Treatment explanation production system, and program of treatment explanation production system

Abstract

To provide a treatment explanation production system in which a user can use an uploaded program while referring to a treatment explanation file.SOLUTION: A treatment explanation production system 100 comprises a user terminal 110A, a first server 120 and a second server 130. When any program data PG is uploaded to the first server from the user terminal, the first server creates ABI information ST which is specification and standard information on the basis of the program data. The first server transmits command information PT for generating the program data, the ABI information and treatment explanation to the second server. The second server creates a treatment explanation file ML of a program by using previously learned artificial intelligence means 131 to transmit it to the first server, on the basis of the command information for generating the program data, the ABI information and the treatment explanation. The first server associates the treatment explanation file ML with the program data PG to register it with a database 121.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an instruction manual generating system that has a user terminal and a server and generates an instruction manual file for any program, and to a program for carrying out the same. [Background technology]

[0002] 2. Description of the Related Art Conventionally, there is known an operation display system configured with an external server and a terminal device, in which the external server includes an upload unit, a processing unit, a program storage unit, and a transmission / reception unit (for example, Patent Document 1). The upload unit had a function of uploading the operation program received by the transmitting / receiving unit and saving (memorizing) it in the program storage unit. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2019-99155 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, the conventional operation display system described above was configured only to upload the operation program to an external server, so when other users wanted to use the operation program, it was difficult for them to understand the structure and contents of the program, and they did not know how to use it.

[0005] Therefore, the present invention is intended to solve the problems of the prior art as described above, that is, the object of the present invention is to provide an instruction manual generation system and a program therefor that allow a user to use an uploaded program while referring to an instruction manual file. [Means for solving the problem]

[0006] The invention of claim 1 is an instruction manual generation system which has a user terminal and a server and generates an instruction manual file for any program, and when any program data is uploaded from the user terminal to a first server, the first server generates ABI information, which is specification and standard information that guarantees program compatibility at a binary level, based on the program data, and the first server transmits the program data, ABI information, and command information to generate an instruction manual to a second server different from the first server, and the second server generates an instruction manual file for the program using a pre-trained artificial intelligence means based on the received program data, ABI information, and command information to generate an instruction manual and transmits it to the first server, and the first server associates the received instruction manual file for the program with the program data and registers it in a database, thereby solving the above-mentioned problem.

[0007] The invention according to claim 2 is an instruction manual generation system having a user terminal and a server for generating an instruction manual file for an arbitrary program, wherein when arbitrary program data is uploaded from the user terminal to a first server, the first server generates ABI information, which is specification / standard information that guarantees program compatibility at a binary level, based on the program data, and divides the program data and the ABI information so as to be separated by functions, which are the contents of the program data, and stores the divided program division data and ABI division information in a database, and the first server sequentially transmits the divided program division data and ABI division information to a second server different from the first server, and the second server converts the sequentially received program division data and ABI division information into vector coordinate system program division data and vector coordinate system ABI division information, which are vector coordinate system data, using vector coordinate conversion means, which is a pre-trained artificial intelligence means, and sequentially transmits the vector coordinate system program division data and vector coordinate system ABI division information to the first server. the first server records the vector coordinate system program partition data and vector coordinate system ABI partition information sequentially received in a spatial map and stores the data in a database in association with the program partition data and ABI partition information; the first server sequentially transmits to a second server command information to generate instruction manuals corresponding to the program partition data and ABI partition information; the second server converts the sequentially received instruction information to generate instruction manuals into vector coordinate system command information, which is vector coordinate system data, using a vector coordinate conversion means which is a pre-trained artificial intelligence means, and sequentially transmits the vector coordinate system command information to the first server; the first server refers to the spatial map based on the sequentially received vector coordinate system command information, selects program partition data and ABI partition information corresponding to the vector coordinate system program partition data and vector coordinate system ABI partition information that are most closely related to the vector of the vector coordinate system command information, and sequentially transmits to a third server different from the first server the program partition data, ABI partition information, and command information to generate instruction manuals corresponding to these; the third serverBased on the sequentially received program division data, ABI division information, and command information for generating corresponding instruction manuals, partial instruction manual files for the program division data are generated using a text generation means, which is a pre-trained artificial intelligence means, and are sequentially transmitted to a first server, and the first server synthesizes the sequentially received partial instruction manual files into one overall instruction manual file, and registers the synthesized overall instruction manual file in a database in association with the program data, thereby solving the above-mentioned problem.

[0008] The invention of claim 3 further solves the above-mentioned problem by being configured such that, in addition to the configuration of the instruction manual generation system described in claim 2, when the first server sends command information to the second server to generate instruction manuals corresponding to the program partition data and ABI partition information, the first server determines which of the functions of the program partition data and ABI partition information is an initialization function, a function function, or a recording function, and based on this determination, selects command information to generate instruction manuals for each template prepared in advance in accordance with the initialization function, function function, or recording function, which are the functions of the program partition data and ABI partition information, and sends command information to the second server to generate the selected instruction manuals.

[0009] The invention of claim 4 further solves the above-mentioned problem by being configured in such a way that, in addition to the configuration of the instruction manual generation system described in claim 2 or claim 3, when any program data is uploaded from the user terminal to a first server, the first server accesses a fourth server different from the first server based on reference information input from the user terminal or the contents of the uploaded program data or ABI information to obtain related information related to the contents of the program data or ABI information, and adds it to command information to generate instruction instructions.

[0010] The invention of claim 5 further solves the above-mentioned problem by being configured such that, in addition to the configuration of the instruction manual generation system described in claim 1, the first server determines whether the capacity of the program data is equal to or greater than a predetermined capacity before transmitting the program data, ABI information, and command information to generate instruction manuals to the second server, and if the capacity is equal to or greater than the predetermined capacity, the first server divides the program data into multiple parts and transmits the first program part data containing the first part of the entire program data to the second server as program data.

[0011] The invention of claim 6 further solves the above-mentioned problem by having, in addition to the configuration of the instruction manual generation system described in claim 5, a configuration in which, when the first server receives a program instruction manual file from the second server, it inserts subsequent program portion data that was not divided and sent to the second server into the received program instruction manual file to create a supplemental instruction manual file, and associates the supplemental instruction manual file with the program data and registers it in a database as an instruction manual file.

[0012] The invention of claim 7 is a program for an instruction manual generation system for generating an instruction manual file of an arbitrary program, which includes an upload determination step for determining whether arbitrary program data has been uploaded from a user terminal to a first server, an ABI information generation step for the first server to generate ABI information, which is specification and standard information that guarantees program compatibility at a binary level based on the program data when the data has been uploaded, a prompt transmission step for the first server to transmit the program data, ABI information, and command information for generating instruction manuals to a second server different from the first server, an instruction manual generation step for the second server to generate an instruction manual file for the program using a pre-trained artificial intelligence means based on the received program data, ABI information, and command information for generating instruction manuals, an instruction manual transmission step for the second server to transmit the generated instruction manual file for the program to the first server, and an instruction manual registration step for the first server to register the received instruction manual file for the program in a database in association with program data.

