI device 2 is composed of, for example, a server-type computer or a cloud-type computer. The generative AI device 2 includes, for example, a generative artificial intelligence model 20, also known as generative AI or generative AI. The generative artificial intelligence model 20 receives instructions from prompt information D3 and generates response information D4 in response to those instructions.

[0013] The generative artificial intelligence model 20 is a trained model constructed using a vast learning dataset and deep learning techniques, and a base model (e.g., a large language model, etc.) or a multimodal model, etc. is used. Note that the generative artificial intelligence model 20 may utilize what is provided as a service by an external contractor, and in that case, the generative AI device 2 may be omitted from the components of the information providing system 1.

[0014] The information providing device 3 is composed of, for example, a server-type computer or a cloud-type computer. When the information providing device 3 receives the question information D1 from the user terminal device 4, it generates prompt information D3 for the generative artificial intelligence model 20 and transmits it to the generative AI device 2. Also, when the information providing device 3 receives the answer information D4 from the generative AI device 2 generated by the generative artificial intelligence model 20, it provides the answer information D4 to the user terminal device 4.

[0015] The user terminal device 4 is composed of, for example, a stationary computer or a portable computer. The user terminal device 4 has programs such as an application or a browser installed, accepts various input operations, and outputs various information via a display screen or voice. The user terminal device 4 displays a display screen for accepting an input operation for inputting the question information D1 or outputting the answer information D4, etc., and transmits and receives various data such as the question information D1 and the answer information D4, etc. to and from the information providing device 3.

[0016] The database device 5 is composed of, for example, a server-type computer or a cloud-type computer. The database device 5 includes a reference information management database 50 (described later). The reference information D2 is stored in the reference information management database 50. The reference information management database 50 is accessible from each of the devices 2 to 4, and is, for example, referred to when the generative artificial intelligence model 20 generates the answer information D4. Also, the reference information management database 50 undergoes editing operations such as addition, deletion, and modification via the information providing device 3 or the user terminal device 4.

[0017] (Overview of Each Piece of Information) The question information D1 is data regarding inquiries, instructions, or situations related to a specific task or objective for the generative artificial intelligence model 20, and is information used to prompt the generation of the response information D4 by the model. The question information D1 may be text data in natural language, image data, audio data, numerical data, structured data, or a combination thereof, and is not limited thereto as long as it is interpretable by the generative artificial intelligence model 20.

[0018] The reference information D2 is information retrieved from an external database, knowledge base, or other information source and used as part of the prompt information D3 to enhance the accuracy, comprehensiveness, or specificity (specialization) when the generative artificial intelligence model 20 generates the response information D4. The reference information D2 may be dynamically retrieved by a search module within the information providing system 1 or an external database search function based on the question information D1 or the intermediate response information D4 by the generative artificial intelligence model 20 in the Retrieval-Augmented Generation (RAG) mechanism, or may be directly included in and provided with the question information D1. The reference information D2 may be text data in natural language (e.g., documents, articles, papers), structured data (e.g., tabular data, JSON data), or a specific data structure such as a knowledge graph, and is not limited thereto as long as it is usable by the generative artificial intelligence model 20.

[0019] The prompt information D3 is information input to the generative artificial intelligence model 20 in a form including the question information D1, instructions, constraints, and / or reference information D2 in order to cause the generative artificial intelligence model 20 to execute a specific task or generate the response information D4 in a specific format. The prompt information D3 may mainly be text data in natural language, and is not limited thereto as long as it can correspond to the input format of the generative artificial intelligence model 20.

[0020] The response information D4 is data generated by the generative artificial intelligence model 20 based on the input question information D1, and is information that contributes to solving the problem indicated by the question information D1 or to achieving the objective, or provides new insights or creative works. The response information D4 can take various forms depending on the function and purpose of use of the generative artificial intelligence model 20, such as natural language text data (e.g., text, summary, code), image data, audio data, video data, or structured data (e.g., JSON data, XML data), but is not limited to these as long as it can be generated by the generative artificial intelligence model 20.

[0021] (Configuration of Information Provisioning Device 3) Figure 2 is a block diagram showing an example of an information providing device 3. The information providing device 3 comprises a control unit 30 composed of a processor, a storage unit 31 composed of an HDD, SSD, memory, etc., a communication unit 32 which is a communication interface with the network 6, an input unit 33 composed of a keyboard, mouse, etc., and a display unit 34 composed of a display, etc. Note that the input unit 33 and the display unit 34 may be omitted.

[0022] The memory unit 31 stores the prompt syntax data 310 and the information processing program 311, as well as the operating system, other programs, various types of data, etc.

[0023] The prompt syntax data 310 is used to generate prompt information D3 from question information D1, and includes multiple syntaxes. Each syntax specifies reference information D2 to be referenced when generating answer information D4, or specifies a task to be imposed on the generative artificial intelligence model 20.

[0024] The control unit 30 functions as a question receiving unit 300, a prompt generation unit 301, and an answer provision unit 302 by executing an information processing program 311 stored in the storage unit 31. Each of the units 300 to 302 of the control unit 30 functions as a user interface with the user U by transmitting display information to the user terminal device 4 to display various display screens on the user terminal device 4 and accepting various input operations through the display screens.

[0025] The question reception unit 300 performs question reception processing to receive question information D1.

[0026] The prompt generation unit 301 performs a prompt generation process that generates prompt information D3, which instructs the generative artificial intelligence model 20 to output answer information D4 while referring to reference information D2, based on the question information D1.

[0027] The response provision unit 302 performs a response provision process that provides response information D4 output from the generative artificial intelligence model 20 by inputting prompt information D3 to the generative artificial intelligence model 20.