[0013] The invention of claim 8 is a program for an instruction manual generation system for generating an instruction manual file for an arbitrary program, comprising: an upload determination step for determining whether arbitrary program data has been uploaded from a user terminal to a first server, which is a server; an ABI information generation step in which the first server generates ABI information, which is specification / standard information that guarantees program compatibility at a binary level, based on the program data; a division / storage step in which the first server divides the program data and the ABI information so as to separate them by functions, which are the contents of the program data, and stores the divided program division data and the ABI division information in a database; a division information transmission step in which the first server sequentially transmits the divided program division data and the ABI division information to a second server different from the first server; and a division information transmission step in which the second server sequentially converts the program division data and ABI division information received by the second server into vector coordinate system program division data and vector coordinate system program division data, which are vector coordinate system data, using a vector coordinate conversion means, which is a pre-trained artificial intelligence means. a vector coordinate system information conversion step of converting the vector coordinate system program partition data and the vector coordinate system ABI partition information received by the second server into vector coordinate system ABI partition information; a vector coordinate system information transmission step of the second server sequentially transmitting the vector coordinate system program partition data and the vector coordinate system ABI partition information to the first server; a map recording and storage step of the first server recording the vector coordinate system program partition data and the vector coordinate system ABI partition information received by the first server in a spatial map and storing the data in a database in association with the program partition data and the ABI partition information; a first prompt transmission step of the first server sequentially transmitting to the second server command information for generating an instruction manual corresponding to the program partition data and the ABI partition information; a prompt vector coordinate system information conversion step of the second server converting the sequentially received command information for generating an instruction manual into vector coordinate system command information which is vector coordinate system data using a vector coordinate conversion means which is a pre-trained artificial intelligence means; a vector coordinate system prompt transmission step of the second server sequentially transmitting the vector coordinate system command information to the first server;The above-mentioned problem is solved by the second prompt transmission step of selecting program partition data and ABI partition information corresponding to the vector coordinate system program partition data and vector coordinate system ABI partition information that are most closely related to the vector of the vector coordinate system command information by referring to a space map based on the sequentially received vector coordinate system command information, and sequentially transmitting the program partition data, ABI partition information, and command information for generating the corresponding instruction manual to a third server different from the first server; an instruction manual generation step of the third server generating a partial instruction manual file for the program partition data using a text generation means which is a pre-trained artificial intelligence means based on the sequentially received program partition data, ABI partition information, and command information for generating the corresponding instruction manual; an instruction manual transmission step of the third server sequentially transmitting the generated partial instruction manual files to the first server; and an instruction manual synthesis / registration step of the first server synthesizing the sequentially received partial instruction manual files into one overall instruction manual file and registering the synthesized overall instruction manual file in a database in association with program data. Effect of the Invention

[0014] The instruction manual generating system of the present invention, having a user terminal and a server, not only enables communication between the user terminal and the server, but also provides the following unique effects.

[0015] According to the instruction manual generation system of the invention of claim 1, after a first user uploads any program data from a user terminal, the program data and the instruction manual file for that program are registered as a set on the first server, so that a second user can use the program while referring to the instruction manual file. Furthermore, since the pre-trained artificial intelligence means refers to not only program data but also ABI information based on a prompt from command information to generate instruction manuals, a high-quality program instruction manual file that is easy to understand for the second user who uses the program instruction file can be provided, compared to a configuration in which the pre-trained artificial intelligence means generates a program instruction manual file without referring to ABI information.

[0016] According to the instruction manual generation system of the invention of claim 2, similar to the effect achieved by the invention of claim 1, after a first user uploads any program data from his / her user terminal, the program data and the instruction manual file for that program are registered as a set on the first server, so that a second user can use the program while referring to the instruction manual file. Furthermore, since the pre-trained artificial intelligence means refers to not only program data but also ABI information based on a prompt from command information to generate instruction manuals, a high-quality program instruction manual file that is easy to understand for the second user who uses the program instruction file can be provided, compared to a configuration in which the pre-trained artificial intelligence means generates a program instruction manual file without referring to ABI information. In addition, the program data and ABI information are respectively divided so that they are separated by the functions / functionality which are the contents of the program data, the divided program division data and ABI division information are converted into vector coordinate system program division data and vector coordinate system ABI division information, the vector coordinate system program division data and vector coordinate system ABI division information are recorded in a spatial map, the program division data and ABI division information corresponding to the vector coordinate system program division data and vector coordinate system ABI division information which are most closely related to the vector of the vector coordinate system command information are selected, and partial instruction manual files for the program division data are generated by a text generation means which is a pre-trained artificial intelligence means based on command information to generate instruction manuals corresponding to these, so that partial instruction manual files with high content accuracy can be obtained for each function / functionality. Since a plurality of partial instruction manual files are synthesized into one overall instruction manual file, it is possible to obtain an overall instruction manual file with good overall content accuracy.

[0017] According to the instruction manual generation system of the invention of claim 3, in addition to the effects achieved by the invention of claim 2, command information is selected to generate instruction manuals corresponding to the functions of program partition data and ABI partition information that are each divided so as to be separated by function / functionality, thereby further improving the accuracy when the first server refers to a spatial map based on the vector coordinate system command information received sequentially and selects program partition data and ABI partition information that correspond to the vector coordinate system program partition data and vector coordinate system ABI partition information that are most closely related to the vector of the vector coordinate system command information. Furthermore, since the accuracy in selecting program division data and ABI division information is improved, the accuracy of the content can be improved when the third server generates a partial instruction manual file for the program division data using a text generation means, which is a pre-trained artificial intelligence means, based on the program division data, ABI division information, and command information for generating corresponding instruction manuals that are received sequentially.

[0018] According to the instruction manual generation system of the invention of claim 4, in addition to the effects of the invention of claim 2 or claim 3, relevant information related to the contents of the program data or ABI information is added to the command information for generating instruction manuals, thereby making it possible to further improve the accuracy of the content of the command information for generating instruction manuals. Furthermore, since the content accuracy of the command information to generate instruction manuals is improved, the content accuracy can be further improved when the third server generates a partial instruction manual file for the program division data using a text generation means, which is a pre-trained artificial intelligence means, based on the program division data, ABI division information, and command information to generate instruction manuals corresponding to these that are received sequentially.

[0019] According to the instruction manual generation system of the invention of claim 5, in addition to the effects achieved by the invention of claim 1, when the second server generates an instruction manual file, the ABI information is used to supplement the information other than the beginning, so that even if the pre-trained artificial intelligence means has input capacity limitations, an instruction manual file for a program including information other than the beginning can be generated.

[0020] According to the instruction manual generation system of the invention of claim 6, in addition to the effect of the invention of claim 5, the supplementary instruction manual file contains the entire program data before division, making it easier for a second user of the program to understand.

[0021] According to the program of the instruction manual generation system of the invention of claim 7, similar to the effect achieved by the invention of claim 1, after a first user uploads any program data from his / her user terminal, the program data and the instruction manual file for that program are registered as a set on the first server, so that a second user can use the program while referring to the instruction manual file. Furthermore, since the pre-trained artificial intelligence means refers to not only program data but also ABI information based on a prompt from command information to generate instruction manuals, a high-quality program instruction manual file that is easy to understand for the second user who uses the program instruction file can be provided, compared to a configuration in which the pre-trained artificial intelligence means generates a program instruction manual file without referring to ABI information.

[0022] According to the program of the instruction manual generation system of the invention of claim 8, similar to the effects achieved by the inventions of claims 1 and 2, after a first user uploads any program data from his / her user terminal, the program data and the instruction manual file for that program are registered as a set on the first server, so that a second user can use the program while referring to the instruction manual file. Furthermore, since the pre-trained artificial intelligence means refers to not only program data but also ABI information based on a prompt from command information to generate instruction manuals, a high-quality program instruction manual file that is easy to understand for the second user who uses the program instruction file can be provided, compared to a configuration in which the pre-trained artificial intelligence means generates a program instruction manual file without referring to ABI information. In addition, the program data and ABI information are respectively divided so that they are separated by the functions / functionality which are the contents of the program data, the divided program division data and ABI division information are converted into vector coordinate system program division data and vector coordinate system ABI division information, the vector coordinate system program division data and vector coordinate system ABI division information are recorded in a spatial map, the program division data and ABI division information corresponding to the vector coordinate system program division data and vector coordinate system ABI division information which are most closely related to the vector of the vector coordinate system command information are selected, and partial instruction manual files for the program division data are generated by a text generation means which is a pre-trained artificial intelligence means based on command information to generate instruction manuals corresponding to these, so that partial instruction manual files with high content accuracy can be obtained for each function / functionality. Since a plurality of partial instruction manual files are synthesized into one overall instruction manual file, it is possible to obtain an overall instruction manual file with good overall content accuracy. [Brief description of the drawings]

[0023] [Figure 1] 1 is a diagram showing the concept of an instruction manual generating system according to a first embodiment of the present invention; [Diagram 2] FIG. 2 is a chart showing an example of the operation of the instruction manual generating system according to the first embodiment of the present invention. [Diagram 3] 1A and 1B are explanatory diagrams of the concept of dividing program data in the instruction manual generation system according to the first embodiment of the present invention, and an explanatory diagram of the first program portion data of the divided program data being sent together with a prompt to a second server. [Figure 4] 1A and 1B are explanatory diagrams for creating a supplemented instruction manual file by inserting the instruction manual file in the first server of the instruction manual creating system according to the first embodiment of the present invention. [Diagram 5] FIG. 13 is a diagram showing the concept of an instruction manual generating system according to a second embodiment of the present invention. [Figure 6]FIG. 6 is a chart showing an example of the operation of the instruction manual generating system according to the second embodiment of the present invention. [Figure 7] FIG. 11 is a diagram showing the concept of data exchange in the instruction manual generating system according to the second embodiment of the present invention. [Figure 8] FIG. 11 is a diagram showing an example of command information for generating an instruction manual of the instruction manual generating system according to the second embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0024] The instruction manual generation system of the present invention has a user terminal and a server, and when any program data is uploaded from the user terminal to a first server, the first server generates ABI information, which is specification and standard information that guarantees program compatibility at a binary level, based on the program data, and the first server transmits the program data, ABI information, and command information to generate the instruction manual to a second server different from the first server, and the second server generates an instruction manual file for the program using a pre-trained artificial intelligence means based on the received program data, ABI information, and command information to generate the instruction manual, and transmits it to the first server, and the first server registers the received instruction manual file for the program in association with the program data in a database. Therefore, after any program data is uploaded from the user terminal by the first user, the second user can use the program while referring to the instruction manual file on the user terminal, and any specific embodiment may be used as long as it is possible to provide a high-quality program instruction manual file that is easy to understand for the second user who uses the program instruction file, compared to a configuration in which the pre-trained artificial intelligence means generates an instruction manual file for the program without referring to the ABI information. The program of the instruction manual generating system of the present invention includes an upload determination step of determining whether any program data has been uploaded from a user terminal to a first server, an ABI information generation step in which the first server generates ABI information, which is specification and standard information that guarantees program compatibility at a binary level, based on the program data when the data is uploaded, a prompt transmission step in which the first server transmits the program data, the ABI information, and instruction manual generation command information to a second server different from the first server, an instruction manual generation step in which the second server generates an instruction manual file for the program using pre-trained artificial intelligence means based on the received program data, ABI information, and instruction manual generation command information, and The specific implementation may be any as long as it has an instruction manual sending step of sending the RAM's instruction manual file to a first server, and an instruction manual registration step of the first server associating the received program's instruction manual file with the program data and registering it in a database, so that after a first user uploads any program data from his / her user terminal, a second user can use the program while referring to the instruction manual file on the user terminal, and a good-quality program instruction manual file that is easy to understand for the second user who uses the program's instruction manual file can be provided, compared to a configuration in which the pre-trained artificial intelligence means generates a program's instruction manual file without referring to ABI information.

[0025] For example, each of the first server and the second server may be a single server or multiple servers on a cloud. The pre-trained artificial intelligence means may be composed of an embedding model, which is an example of a vector coordinate conversion means that converts input data into data in a vector coordinate system, or an interactive text generation means that generates text data and is also called a large-scale language model, such as ChatGPT (Generative Pre-trained Transformer) (hereinafter referred to as ChatGPT), and may be configured to be installed on one server or multiple servers on the cloud. The user terminal may be a desktop personal computer terminal, a notebook personal computer terminal, a smartphone terminal, a tablet terminal, or the like that sends and receives information, and may be any terminal that can be freely connected to a server via a communication network, including a wide area network such as the Internet, a local network, or a telephone line. The upload operation by the (first) user is performed at a user terminal, the first user terminal. The use (viewing) of the program's instruction manual file by the (second) user is carried out on the second user terminal, which is a user terminal. EXAMPLES

[0026] An instruction manual generating system 100 according to a first embodiment of the present invention will be described below with reference to FIGS. 1 to 4(B). Here, Figure 1 is a diagram showing the concept of the instruction manual generating system 100 which is the first embodiment of the present invention, Figure 2 is a chart diagram showing an example of the operation of the instruction manual generating system 100 which is the first embodiment of the present invention, Figure 3 (A) is an explanatory diagram of the concept of dividing the program data PG of the instruction manual generating system 100 which is the first embodiment of the present invention, Figure 3 (B) is an explanatory diagram of sending the first program portion data PG1 of the divided program data PG together with a prompt to the second server 130, Figure 4 (A) is an explanatory diagram of an example of an instruction manual file ML generated in the second server 130 of the instruction manual generating system 100 which is the first embodiment of the present invention based on the first program portion data PG1, ABI information ST and command information PT for generating instruction manuals, and Figure 4 (B) is an explanatory diagram of inserting into the instruction manual file ML in the first server 120 of the instruction manual generating system 100 which is the first embodiment of the present invention to create a supplemented instruction manual file ML1.

[0027] As shown in FIG. 1, the first embodiment of the present invention, an instruction manual generation system 100, includes a first user terminal 110A and a second user terminal 110B as examples of user terminals, a first server 120, and a second server 130. Among these, the second server 130 has chat GPT, which is an example of a text generation means 131a also called a large-scale language model as the pre-trained artificial intelligence means 131. The instruction manual generating system 100 is configured to generate an instruction manual file ML for a program by using chat GPT, which is an example of the text generating means 131a. More specifically, a first user uses a first user terminal 110A to upload any program data PG (also called a smart contract), such as a program created by the first user, to a first server 120. Here, the upload may be based on an operation by the first user, or may be performed by an automatic program or the like. Then, when any program data PG is uploaded from the first user terminal 110A to the first server 120, the first server 120 generates ABI information ST (Application Binary Interface information), which is specification and standard information that guarantees program compatibility at the binary level based on the program data PG.