[0028] (Hardware configuration of each device) Figure 3 is a hardware configuration diagram showing an example of the computer 900 that constitutes each device. Each device 2 to 5 in the information provision system 1 is composed of a general-purpose or dedicated computer 900.

[0029] As shown in Figure 3, the computer 900 comprises, as its main components, a bus 910, a processor 912, memory 914, an input device 916, an output device 917, a display device 918, a storage device 920, a communication interface unit 922, an external device interface unit 924, an I / O device interface unit 926, and a media input / output unit 928. Note that the above components may be omitted as appropriate depending on the intended use of the computer 900.

[0030] The processor 912 consists of one or more arithmetic processing units (CPU (Central Processing Unit), MPU (Micro-Processing Unit), DSP (Digital Signal Processor), GPU (Graphics Processing Unit), NPU (Neural Processing Unit), etc.) and operates as a control unit that oversees the entire computer 900. The memory 914 stores various data and programs 930 and consists of volatile memory (DRAM, SRAM, etc.) that functions as main memory, and non-volatile memory (ROM), flash memory, etc.

[0031] The input device 916 consists of, for example, a keyboard, mouse, numeric keypad, or electronic pen, and functions as an input unit. The output device 917 consists of, for example, a sound (voice) output device or a vibration device, and functions as an output unit. The display device 918 consists of, for example, a liquid crystal display, an organic EL display, electronic paper, or a projector, and functions as an output unit. The input device 916 and the display device 918 may be configured as an integrated unit, such as a touch panel display. The storage device 920 consists of, for example, an HDD or SSD, and functions as a storage unit. The storage device 920 stores various data necessary for the execution of the operating system and program 930.

[0032] The communication I / F unit 922 is connected by wire or wireless to a network 940 such as the Internet or an intranet (which may be the same as network 6 in Figure 1) and functions as a communication unit that sends and receives data with other computers according to a predetermined communication standard. The external device I / F unit 924 is connected by wire or wireless to external devices 950 such as cameras, printers, scanners, and reader / writers and functions as a communication unit that sends and receives data with external devices 950 according to a predetermined communication standard. The I / O device I / F unit 926 is connected to I / O devices 960 such as various sensors and actuators and functions as a communication unit that sends and receives various signals and data with the I / O devices 960, such as detection signals from sensors and control signals to actuators. The media input / output unit 928 consists of, for example, a drive device such as a DVD drive or CD drive, a memory card slot, and a USB connector, and reads and writes data to media (non-temporary storage media) 970 such as DVDs, CDs, memory cards, and USB memory.

[0033] In the computer 900 having the above configuration, the processor 912 calls and executes the program 930 stored in the storage device 920 in the memory 914, and controls various parts of the computer 900 via the bus 910. The program 930 may also be stored in the memory 914 instead of the storage device 920. The program 930 may be recorded on the media 970 in an installable or executable file format and provided to the computer 900 via the media input / output unit 928. The program 930 may also be provided to the computer 900 by downloading it via the network 940 through the communication interface unit 922. Furthermore, the computer 900 may implement various functions realized by the processor 912 executing the program 930 using hardware such as an FPGA (Field-Programmable Gate Array) or ASIC (Application Specific Integrated Circuit).

[0034] Computer 900 is an electronic device of any form, consisting of, for example, a stationary computer or a portable computer. Computer 900 may be a client computer, a server computer, a cloud computer, or an embedded computer such as a control panel or controller (including microcontrollers, programmable logic controllers, and sequencers).

[0035] (Operation of Information Provision System 1) The following describes a series of operations performed by the information provision system 1. This series of operations is performed through the cooperation of each part 300-302 of the information provision device 3 (each process of the information provision method executed by the information processing program 311), the generative AI device 2, the user terminal device 4, and the database device 5.

[0036] (In the case of a single prompt) Figure 4 is a flowchart illustrating an example of the operation (one stage) of the information provision system 1. The following explanation assumes that reference information D2 is registered in the reference information management database 50.

[0037] First, in step S100, the user terminal device 4 transmits question information D1 based on the user U's input operation on the display screen to the information providing device 3.

[0038] Next, in step S110 (question reception processing), the question reception unit 300 of the information providing device 3 receives the question information D1 transmitted in step S100, thereby accepting the question information D1.

[0039] Next, in step S200 (prompt generation process), the prompt generation unit 301 generates prompt information D3 that instructs the generative artificial intelligence model 20 to output answer information D4 while referring to reference information D2, based on the question information D1 received in step S110.

[0040] Next, in step S210 (answer provision process), the answer provision unit 302 inputs the prompt information D3 generated in step S200 into the generative artificial intelligence model 20, thereby generating answer information D4 as an output result from the generative artificial intelligence model 20.

[0041] Next, in step S220 (answer provision processing), the answer provision unit 302 sends display screen information for displaying the answer information D4 generated in step S210 to the user terminal device 4, which is the source of the question information D1. At this time, the display screen information may also include the prompt information D3 generated in step S200.

[0042] Next, in step S230, when the user terminal device 4 receives the display screen information transmitted in step S220, it displays a display screen containing the response information D4 based on that display screen information.

[0043] Through the above series of processes, the answer information D4 to the question information D1 is provided to user U. The same series of processes will be performed again if user U inputs new question information D1.

[0044] (In the case of a two-step prompt) Figure 5 is a flowchart showing an example of the operation (two stages) of the information provision system 1. The flow shown in Figure 5 differs from the flow shown in Figure 4 in that it uses a multi-stage prompt (two stages). In the following explanation, it is assumed that reference information D2 is registered in the reference information management database 50.