[0028] Then, the first server 120 transmits to the second server 130 the program data PG, the ABI information ST, and a prompt (also called a script) which is command information PT for generating an instruction manual. Then, the second server 130 generates a program instruction manual file ML (also called a structured document) using chat GPT, which is an example of a text generation means 131a as the pre-trained artificial intelligence means 131, based on the received program data PG, ABI information ST, and command information PT to generate instruction manuals. Then, the second server 130 transmits the generated instruction manual file ML to the first server 120. Then, the first server 120 is configured to register the received program instruction manual file ML in the database 121 in association with the program data PG.

[0029] As a result, after any program data PG is uploaded from first user terminal 110A, the program data PG and the instruction manual file ML for that program are registered as a set in first server 120. As a result, the second user can use the program while referring to the instruction manual file ML on the second user terminal 110B. Furthermore, chat GPT, which is an example of text generation means 131a as the pre-trained artificial intelligence means 131, refers to not only the program data PG but also the ABI information ST based on a prompt of command information PT to generate an instruction manual. As a result, compared to a configuration in which chat GPT, which is an example of a text generation means 131a as a pre-trained artificial intelligence means 131, generates a program instruction file ML without referring to the ABI information ST, it is possible to provide a high-quality program instruction file ML that is easy to understand for the second user who uses the program instruction file ML.

[0030] Next, an example of the operation of the instruction manual generating system 100 will be described in more detail. As shown in FIG. 2, in step S1, as an upload determination step, the first server 120 determines whether any program data PG has been uploaded to the first server 120 from a first user terminal 110A as an example of a user terminal. If it is determined that the data has been uploaded, the process proceeds to step S2. On the other hand, if it is determined that the data has not yet been uploaded, step S1 is repeated.

[0031] In step S2, as an ABI information generating step, the first server 120 generates ABI information ST, which is specification and standard information that guarantees program compatibility at the binary level, based on the program data PG. In step S3, which is a prompt transmission step, the first server 120 transmits to the second server 130 a prompt including the program data PG, the ABI information ST, and command information PT for generating an instruction manual.

[0032] In step S4, as an instruction manual generation step, the second server 130 generates an instruction manual file ML for the program using chat GPT, which is an example of a text generation means 131a as the pre-trained artificial intelligence means 131, based on the received program data PG, ABI information ST, and a prompt from the command information PT to generate instruction manuals. In step S5, as an instruction manual sending step, second server 130 sends the generated instruction manual file ML of the program to first server 120. In step S6, as an instruction manual registration step, first server 120 registers the received instruction manual file ML of the program in database 121 in association with program data PG.

[0033] As a result, after any program data PG is uploaded from first user terminal 110A, the program data PG and the instruction manual file ML for that program are registered as a set in first server 120, as described above. As a result, the second user can use the program while referring to the instruction manual file ML on the second user terminal 110B. Furthermore, chat GPT, which is an example of text generation means 131a as the pre-trained artificial intelligence means 131, refers to not only the program data PG but also the ABI information ST based on a prompt of command information PT to generate an instruction manual. As a result, compared to a configuration in which chat GPT, which is an example of a text generation means 131a as a pre-trained artificial intelligence means 131, generates a program instruction file ML without referring to the ABI information ST, it is possible to provide a high-quality program instruction file ML that is easy to understand for the second user who uses the program instruction file ML.

[0034] Furthermore, in this embodiment, before the first server 120 in step S3 transmits the program data PG, ABI information ST, and command information PT to generate instruction manuals to the second server 130, the first server 120 may determine whether the capacity of the program data PG is equal to or greater than a predetermined capacity. Then, as shown in FIG. 3(A), when it is determined that the capacity is equal to or greater than a predetermined capacity, the first server 120 divides the program data PG into a plurality of parts. For example, a program file whose program data PG has a capacity equal to or larger than a predetermined capacity is divided into four files, that is, file No. 1, file No. 2, file No. 3, and file No. 4. Then, as shown in FIG. 3(B), a file No. 1 which is head program portion data PG1 including the head portion of the entire program data PG may be transmitted to the second server 130 as the program data PG.

[0035] As a result, as shown in FIG. 4(A), when the second server 130 generates the instruction manual file ML, the information of the portions other than the beginning, that is, file No. 2, file No. 3, and file No. 4, is supplemented by the ABI information ST. As a result, even if there is an input capacity limit in the chat GPT, which is an example of a text generation means 131a as a pre-trained artificial intelligence means 131, it is possible to generate a program instruction manual file ML that includes information on the parts other than the beginning, namely file No. 2, file No. 3, and file No. 4.

[0036] Also in this embodiment, when the first server 120 receives the program instruction manual file ML from the second server 130 in steps S5 to S6, the information of the parts of file No. 2, file No. 3, and file No. 4, which are the subsequent program partial data PG2 that was not divided from the above-mentioned program data PG and sent to the second server 130, is inserted into the received program instruction manual file ML as shown in Figure 4 (B) to create a supplemented version instruction manual file ML1. The first server 120 may then be configured to register the supplementary instruction manual file ML1 in the database 121 as an instruction manual file ML in association with the program data PG.

[0037] As a result, even if the program data PG is divided into several parts and only file No. 1, which is the first program part data PG1 containing the first part of the entire program data PG, is sent to the second server 130 as program data PG, and information on the parts other than the first part, that is, file No. 2, file No. 3, and file No. 4, is not sent, the supplementary instruction manual file ML1 will contain information on files No. 1 to No. 4, which are the entire program data PG before division. As a result, the program can be more easily understood by a second user.

[0038] The instruction manual generating system 100, which is the first embodiment of the present invention obtained in this manner, has a first user terminal 110A and a second user terminal 110B as user terminals, a first server 120, and a second server 130. When any program data PG is uploaded from the first user terminal 110A to the first server 120, the first server 120 generates ABI information ST, which is specification and standard information that guarantees program compatibility at a binary level, based on the program data PG. The first server 120 transmits the program data PG, the ABI information ST, and command information PT to generate instruction manuals to the second server 130. The second server 130 generates text data, which is an example of a text generation means 131a serving as a pre-trained artificial intelligence means 131, based on the received program data PG, the ABI information ST, and the command information PT to generate instruction manuals. The configuration is such that a program instruction file ML is generated using chat GPT and sent to the first server 120, and the first server 120 associates the received program instruction file ML with the program data PG and registers it in the database 121. Therefore, after any program data PG is uploaded from the first user terminal 110A, the second user can use the program while referring to the instruction file ML on the second user terminal 110B. Furthermore, compared to a configuration in which chat GPT, which is an example of a text generation means 131a as the pre-trained artificial intelligence means 131, generates a program instruction file ML without referring to the ABI information ST, it is possible to provide a high-quality program instruction file ML that is easy for the second user who uses the program instruction file ML to understand.