[0045] In the flow shown in Figure 5, prompt information D3 refers to either the first prompt information D3A or the second prompt information D3B. Similarly, response information D4 refers to either the first response information D4A or the second response information D4B. Furthermore, reference information D2 refers to either the first reference information D2A or the second reference information D2B.

[0046] The first response information D4A is the first output result obtained from the generative artificial intelligence model 20. The first prompt information D3A is information that instructs the generative artificial intelligence model 20 to output the first response information D4A based on the question information D1 and while referring to the first reference information D2A.

[0047] The second response information D4B is the second output result obtained from the generative artificial intelligence model 20, and is the final answer to the question information D1. The second prompt information D3B is information that instructs the generative artificial intelligence model 20 to output the second response information D4B based on the first response information D4A and while referring to the second reference information D2B.

[0048] First, in step S100, the user terminal device 4 transmits question information D1 based on the user U's input operation on the display screen to the information providing device 3.

[0049] Next, in step S110 (question reception processing), the question reception unit 300 of the information providing device 3 receives the question information D1 transmitted in step S100, thereby accepting the question information D1.

[0050] Next, in step S200 (first prompt generation process), the prompt generation unit 301 generates first prompt information D3A, which instructs the generative artificial intelligence model 20 to output first answer information D4A while referring to first reference information D2A, based on the question information D1 received in step S110.

[0051] Next, in step S210 (first response provision process), the response provision unit 302 inputs the first prompt information D3A generated in step S200 into the generative artificial intelligence model 20, thereby generating the first response information D4A as an output result from the generative artificial intelligence model 20.

[0052] Next, in step S220 (first answer provision process), the answer provision unit 302 sends display screen information for displaying the first answer information D4A generated in step S210 to the user terminal device 4, which is the source of the question information D1. At this time, the display screen information may also include the first prompt information D3A generated in step S200.

[0053] Next, in step S230, when the user terminal device 4 receives the display screen information transmitted in step S220, it displays a display screen including the first response information D4A based on that display screen information.

[0054] Next, in step S240 (second prompt generation process), the prompt generation unit 301 generates second prompt information D3B, which instructs the generative artificial intelligence model 20 to output second response information D4B while referring to second reference information D2B, based on the first response information D4A generated in step S210.

[0055] Next, in step S250 (second response provision process), the response provision unit 302 inputs the second prompt information D3B generated in step S240 into the generative artificial intelligence model 20, thereby generating the second response information D4B as an output result from the generative artificial intelligence model 20.

[0056] Next, in step S260 (second answer provision process), the answer provision unit 302 sends display screen information for displaying the second answer information D4B generated in step S250 to the user terminal device 4, which is the source of the question information D1. At this time, the display screen information may also include the second prompt information D3B generated in step S240.

[0057] Next, in step S270, when the user terminal device 4 receives the display screen information transmitted in step S260, it displays a display screen including the second response information D4B based on that display screen information.

[0058] Through the above series of processes, the second answer information D4B, which is the final answer to the question information D1, is provided to user U. The same series of processes will be performed again if user U inputs new question information D1.

[0059] In the following first and second embodiments, the explanation will focus on the case where, when a landscape gardener (user) using the user terminal device 4 receives a request for landscaping from a customer, the information providing device 3 generates planning information showing a plan for carrying out landscaping on the construction site and provides said planning information to the user.

[0060] <First Embodiment> Next, a first embodiment for carrying out the present invention will be described with reference to the above-mentioned common embodiment drawings. Note that detailed explanations of parts similar to those in the common embodiment regarding the functions and connections of each component in the system, the various data handled by each component, the configuration and operation of each part in each device, the various data handled by each device, and the operation and configuration in the flowchart will be omitted.

[0061] (Overview of the system according to the first embodiment) Information Provision System 1 functions as a system that provides plans for carrying out landscaping on the construction site. Landscaping is a broad concept that includes not only a series of actions to design and construct gardens, parks, green spaces, etc., as aesthetically or functionally, but also maintenance to keep the completed space in good condition. The construction site is the land, area, or part of a building (e.g., garden, veranda, rooftop) on which the landscaping will be carried out.

[0062] Here, "process" in this disclosure refers to a work unit in landscaping (for example, "arranging landscape stones" or "pruning trees"), and "construction" refers to the act of specifically carrying out those processes, including the manner in which work is carried out based on real-time instructions and assistance provided by the information provision system 1. On the other hand, "plan" refers to information that presents workers with information on the procedures for the work to be specifically performed within the work unit, the duration of each work, the costs associated with the work, material list information for each work, and the completed image for each work, and also refers to information that includes various information necessary for construction, such as the overall design drawings and execution procedure manuals for the landscaping project, and further encompasses information covering all stages of landscaping work, such as specific work instructions provided during construction and the annual maintenance schedule after completion.

[0063] (Summary of each piece of information according to the first embodiment) Question information D1 is construction target information regarding the landscaping work to be carried out. Question information D1 may also include text data containing sentences representing information about the construction target from the user, along with the construction target information. The construction target information may be entered in any data format, either by a file containing the construction target information or by the file name of a reference to access the file containing the construction target information.

[0064] The construction target information may include, for example, image data representing the situation before, during, or after the landscaping work is carried out on the construction target, or requirement information that is the requirement specification for the landscaping work. More specifically, real-time video data taken with smart glasses or the like worn by the landscaping contractor during the work, or still image data of plants suspected of being affected by pests or diseases, may also be included in the construction target information.