[0039] Furthermore, before the first server 120 sends the program data PG, ABI information ST, and command information PT to generate instruction manuals to the second server 130, it determines whether the capacity of the program data PG is equal to or greater than a predetermined capacity, and if it is equal to or greater than the predetermined capacity, the first server 120 divides the program data PG into multiple parts and sends the first program part data PG1 containing the first part of the entire program data PG to the second server 130 as the program data PG.This makes it possible to generate an instruction manual file ML for the program that contains information on parts other than the first part, even if there is an input capacity limit in the chat GPT, which is an example of a text generation means 131a as the pre-trained artificial intelligence means 131.

[0040] In addition, when the first server 120 receives a program instruction manual file ML from the second server 130, the subsequent program portion data PG2 that was not divided from the program data PG and sent to the second server 130 is inserted into the received program instruction manual file ML to create a supplemental instruction manual file ML1, and the supplemental instruction manual file ML1 is associated with the program data PG as an instruction manual file ML and registered in the database 121, thereby making it easier for the second user who uses the program to understand.

[0041] Furthermore, the program of the instruction manual generation system 100 according to the first embodiment of the present invention includes an upload determination step S1 for determining whether any program data PG has been uploaded to the first server 120 from the first user terminal 110A as a user terminal; an ABI information generation step S2 for the first server 120 to generate ABI information ST, which is specification and standard information that guarantees program compatibility at a binary level, based on the program data PG when the data has been uploaded; a prompt transmission step S3 for the first server 120 to transmit to the second server 130 the program data PG, the ABI information ST, and command information PT for generating instruction manuals; and an instruction manual generation step S4 for the second server 130 to generate an instruction manual file ML for the program using a chat GPT, which is an example of a text generation means 131a as the pre-trained artificial intelligence means 131, based on the received program data PG, the ABI information ST, and the command information PT for generating instruction manuals. Having step S4, an instruction manual sending step S5 in which the second server 130 sends the generated program instruction file ML to the first server 120, and an instruction manual registration step S6 in which the first server 120 associates the received program instruction file ML with the program data PG and registers it in the database 121, after any program data PG is uploaded from the first user terminal 110A, the second user can use the program while referring to the instruction manual file ML on the second user terminal 110B.Furthermore, compared to a configuration in which chat GPT, an example of a text generation means 131a as the pre-trained artificial intelligence means 131, generates a program instruction file ML without referring to the ABI information ST, it is possible to provide a high-quality program instruction file ML that is easy to understand for the second user who uses the program instruction file ML, and the effects are enormous. EXAMPLES

[0042] A system 200 for generating instructions according to a second embodiment of the present invention will now be described with reference to FIGS. Here, Figure 5 is a diagram showing the concept of instruction manual generating system 200 which is the second embodiment of the present invention, Figure 6 is a chart showing an example of operation of instruction manual generating system 200 which is the second embodiment of the present invention, Figure 7 is a diagram showing the concept of data exchange in instruction manual generating system 200 which is the second embodiment of the present invention, and Figure 8 is a diagram showing an example of command information PT for generating instruction manuals in instruction manual generating system 200 which is the second embodiment of the present invention. The second embodiment of the instruction manual generating system 200 replaces the pre-trained artificial intelligence means 131 of the second server 130 of the instruction manual generating system 100 of the first embodiment with two pre-trained artificial intelligence means 231 of the second server 230 and pre-trained artificial intelligence means 241 of the third server 240, and increases the amount of data exchange.Since many elements are common to the instruction manual generating system 100 of the first embodiment, detailed explanations of the common points will be omitted and only symbols in the 200 series with the last two digits being common will be used.

[0043] The second embodiment of the present invention, an instruction manual generation system 200, includes a first user terminal 210A and a second user terminal 210B as examples of user terminals, a first server 220, a second server 230, and a third server 240, as shown in FIG. 5. Of these, the second server 230 has vector coordinate conversion means 231b as the pre-trained artificial intelligence means 231.

[0044] The vector coordinate conversion means 231b is also called an embedding model and is provided to convert input data into data in a vector coordinate system. The third server 240 also has chat GPT, which is an example of a text generation means 241a also called a large-scale language model, as the pre-trained artificial intelligence means 241. As a technical concept, third server 240 only needs to be different from first server 220, and may have a configuration that is physically common to second server 230, or may have a physically different configuration.

[0045] In this embodiment, when any program data PG is uploaded from a user terminal to the first server 220, the first server 220 generates ABI information ST, which is specification and standard information that guarantees program compatibility at the binary level, based on the program data PG. At the same time, the first server 220 divides the program data PG and the ABI information ST so as to separate them according to the functions that are the contents of the program data PG, and stores the divided program division data DPG and ABI division information DST in the database 221.

[0046] Then, the first server 220 sequentially transmits the divided program division data DPG and the ABI division information DST to a second server 230 different from the first server 220. Then, the second server 230 converts the sequentially received program partition data DPG and ABI partition information DST into vector coordinate system program partition data VPG and vector coordinate system ABI partition information VST, which are vector coordinate system data, using an Embedding model, which is an example of a vector coordinate conversion means 231b, which is the pre-trained artificial intelligence means 231. Then, the second server 230 sequentially transmits the vector coordinate system program partition data VPG and the vector coordinate system ABI partition information VST to the first server 220.

[0047] Then, the first server 220 records the vector coordinate system program partition data VPG and the vector coordinate system ABI partition information VST that are sequentially received in the spatial map. At the same time, the first server 220 stores the program partition data DPG and the ABI partition information DST in the database 221 in association with each other. That is, the vector coordinate system program partition data VPG and the vector coordinate system ABI partition information VST are tagged with the program partition data DPG and the ABI partition information DST and saved. Thereafter, the first server 220 sequentially transmits to the second server 230 command information PT to generate instruction manuals corresponding to the program division data DPG and the ABI division information DST. Then, the second server 230 converts the sequentially received command information PT for generating instruction manuals into vector coordinate system command information VPT, which is vector coordinate system data, using an Embedding model, which is an example of the vector coordinate conversion means 231b, which is the pre-trained artificial intelligence means 231.

[0048] Then, the second server 230 transmits the vector coordinate system command information VPT to the first server 220 in sequence. Then, the first server 220 refers to a spatial map based on the vector coordinate system command information VPT received sequentially and selects the program partition data DPG and ABI partition information DST corresponding to the vector coordinate system program partition data VPG and vector coordinate system ABI partition information VST that are most closely related to the vector of the vector coordinate system command information VPT.

[0049] Then, the first server 220 sequentially transmits the program division data DPG, the ABI division information DST, and command information PT for generating corresponding instruction manuals to a third server 240 different from the first server 220. Then, the third server 240 generates a partial instruction manual file DML for the program division data DPG using chat GPT, which is an example of a text generation means 241a, which is the pre-trained artificial intelligence means 241, based on the program division data DPG, ABI division information DST, and command information PT to generate corresponding instruction manuals that are received sequentially.

[0050] Then, the third server 240 sequentially transmits the partial instruction manual file DML to the first server 220. Then, the first server 220 sequentially combines the plurality of partial instruction manual files DML received into one whole instruction manual file. The first server 220 is configured to register the synthesized overall instruction manual file in the database 221 in association with the program data PG.