[0065] Image data representing the condition of the construction site before landscaping work is carried out may include, for example, two-dimensional still image data (JPEG, PNG format, etc.) taken with a digital camera or smartphone, video data (MP4 format, etc.), design drawing data (CAD data, etc.), three-dimensional point cloud data or mesh data acquired using sensors such as LiDAR, etc., and is not limited to these, as long as it is data that visually shows the current state of the construction site. Furthermore, image data and video data may include supplementary information such as location information at the time of shooting and dimensional information of the objects being photographed.

[0066] The requirements information is the specification for the landscaping, indicating the specific requests and constraints that the customer has for the landscaping. The requirements information may include, for example, design requests entered as natural language text data (e.g., "I want a calm, Japanese-style atmosphere," "I want a lawn space where children can play"), budget amounts entered as numerical data, garden styles entered as selection forms (e.g., "English garden," "modern"), or specifications of existing trees or structures to be preserved. It may also include more conceptual or vague requests such as "I want a stream" or "I want to be able to enjoy flowers in each season," and is not limited to these, as long as it is information that identifies the specifications of the landscaping desired by the customer.

[0067] Reference information D2 is landscape knowledge information that systematizes knowledge about landscape architecture. Landscape knowledge information functions as a knowledge base for the generative artificial intelligence model 20 to behave like an experienced landscape expert. Landscape knowledge information is not just a list of data, but is stored in the reference information management database 50 as a structured database in which various landscape knowledge is interconnected and the most appropriate knowledge can be dynamically referenced according to the context of the question information D1.

[0068] Landscape knowledge information may include, for example, at least one of the following: construction process know-how information, construction know-how information for each process, plant information, material information, design style information, soil and environmental information, tool and heavy equipment information, and cost and estimation information.

[0069] Construction process know-how information is systematized knowledge about the overall workflow of landscaping work, indicating what procedures (combinations and sequences of processes) should be followed depending on the purpose and conditions of the landscaping. For example, the reference information management database 50 has common landscaping processes such as "ground preparation and undulation," "water features (streams and ponds) foundation construction," "landscape stone installation," "garden path sub-base and partial finishing," "tea garden elements (stepping stones, water basins, etc.)," ​​"hedges, stone walls, and earthen walls," "planting arrangement," and "pruning and maintenance" registered as master data. When the customer's request information is "Japanese garden (dry landscape garden)," the information providing device 3 refers to the master data that constitutes the construction process know-how information and defines a series of process flows such as "ground preparation, weed control sheet laying, landscape stone arrangement, white sand laying, and sand pattern creation." On the other hand, when the request information is "English garden," it defines a different process flow such as "soil improvement, garden path construction, structure installation, tall tree planting, perennial planting, and mulching." Construction process know-how information may include numerous past construction examples (for example, before and after images of construction, videos of work in progress, notes from designers and craftsmen, etc.) or a systematization of standard work procedures according to design style and customer requests. By referring to construction process know-how information, the generative artificial intelligence model 20 can plan the optimal construction schedule according to customer requests.

[0070] Process-specific construction know-how information is systematized information that contains detailed procedures, techniques, and points to note regarding how to specifically carry out each process (task) defined by the construction process know-how information. For example, for the process of "arranging landscape stones," the specific know-how for "arranging landscape stones" would include specialized technical information such as "how to determine the orientation of stones (top, bottom, front, back)," "the basic composition and balancing of a three-stone arrangement," and "guidelines for embedding depth to enhance stone stability," as well as points to note such as "safety management matters when transporting heavy objects" and "criteria for stopping work in case of rain." Furthermore, process-specific construction know-how information may include not only textual procedural information, but also image data or video data that visually shows the actual work, or notes recording tips from skilled craftsmen, among other diverse forms of information.

[0071] The following are examples of other know-how included in the construction know-how information for each process. In the construction of stone masonry, detailed structural information such as cross-sectional diagrams of rough masonry (diagrams showing the components such as top stones, backfill concrete, drainage pipes, foundation stones, and foundation concrete, their relationship to the ground, and construction methods using formwork, bracing, bracing, or support materials), information on the construction methods of structures such as stone fences using logs, battens, and thick planks (diagrams including information on sawmill lines, log texture, and rough logs viewed from above), specialized technical information such as laying crushed stone after excavating the foundation, assembling from the foundation stones in order, and ensuring the slope and drainage holes, and points to note that the strength of the foundation and drainage treatment are the most important points that affect durability are all remembered in association with each other.

[0072] Furthermore, when pruning garden trees, methods for identifying unnecessary branches (such as water sprouts, inverted branches, upright branches, inner branches, parallel branches, and suckers) and pruning methods corresponding to those branch types (including diagrams showing specific cutting techniques such as cutting back, thinning branches, pruning tangled branches, blocking branches, pruning methods for thick branches, and pruning methods for hedges) are associated with and memorized.

[0073] Furthermore, the construction know-how information for each process also includes specific construction procedures for fence construction (traditional types such as Kenninji fences), illustrated with a series of work steps such as processing the posts, digging holes, erecting the posts, marking the layout, selecting and tying the bamboo, attaching split bamboo and bamboo trim, and applying preservatives.

[0074] Plant information is a systematic collection of knowledge regarding the cultivation, growth, selection, and placement of plants used in landscaping. This information includes, for example, plant names, scientific names, tree shapes, tree heights, flowering seasons, sunlight requirements (full sun / partial shade / shade), cold tolerance, required soil, pruning timing and methods, and pest and disease information. It also systematically includes information on various flowering trees (e.g., oleander, Hypericum, camellia, forsythia, etc.) and plant types and placement suitable for mixed flowerbeds.

[0075] Material information is a systematic collection of knowledge about various materials and components used in landscaping work. This information includes, for example, catalog information for stone (type, origin, color, unit price), wood (type, durability, presence or absence of preservative treatment), bricks and tiles (product name, dimensions, color), and exterior products (fences, lighting, etc.). Specifically for stone, it may also include information on types of stone suitable for natural stone masonry, foundation concrete, backfill concrete, and sandstone used for foundations. For wood, information on types, durability, and presence or absence of preservative treatment for logs, bamboo (e.g., moso bamboo, bleached bamboo, cedar logs, etc.), timber, formwork materials, bracing, support materials, wooden pegs, and copper wire is also included. Furthermore, roofing materials such as battens and thick boards, and drainage pipes are also included in the material information.

[0076] Landscape gardening knowledge information includes design style information (historical background and components of each style), soil and environment information (types of soil and improvement methods), tool and heavy machinery information (types and uses of tools used in each process), and cost and estimation information (unit prices of materials, standard labor unit prices, and rates for various expenses). Design style information also includes floor plans, elevations, detailed structural drawings, and photographs of completed images of specific fences such as Kenninji-gaki, Kinkakuji-gaki, and Otsu-gaki, as well as information on components of tea gardens such as middle gates, waiting areas with benches, and stone basins.

[0077] Figure 6 is a data structure diagram showing an example of a reference information management database 50. The landscape knowledge information listed above is stored in the reference information management database 50 as a relational database, for example. The database is designed so that multiple tables are interconnected.

[0078] For example, the reference information management database 50 includes a "design style table" that stores information about design styles and a "process master table" that defines individual construction processes. The reference information management database 50 also includes a "style-specific process flow table" that links design styles with process masters and defines the order of construction. The "style-specific process flow table" forms the core of the construction process know-how information.

[0079] Furthermore, the reference information management database 50 includes a "process-specific know-how information table" that stores detailed technical information and points to note for each process. The "process-specific know-how information table" forms the core of the process-specific construction know-how information and is linked to the "process master table".

[0080] In addition, the reference information management database 50 includes a "plant master table," a "material master table," and a "construction case study table," which store information on the aforementioned plants, materials, and past construction case studies, respectively.

[0081] By using a structured database as described above, the generative artificial intelligence model 20 can efficiently search and extract relevant information in response to user requests, combine it, and generate logical, consistent, and high-quality planning information.

[0082] Response information D4 is planning information that shows the plan for carrying out landscaping on the construction site. Planning information may include, for example, construction procedure information that shows the steps required for construction in landscaping.

[0083] Furthermore, the response information D4 may also include, in addition to the planning information, image data representing the state of the landscaping at the construction site after its completion. The image data representing the state of the landscaping at the construction site after its completion may be, for example, a photorealistic completed image generated by the generative artificial intelligence model 20, a design perspective, or a simple floor plan, and is not limited to these, as long as it is data that visually shows what the planned landscaping will look like when completed.

[0084] Construction procedure information is information that shows the procedures required for construction in landscaping. For example, construction procedure information may include detailed work instructions for each process, specifying what work should be done, what materials and tools should be used, and at what point within the construction area (using coordinates or markers on the drawing). Furthermore, construction procedure information may include reference images showing the state when each process is completed, and images showing the progress of the work in progress. More specifically, construction procedure information may be provided as AR (augmented reality) information superimposed on a display such as smart glasses, or as voice guidance output from smart earphones. In addition, in the "landscape stone placement" process, specific instructions such as what angle the stone in front should be rotated and by how many more degrees may be provided in real time as construction procedure information. Also, in the tree pruning process, construction procedure information may be image data that specifically indicates the position of the branches to be pruned using circles or other symbols on a photograph of the tree before pruning. Furthermore, if an image of a plant in which pests or diseases have been confirmed is input, the construction procedure information may also include information that identifies the pests or diseases and instructs on the recommended type of pesticide, dilution ratio, and application method.

[0085] Furthermore, the construction procedure information may be a chronological list of the construction process, or a detailed instruction sheet including detailed work instructions for each process, a list of necessary tools, an estimated work period, safety precautions, etc. It is not limited to these forms, as long as it allows the user to understand the specific work required to realize the planned landscaping. By providing construction procedure information, even inexperienced workers can accurately understand the target quality and proceed with the work.

[0086] Response information D4 may include not only image data and construction procedure information, but also estimation information, material list information, and maintenance plan information. Estimate information is information showing the costs required to realize the plan, and includes, for example, a breakdown of material costs, labor costs, heavy equipment lease fees, design fees, and miscellaneous expenses. Material list information is information showing the names, part numbers, quantities, and unit prices of plants and materials to be used. Maintenance plan information is information for maintaining the garden after completion, and includes, for example, pruning times and fertilization timings for each plant, and equipment inspection schedules. Maintenance plan information may also include information showing how to deal with individual problems such as sudden occurrences of pests and diseases or tree weakening.

[0087] Figure 7 shows an example of construction target information and planning information. In Figure 7, as construction target information, image data representing the condition of the construction target before landscaping is carried out is exemplified as "Before Construction". Similarly, as planning information, image data representing the condition of the construction target after landscaping is carried out is exemplified as "After Construction (After Installation of Landscape Stones)". As shown in Figure 7, planning information is not limited to images representing the condition of the construction target after all construction processes are completed, but may also be images representing the condition of the construction target during construction after each process is completed. Furthermore, images before and after construction, as shown in Figure 7, may be associated with each other and stored as landscaping knowledge information.