[0051] This makes it possible to obtain the same effects as in the first embodiment described above, and a detailed description thereof will be omitted. Furthermore, as a unique effect of this embodiment, the program data PG and ABI information ST are respectively divided so as to be separated by the functions which are the contents of the program data PG, and the divided program division data DPG and ABI division information DST are converted into vector coordinate system program division data VPG and vector coordinate system ABI division information VST, the vector coordinate system program division data VPG and vector coordinate system ABI division information VST are recorded in a spatial map, the program division data DPG and ABI division information DST corresponding to the vector coordinate system program division data VPG and vector coordinate system ABI division information VST which are most closely related to the vector of the vector coordinate system command information VPT are selected, and based on the command information PT to generate instruction manuals corresponding to these, a partial instruction manual file DML for the program division data DPG is generated by chat GPT, which is an example of text generation means 241a which is the pre-trained artificial intelligence means 241. As a result, a partial instruction manual file DML with high accuracy for each function can be obtained. Also, multiple partial instruction manual files DML are combined into one complete instruction manual file. As a result, it is possible to obtain a complete instruction manual file with high overall content accuracy.

[0052] Next, an example of the operation of the instruction manual generating system 200 of this embodiment will be described in more detail. As shown in FIG. 6, in step S11, as an upload determination step, the first server 220 determines whether any program data PG has been uploaded to the first server 220 from a first user terminal 210A as an example of a user terminal. If it is determined that the data has been uploaded, the process proceeds to step S12. On the other hand, if it is determined that the data has not yet been uploaded, step S11 is repeated. In step S12, as an ABI information generating step, the first server 220 generates ABI information ST, which is specification and standard information that guarantees program compatibility at the binary level, based on the program data PG.

[0053] In step S13, which is a division / storage step, the first server 220 divides the program data PG and ABI information ST into sections each corresponding to a function, which is the content of the program data PG, and stores the divided program division data DPG and ABI division information DST in the database 221. The number of divisions does not necessarily have to be the same for the program data PG and the ABI information ST. In step S14, as shown in FIG. 7, as a division information transmission step, the first server 220 sequentially transmits one divided program division data DPG and the ABI division information DST corresponding to this program division data DPG to a second server 230 different from the first server 220. In step S15, as a vector coordinate system information conversion step, the second server 230 converts the sequentially received program partition data DPG and ABI partition information DST into vector coordinate system program partition data VPG and vector coordinate system ABI partition information VST, which are vector coordinate system data, using an Embedding model, which is an example of vector coordinate conversion means 231b, which is the pre-trained artificial intelligence means 231.

[0054] In step S16, as a vector coordinate system information transmitting step, the second server 230 sequentially transmits the vector coordinate system program partition data VPG and the vector coordinate system ABI partition information VST to the first server 220. In step S17, as a map recording and saving step, the first server 220 records the vector coordinate system program partition data VPG and vector coordinate system ABI partition information VST which are sequentially received in the spatial map. The spatial map is not limited to being stored in the database 221 of the first server 220, but may be stored in a storage area (storage unit) including a memory, or in a file. Then, the first server 220 stores the vector coordinate system program partition data VPG and the vector coordinate system ABI partition information VST in the database 221 in association with the program partition data DPG and the ABI partition information DST.

[0055] In step S18, as a first prompt transmission step, the first server 220 transmits to the second server 230 command information PT to generate instruction manuals corresponding to the program division data DPG and the ABI division information DST. In step S19, as a prompt vector coordinate system information conversion step, the second server 230 converts the received command information PT for generating instruction manuals into vector coordinate system command information VPT, which is vector coordinate system data, using an Embedding model, which is an example of the vector coordinate conversion means 231b, which is the pre-trained artificial intelligence means 231. In step S20, the second server 230 transmits vector coordinate system command information VPT to the first server 220 as a vector coordinate system prompt transmitting step.

[0056] In step S21, as a second prompt sending step, the first server 220 refers to a spatial map based on the received vector coordinate system command information VPT and selects program partition data DPG and ABI partition information DST corresponding to the vector coordinate system program partition data VPG and vector coordinate system ABI partition information VST that are most closely related to the vector of the vector coordinate system command information VPT. Then, the first server 220 transmits the program division data DPG, the ABI division information DST, and command information PT for generating corresponding instruction manuals to a third server 240 different from the first server 220. In step S22, as an instruction manual generation step, the third server 240 generates a partial instruction manual file DML for the program division data DPG using chat GPT, which is an example of a text generation means 241a, which is the pre-trained artificial intelligence means 241, based on the received program division data DPG, ABI division information DST, and command information PT to generate instruction manuals corresponding to these. In step S23, as an instruction manual sending step, the third server 240 sends the generated partial instruction manual file DML to the first server 220.

[0057] In step S24, as an example, as a determination step for determining whether partial instruction manual files DML have been generated equal to the number into which the ABI information ST was divided in step S13, the first server 220 counts the number of partial instruction manual files DML received and determines whether the number has reached the number into which it was divided in step S13. If it is determined that the time has been reached, the process proceeds to step S25, whereas if it is determined that the time has not been reached yet, the process returns to step S18 and steps S18 to S23 are repeated. That is, steps S18 to S23 are repeated until a partial instruction manual file DML is generated for each of the divided ABI division information DST. In this embodiment, as an example, the loop of steps S14 to S17 in which sequential transmission and sequential reception are performed is separated from the loop of steps S18 to S23, but the configuration may be such that steps S14 to S23 are repeated by returning to step S14. In steps S14 to S23, as long as the ABI information ST can be transmitted and received in sequence by the number of pieces of ABI division information DST obtained by dividing the ABI information ST, any configuration may be used with respect to the number of loops and the presence or absence of loops. In step S25, as an instruction manual synthesis / registration step, the first server 220 synthesizes the received partial instruction manual files DML into one whole instruction manual file. As for the method of synthesis, the partial instruction manual files DML received in sequence may be added in sequence, or a plurality of partial instruction manual files DML received in sequence may be synthesized all at once. Then, the first server 220 registers the synthesized overall instruction manual file in the database 221 in association with the program data PG.

[0058] As a result, as described above, a unique effect of this embodiment is that the program data PG and ABI information ST are each divided so that they are separated by the functions that are the contents of the program data PG, and the divided program division data DPG and ABI division information DST are converted into vector coordinate system program division data VPG and vector coordinate system ABI division information VST, the vector coordinate system program division data VPG and vector coordinate system ABI division information VST are recorded in the spatial map, the program division data DPG and ABI division information DST corresponding to the vector coordinate system program division data VPG and vector coordinate system ABI division information VST that are most closely related to the vector of the vector coordinate system command information VPT are selected, and based on the command information PT to generate instruction manuals corresponding to these, a partial instruction manual file DML for the program division data DPG is generated by chat GPT, which is an example of text generation means 241a, which is the pre-trained artificial intelligence means 241. As a result, a partial instruction manual file DML with high accuracy for each function can be obtained. Also, multiple partial instruction manual files DML are combined into one complete instruction manual file. As a result, it is possible to obtain a complete instruction manual file with high overall content accuracy.

[0059] Furthermore, in this embodiment, when the first server 220 sends command information PT to the second server 230 to generate instruction manuals corresponding to the program partition data DPG and ABI partition information DST, it determines whether the function of the program partition data DPG and the ABI partition information DST is an initialization function, a function function, or a recording function. For example, if there is a "constructor" code, it is determined to be an initialization function. Furthermore, if there is a "function" code, it is a function used to read information written in a program or the like, or to update that information, and is determined to be a functional function. Furthermore, if there is an “event” code, it is determined to be a recording function, which is a function that outputs an event that occurs in the program, for example, an input as a trigger and records it on the blockchain. Next, based on this determination, the first server 220 selects command information PT to generate instruction manuals for each template that has been prepared in advance according to the initialization function, function function, and recording function, which are the functions of the program division data DPG and the ABI division information DST. FIG. 8 shows command information PT for generating an instruction manual as a prompt when it is determined that the function is a function. Of course, when selecting command information PT for generating instructions as a prompt, a template may be selected and tuned. In addition, the first server 220 determines which of the functions of the program division data DPG and the ABI division information DST is the initialization function, function function, or recording function, and based on this, selects command information PT to generate instruction manuals for each template.As a result, as shown in Figure 8, the command information PT to generate instruction manuals as a prompt contains one of the functions of the program division data DPG and the ABI division information DST: the initialization function, function function, or recording function. In the example of FIG. 8, a function called "mint" is described as an example. During tuning, information on the initialization function, function function, and recording function, which are functions of the program partition data DPG and the ABI partition information DST, may be reflected in the command information PT for generating an instruction manual. For example, if the function of the program partition data DPG and the ABI partition information DST is function A, command information PT stating "Please explain function A" is reflected, and if the function is function B, command information PT stating "Please explain function B" is reflected. Furthermore, the command information PT for generating an instruction manual may contain common content as well as content that changes for each function of the program partition data DPG and the ABI partition information DST. For example, you might include something like, "Please explain the outline of this program data from a business perspective," or "Please explain how to use this program data from the perspective of an engineer in charge of operations." Then, the first server 220 is configured to transmit to the second server 230 command information PT to generate the selected instruction manual.

[0060] As a result, command information PT is selected for generating instruction manuals corresponding to the functions of the program partition data DPG and the ABI partition information DST, which are respectively partitioned so as to be separated by function. As a result, the first server 220 can improve the accuracy when selecting the program partition data DPG and ABI partition information DST corresponding to the vector coordinate system program partition data VPG and vector coordinate system ABI partition information VST that are most closely related to the vector of the vector coordinate system command information VPT by referring to a spatial map based on the vector coordinate system command information VPT received sequentially. Furthermore, the program partition data DPG and the ABI partition information DST are selected with greater accuracy. As a result, the third server 240 can improve the accuracy of content when generating a partial instruction manual file DML for the program division data DPG using chat GPT, which is an example of a text generation means 241a, which is a pre-trained artificial intelligence means 241, based on the program division data DPG, ABI division information DST, and command information PT for generating corresponding instruction manuals that are received sequentially.

[0061] In addition, in this embodiment, when any program data PG is uploaded from a user terminal to the first server 220, the first server 220 accesses a fourth server (not shown) different from the first server 220 based on the reference information input from the user terminal to obtain related information related to the contents of the program data PG or the ABI information ST. Then, the first server 220 is configured to add the acquired related information to command information PT for generating the instruction manual. Note that the first server 220 may be configured to access a fourth server (not shown) based on the contents of the uploaded program data PG or ABI information ST instead of the input reference information.

[0062] As a result, relevant information related to the contents of the program data PG or the ABI information ST is added to the command information PT for generating the instruction manual. As a result, the content accuracy of the command information PT for generating the operating instructions can be improved. Furthermore, the content of the command information PT for generating the operating instructions is more accurate. As a result, the third server 240 can further improve the accuracy of content when generating a partial instruction manual file DML for the program division data DPG using chat GPT, an example of a text generation means 241a, which is the pre-trained artificial intelligence means 241, based on the program division data DPG, ABI division information DST, and command information PT for generating corresponding instruction manuals that are received sequentially.

[0063] In the instruction manual generating system 200 according to the second embodiment of the present invention thus obtained, when any program data PG is uploaded from a user terminal to a first server 220 which is a server, the first server 220 generates ABI information ST which is specification / standard information for guaranteeing program compatibility at a binary level based on the program data PG, and divides the program data PG and the ABI information ST so as to separate them for each function which is the content of the program data PG, and stores the divided program division data DPG and the ABI division information DST in a database 221. The first server 220 sequentially transmits the divided program division data DPG and the ABI division information DST to a second server 230 different from the first server 220, and the second server 230 sequentially converts the sequentially received program division data DPG and ABI division information DST into vector coordinate system data, which are vector coordinate system data, using an Embedding model which is an example of vector coordinate conversion means 231b which is a pre-trained artificial intelligence means 231. The first server 220 records the sequentially received vector coordinate system program partition data VPG and vector coordinate system ABI partition information VST in a spatial map and stores them in the database 221 in association with the program partition data DPG and ABI partition information DST. The first server 220 sequentially transmits command information PT to the second server 230 to generate instruction manuals corresponding to the program partition data DPG and the ABI partition information DST. The second server 230 converts the sequentially received command information PT to generate instruction manuals into vector coordinate system command information VPT, which is vector coordinate system data, using an Embedding model which is an example of the vector coordinate conversion means 231b which is the pre-trained artificial intelligence means 231. The first server 220 sequentially transmits the vector coordinate system command information VPT to the first server 220.A space map is referenced based on the sequentially received vector coordinate system command information VPT, and program partition data DPG and ABI partition information DST corresponding to the vector coordinate system program partition data VPG and the vector coordinate system ABI partition information VST having the closest relationship to the vector of the vector coordinate system command information VPT are selected, and the program partition data DPG, the ABI partition information DST, and command information PT for generating the corresponding instruction manuals are sequentially transmitted to a third server 240 different from the first server 220, and the third server 240 sequentially transmits the program partition data DPG, the ABI partition information DST, and command information PT for generating the corresponding instruction manuals to the third server 240. Based on T, chat GPT, which is an example of text generation means 241a, which is the pre-trained artificial intelligence means 241, is used to generate partial instruction manual files DML for the program division data DPG and sequentially transmit them to the first server 220, and the first server 220 synthesizes the multiple partial instruction manual files DML received sequentially into one overall instruction manual file, and registers the synthesized overall instruction manual file in the database 221 in association with the program data PG. This makes it possible to obtain partial instruction manual files DML with high content accuracy for each function, and to obtain an overall instruction manual file with high content accuracy as a whole.

[0064] Furthermore, when the first server 220 transmits to the second server 230 command information PT to generate instruction manuals corresponding to the program division data DPG and ABI division information DST, the first server 220 determines whether the function of the program division data DPG and the ABI division information DST is an initialization function, a function function, or a recording function, and based on this determination, selects command information PT to generate instruction manuals for a template prepared in advance according to the functions of the program division data DPG and the ABI division information DST, the initialization function, the function function, or the recording function, and transmits the selected command information PT to the second server 230. This can improve the accuracy when selecting the program partition data DPG and ABI partition information DST corresponding to the vector coordinate system program partition data VPG and vector coordinate system ABI partition information VST that are most closely related to the vector of the vector coordinate system command information VPT by referring to the spatial map, and further improve the accuracy of the content when the third server 240 generates a partial instruction manual file DML for the program partition data DPG using the chat GPT, which is an example of the text generation means 241a, which is the pre-trained artificial intelligence means 241, based on the program partition data DPG, ABI partition information DST, and command information PT for generating instruction manuals corresponding to these that are sequentially received.