[0088] (Operation of the information provision system 1 according to the first embodiment) The information provision system according to the first embodiment operates as shown in the flowchart in Figure 4. In this embodiment, the case in which prompt information D3 is generated based on question information D1 while referring to reference information D2 has been described. However, if it is desired to quickly obtain answer information D4 from the generative artificial intelligence model 20, prompt information D3 may be generated based on question information D1 without referring to reference information D2.

[0089] Specifically, the prompt generation process (S200) involves a control unit 30 (processor) functioning as a prompt generation unit 301, which performs natural language processing as needed on the question information D1 (image data of the construction target and text data of the request information) received from the user terminal device 4. Next, the control unit 30 reads prompt syntax data 310 from the storage unit 31 (memory) and incorporates the content of the question information D1 and the reference information D2 (landscape knowledge information) obtained from the database device 5 into the selected syntax. In this way, the prompt generation process in this embodiment is a unique information processing method tailored to its intended use, in which hardware resources, namely the control unit 30 and the storage unit 31, cooperate to convert specific input data (question information D1) into specific structured output data (prompt information D3) for achieving the objective of the present invention (generation of an optimal landscape plan).

[0090] As described above, according to the information providing device 3 and information providing method of the first embodiment, when prompt information D3 generated based on question information D1 while referring to reference information D2 is input to the generative artificial intelligence model 20, answer information D4 is output from the generative artificial intelligence model 20. Therefore, an optimal landscape plan can be easily generated. Furthermore, when prompt information D3 is generated based on question information D1 without referring to reference information D2, answer information D4 can be efficiently obtained from the generative artificial intelligence model 20.

[0091] <Second Embodiment> Next, a second embodiment for carrying out the present invention will be described with reference to the above-mentioned common embodiment drawings. Note that detailed explanations of parts similar to the common embodiment or the first embodiment regarding the functions and connections of each component in the system, the various data handled by each component, the configuration and operation of each part in each device, the various data handled by each device, and the operation and configuration in the flowchart will be omitted.

[0092] (Summary of each piece of information according to the second embodiment) In the second embodiment, prompt information D3 refers to either first prompt information D3A or second prompt information D3B. Answer information D4 refers to either first answer information D4A or second answer information D4B. Reference information D2 refers to either first reference information D2A or second reference information D2B. First answer information D4A is process information that identifies the construction process required in landscaping, and first prompt information D3A is information that instructs the generative artificial intelligence model 20 to output first answer information D4A based on question information D1.

[0093] The second response information D4B is planning information that shows the plan for carrying out landscaping on the construction site, and the second prompt information D3B is information that instructs the generative artificial intelligence model 20 to output the second response information D4B based on the first response information D4A. Specifically, the second prompt information D3B instructs the generative artificial intelligence model 20 to integrally generate the final planning information by considering both the process information (list and sequence of construction processes) obtained as the first response information D4A and the original question information D1 (image data of the construction site and customer requests). This generates high-quality planning information that is both logically consistent and specific.

[0094] Question information D1 is construction target information. Since the construction target information is the same as the construction target information in the first embodiment, a detailed explanation is omitted (see paragraphs 0063-0066).

[0095] The first reference information D2A is landscape knowledge information, and in particular, construction process know-how information. Since the landscape knowledge information is the same as the landscape knowledge information in the first embodiment, a detailed explanation is omitted (see paragraphs 0067-0081).

[0096] The second reference information D2B is landscape knowledge information, and in particular, construction know-how information for each process. Since the landscape knowledge information is the same as the landscape knowledge information in the first embodiment, a detailed explanation is omitted (see paragraphs 0067-0081).

[0097] The first response information, D4A, is process information that identifies the construction processes required in landscaping. The process information may be in the form of a list of data (e.g., JSON format) in which the names of each process are arranged in the order of construction, or it may be in the form of graph data showing the dependencies between each process (e.g., "soil improvement" should be done before "planting"). It is not limited to these formats as long as it can be used in subsequent processing.

[0098] For example, if the generative artificial intelligence model 20 recognizes the presence of a large stone from the image data included in the question information D1 (construction target information), or if the request information includes the request for a "garden using stones," it will determine (identify) that the "landscape stone installation" process is necessary. In this way, the first response information D4A (process information) is generated as a result of identifying and organizing all the necessary work items from the given conditions without any excess or deficiency.

[0099] The second response information, D4B, is planning information. Since the planning information is identical to that in the first embodiment, a detailed explanation is omitted (see paragraphs 0082-0086).

[0100] (Operation of the information provision system 1 according to the second embodiment) The information provision system according to the second embodiment operates as shown in the flowchart in Figure 5. However, steps S220 and S230 may be omitted. That is, the first response information D4A may be treated as internal intermediate data of the information provision device 3 without being presented to the user U, and may be used directly in the subsequent second prompt generation process (step S240).

[0101] In this embodiment, the case in which the second prompt information D3B is generated based on the first answer information D4A while referring to the second reference information D2B has been described. However, the second prompt information D3B may also be generated based on the question information D1 and the first answer information D4A.

[0102] In this embodiment, the case in which the first prompt information D3A is generated based on the question information D1 while referring to the first reference information D2A has been described. However, if it is desired to quickly obtain the first answer information D4A from the generative artificial intelligence model 20, the first prompt information D3A may be generated based on the question information D1 without referring to the first reference information D2A. Also, in this embodiment, the case in which the second prompt information D3B is generated based on the first answer information D4A while referring to the second reference information D2B has been described. However, if it is desired to quickly obtain the second answer information D4B from the generative artificial intelligence model 20, the second prompt information D3B may be generated based on the first answer information D4A without referring to the second reference information D2B.

[0103] As described above, according to the information providing device 3 and information providing method of the second embodiment, when the first prompt information D3A generated based on the question information D1 while referring to the first reference information D2A is input to the generative artificial intelligence model 20, the first answer information D4A is output from the generative artificial intelligence model 20. When the second prompt information D3B generated based on the first answer information D4A while referring to the second reference information D2B is input to the generative artificial intelligence model 20, the second answer information D4B is output from the generative artificial intelligence model 20. Therefore, it is possible to easily generate an optimal landscape plan with higher accuracy. Furthermore, when generating the first prompt information D3A or the second prompt information D3B without referring to the reference information D2, the answer information D4 can be efficiently obtained from the generative artificial intelligence model 20.

[0104] <Other Embodiments> The present invention is not limited to the embodiments described above, and can be implemented with various modifications without departing from the spirit of the invention. All such modifications are included in the technical concept of the present invention.

[0105] In the above embodiment, the functions of each part of the information providing device 3 were described as being realized in a single device, but the functions of each part may be distributed among multiple devices to be realized in multiple devices. Alternatively, the control unit of the user terminal device 4 or the control unit of the database device 5 may execute the information processing program 311 to function as the information providing device 3.

[0106] In the above embodiment, the case in which the database device 5 includes a reference information management database 50 was described, but part or all of the reference information management database 50 may be stored in an external device (or more) connected to the network 6 or in any storage medium. Alternatively, the reference information management database 50 may be stored in the storage unit 31 of the information providing device 3.

[0107] In the above embodiment, the case in which the information providing device 3 operates according to the flowchart shown in Figure 4 has been described, but the execution order of each step may be changed as appropriate, some steps may be omitted, or other steps may be added. For example, when the prompt generation unit 301 generates prompt information D3 in the prompt generation process (step S200), a step may be added in which the prompt information D3 is provided to the user terminal device 4, and a user U input operation to change the prompt information D3 is received via the display screen of the user terminal device 4, and the prompt information D3 is changed.

[0108] In the above embodiment, the generative artificial intelligence model 20 generates response information D4 by referring to reference information D2 stored in the reference information management database 50 when prompt information D3 is input. In contrast, the generative artificial intelligence model 20 may use a trained model in which reference information D2 has been systematically incorporated by including reference information D2 as part of the training dataset. In that case, the generative artificial intelligence model 20 can generate response information D4 by referring to the reference information D2 incorporated into the generative artificial intelligence model 20 based on the instructions of prompt information D3.

[0109] The above embodiment describes the application of the present invention to the field of landscape gardening, but the scope of application of the present invention is not limited to this. For example, the present invention can also be applied to the field of ikebana (Japanese flower arrangement). In that case, the input information for the question includes "ikebana material information," which includes the flowers to be used, the vases, and image data showing the conditions of the space in which the work will be installed. The output information for the answer is "ikebana knowledge information," which is reference information that systematizes knowledge such as the styles of each school of ikebana and the combination of flowers to be used. This includes an image of the completed work and "ikebana plan information" showing the procedure for arranging the flowers. As a result, it becomes possible to easily create harmonious ikebana works even without specialized knowledge.

[0110] In the above embodiment, the case where the plan information, which is the response information, is displayed as a two-dimensional image or text on the display of the user terminal device 4 was described. In contrast, it is also possible to provide plan information using augmented reality (AR) technology by using smart glasses, a tablet, a smartphone, etc., as the user terminal device 4. In that case, when the user looks at the construction target (for example, an actual garden) through these devices, the generated completed image (for example, landscape stones and trees) and specific work instructions during construction (for example, markers indicating the angle at which landscape stones should be placed or branches to be pruned) are superimposed on the real landscape and displayed three-dimensionally. This allows the user to proceed with construction while intuitively and comprehensively checking the image after construction and the work content in a hands-free or near-hands-free state. It is also possible to instruct the user to make fine adjustments to the size and position of virtually placed objects on the spot. Furthermore, by linking with smart earphones, it is conceivable to provide voice guidance linked to the visual information from AR, or to output work instructions to a humanoid robot, etc., to assist or replace physical work.

[0111] In the above embodiment, the information providing device 3 was described as unilaterally providing planning information. However, the information providing system 1 may be configured to learn and evolve by utilizing user feedback. For example, after the user actually performs landscaping work based on the provided planning information, they provide feedback to the information providing device 3, such as photos of the completed work, an evaluation of the planning information, and points noticed during the work. The information providing device 3 stores the feedback as a new "construction example" in the reference information management database 50 and utilizes it for subsequent planning information generation. This allows the knowledge base of the information providing system 1 to be continuously expanded, and the accuracy and practicality of the generated planning information to improve over time.

[0112] The above embodiment mainly described the case where a single plan is generated. In contrast, multiple different plan information may be generated and proposed simultaneously in response to user requests. For example, the information providing device 3 provides multiple plan options based on different optimization criteria, such as a "cost-focused plan," a "design-focused plan," and a "low-maintenance plan." Users and customers can compare and examine the completed image, estimates, and construction procedures of each plan to select the plan that best matches their values ​​and constraints.

[0113] <Example of a prompt> An example of the first prompt information D3A in the second embodiment described above is shown below. Note that the prompt shown here is merely an example, and the role setting, the way data is provided, and the specific instructions can be changed as appropriate without departing from the spirit of the present invention. This prompt provides a specific means for stably obtaining a desired output by giving the generative artificial intelligence model 20 its role, available tools, and constraints.

[0114] role: You are a landscape gardening expert (master craftsman) with extensive knowledge and experience. Instructions: Analyze the following "customer requests" and "current condition of the area to be constructed" to identify the "construction process" required for the landscaping work. Then, rearrange the identified construction process in the optimal order to create the most efficient and high-quality garden. Customer requests: I would like a garden with a calm, traditional Japanese atmosphere. I want to create a small pond and a stream that flows into it. I like the existing Rhododendron molle, so I would like it to be kept. Current status of the construction target: [Attached file:site_photo_01.jpg] Constraints: Please provide only the names of the construction steps in your response, in list format. ・A detailed explanation of the specific tasks involved in each process is not required. Output format: ·JSON format

[0115] <Example of response information D4> An example of the second response information D4B in the second embodiment described above is shown below. Note that the response shown here is merely an example, and its specific expression and format will vary depending on the content of the input question information D1 and the referenced first reference information D2A and second reference information D2B; therefore, it is not limited to this example.

[0116] 1. Final image [Attached file: garden_finished_image.jpg] Based on the client's requests, we designed a quiet and tranquil Japanese-style garden. A small pond and stream are placed in the center, surrounded by landscape stones and plants, harmonizing with the existing Rhododendron molle. 2. Construction Procedure Information Step 1: Creating the ground's contours (land division and leveling) Instructions: Based on the attached file [layout_plan.pdf], set out the layout and determine the depth from the reference height. Excavate the pond and stream areas and create a slope that takes into account overall drainage. Step 2: Creating a flow (stream, waterfall, pond) Instructions: Lay a waterproof sheet over the excavated area and construct the stone wall for the revetment. Install the circulation pump and piping, and adjust the water volume and flow rate during a trial run. Process 3: Landscape stone placement Instructions: Use the three-pronged tool to set up the landscape stones according to the layout diagram in [Attached file:stone_arrangement.pdf]. Pay attention to the center of gravity and slope of the stones, and firmly secure the base stones. Process 4: Planting arrangement Instructions: Plant the main trees first, then arrange the ground cover and undergrowth. Make final adjustments on-site to ensure balance with the existing Rhododendron molle. Step 5: Pruning and maintenance (existing trees) Instructions: [Attachment: pruning_target_kurofune.jpg] Prune the unnecessary branches indicated by the red circles on the existing Kurofune azalea. Since this tree is sensitive to high temperatures, be sure to leave 20-30% more branches and leaves than other trees. 3. Material List Information • Waterproof sheet: 5m x 5m...1 sheet ·Kageishi (Kurama Stone): 2 large, 3 medium • Circulation pump: XX Company YY model...1 unit • Planting: 1 Japanese maple tree, 3 Aucuba japonica plants, 20 pots of Ophiopogon japonicus. 4. Estimated Cost Information Total amount: XXX,XXX yen (Breakdown) Material costs: YYY,YYY yen, Labor costs: ZZZ,ZZZ yen 5. Maintenance Plan Information Rhododendron molle: Prune unnecessary branches every year after the leaves have fallen. We recommend taking measures to protect it from strong direct sunlight in the summer. Ponds and streams: The circulation pump filter needs to be cleaned regularly.

[0117] While several embodiments of this disclosure have been illustrated above, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents. Furthermore, the embodiments described above can be implemented in combination with each other. [Explanation of symbols]

[0118] 1... Information provision system, 2... Generative AI device, 3... Information provision device, 4... User terminal device, 5...Database device, 20...Generative artificial intelligence model, 30...Control unit, 31...Storage unit, 32...Communication unit, 33...Input unit, 34...Display unit, 50...Reference information management database, 300...Question reception unit, 301...Prompt generation unit, 302...Answer provision unit, 310... Prompt syntax data, 311... Information processing program

Claims

1. A method of providing information using a computer, The reception process for receiving question information, A first prompt generation process generates first prompt information that instructs a generative artificial intelligence model to output first answer information based on the aforementioned question information, A first response provision process that provides the first response information output from the generative artificial intelligence model by inputting the first prompt information into the generative artificial intelligence model, A second prompt generation process generates second prompt information that instructs a generative artificial intelligence model to output second response information based on the first response information, The process involves inputting the second prompt information into the generative artificial intelligence model to provide the second response information output from the generative artificial intelligence model, and then performing a second response provision process. The aforementioned question information is, This is information regarding the scope of work for landscaping projects. The above first answer information is, This is process information that identifies the multiple construction processes required in the aforementioned landscaping in the order of execution. The aforementioned second response information is, For each of the construction processes identified by the aforementioned process information, the planning information indicates a plan for carrying out the landscaping on the construction target. Information provision method.

2. An information providing device that provides information using a computer, The reception desk accepts questions and information, A first prompt generation unit generates first prompt information that instructs a generative artificial intelligence model to output first answer information based on the aforementioned question information, A first response providing unit that provides the first response information output from the generative artificial intelligence model by inputting the first prompt information to the generative artificial intelligence model, A second prompt generation unit generates second prompt information that instructs a generative artificial intelligence model to output second response information based on the first response information, The system includes a second response providing unit that provides the second response information output from the generative artificial intelligence model by inputting the second prompt information to the generative artificial intelligence model, The aforementioned question information is, This is information regarding the scope of work for landscaping projects. The above first answer information is, This is process information that identifies the multiple construction processes required in the aforementioned landscaping in the order of execution. The aforementioned second response information is, For each of the construction processes identified by the aforementioned process information, the planning information indicates a plan for carrying out the landscaping on the construction target. Information provision device.

Images