[0065] In addition, when any program data PG is uploaded from a user terminal to the first server 220, the first server 220 accesses a fourth server different from the first server 220 based on the reference information input from the user terminal or the contents of the uploaded program data PG or ABI information ST to obtain related information related to the contents of the program data PG or ABI information ST, and adds it to the command information PT to generate instruction manuals.This configuration makes it possible to further improve the content accuracy of the command information PT to generate instruction manuals.Furthermore, the third server 240 can further improve the content accuracy when generating a partial instruction manual file DML for the program division data DPG using the chat GPT, which is an example of the text generation means 241a, which is the pre-trained artificial intelligence means 241, based on the program division data DPG, ABI division information DST, and command information PT to generate instruction manuals corresponding to these that are received sequentially.

[0066] Furthermore, the program of the instruction manual generating system 200 according to the second embodiment of the present invention includes an upload determination step S11 in which it is determined whether any program data PG has been uploaded from a user terminal to a first server 220, which is a server; an ABI information generation step S12 in which the first server 220 generates ABI information ST, which is specification and standard information that guarantees program compatibility at a binary level, based on the program data PG; a division and storage step S13 in which the first server 220 divides the program data PG and the ABI information ST so as to separate them for each function that is the content of the program data PG, and stores the divided program division data DPG and ABI division information DST in the database 221; a division information transmission step S14 in which the first server 220 sequentially transmits the divided program division data DPG and ABI division information DST to a second server 230 different from the first server 220; a vector coordinate system information conversion step S15 in which an Embedding model, which is an example of the vector coordinate conversion means 231b which is the artificial intelligence means 231, is used to convert the vector coordinate system program partition data VPG and the vector coordinate system ABI partition information VST, which are vector coordinate system data; a vector coordinate system information transmission step S16 in which the second server 230 sequentially transmits the vector coordinate system program partition data VPG and the vector coordinate system ABI partition information VST to the first server 220; a map recording and storage step S17 in which the first server 220 records the vector coordinate system program partition data VPG and the vector coordinate system ABI partition information VST which have been sequentially received in a space map and stores them in the database 221 in association with the program partition data DPG and the ABI partition information DST; a first prompt transmission step S18 in which the first server 220 sequentially transmits to the second server 230 command information PT to generate an instruction manual corresponding to the program partition data DPG and the ABI partition information DST;a prompt vector coordinate system information conversion step S19 for converting the sequentially received command information PT for generating operating instructions into vector coordinate system command information VPT, which is vector coordinate system data, using an Embedding model, which is an example of the vector coordinate conversion means 231b, which is the pre-trained artificial intelligence means 231; a vector coordinate system prompt transmission step S20 for the second server 230 to sequentially transmit the vector coordinate system command information VPT to the first server 220; a second prompt transmission step S21 for the first server 220 to refer to a space map based on the sequentially received vector coordinate system command information VPT, select program partition data DPG and ABI partition information DST corresponding to the vector coordinate system program partition data VPG and vector coordinate system ABI partition information VST that are most closely related to the vector of the vector coordinate system command information VPT, and sequentially transmit the program partition data DPG, the ABI partition information DST, and the command information PT for generating operating instructions corresponding to these to a third server 240 different from the first server 220; The third server 240 generates a partial instruction manual file DML for the program division data DPG using chat GPT, which is an example of the text generating means 241a, which is the pre-trained artificial intelligence means 241, based on the program division data DPG, ABI division information DST, and command information PT for generating the corresponding instruction manuals, which are received in sequence, in an instruction manual generating step S22; the third server 240 sequentially transmits the generated partial instruction manual file DML to the first server 220 in an instruction manual transmitting step S23; and the first server 220 synthesizes the multiple partial instruction manual files DML received in sequence into one overall instruction manual file and registers the synthesized overall instruction manual file in the database 221 in association with the program data PG. This has a significant effect, as it is possible to obtain a partial instruction manual file DML with high content accuracy for each function and to obtain an overall instruction manual file with high content accuracy as a whole. [Explanation of symbols]

[0067] 100, 200 ... Instruction manual generation system 110A, 210A... First user terminal (user terminal) 110B, 210B... Second user terminal (user terminal) 120, 220... First server 121, 221 ··· Database 130, 230...Second server 131, 231 ··· Pre-trained artificial intelligence means 131a Text generation method (Chat GPT) 231b....Vector coordinate conversion means 240 ··· Third Server 241 ··· Pre-trained artificial intelligence means 241a Text generation method (Chat GPT) PG: Program data DPG: Program division data VPG: Vector coordinate system program division data PG1: First program part data PG2: Subsequent program part data ST ··· ABI information (standard information) DST ABI division information VST... Vector coordinate system ABI division information PT: Prompt to generate instructions VPT Vector coordinate system command information ML... Instruction file DML... Partial instruction manual file ML1... Complementary instruction file

Claims

1. An instruction manual generation system having a user terminal and a server for generating an instruction manual file for an arbitrary program, When any program data is uploaded from the user terminal to a first server, the first server generates ABI information, which is specification and standard information that guarantees program compatibility at a binary level, based on the program data; The first server transmits the program data, the ABI information, and command information for generating an instruction manual to a second server different from the first server; The second server generates an instruction manual file for the program using the pre-trained artificial intelligence means based on the received program data, ABI information, and instruction information for generating an instruction manual, and transmits the generated instruction manual file to the first server; The instruction manual generating system is characterized in that the first server is configured to register the received program instruction manual file in a database in association with program data.

2. An instruction manual generation system having a user terminal and a server for generating an instruction manual file for an arbitrary program, When any program data is uploaded from the user terminal to a first server, the first server generates ABI information, which is specification / standard information that guarantees program compatibility at a binary level based on the program data, and divides the program data and the ABI information so as to be divided by functions, which are the contents of the program data, and stores the divided program division data and the ABI division information in a database; The first server sequentially transmits the divided program division data and the ABI division information to a second server different from the first server; The second server converts the sequentially received program partition data and ABI partition information into vector coordinate system program partition data and vector coordinate system ABI partition information, which are vector coordinate system data, using a vector coordinate conversion means that is a pre-trained artificial intelligence means, and sequentially transmits the vector coordinate system program partition data and the vector coordinate system ABI partition information to the first server; The first server records the vector coordinate system program partition data and the vector coordinate system ABI partition information sequentially received in a spatial map, and stores the vector coordinate system program partition data and the ABI partition information in a database in association with each other; The first server sequentially transmits to the second server command information for generating an instruction manual corresponding to the program division data and the ABI division information; The second server converts the sequentially received instruction information for generating the instruction manual into vector coordinate system instruction information, which is vector coordinate system data, using the vector coordinate conversion means, which is the pre-trained artificial intelligence means, and sequentially transmits the vector coordinate system instruction information to the first server; the first server refers to a space map based on the sequentially received vector coordinate system command information, selects program partition data and ABI partition information corresponding to the vector coordinate system program partition data and vector coordinate system ABI partition information that are most closely related to the vector of the vector coordinate system command information, and sequentially transmits the program partition data, the ABI partition information, and command information for generating the corresponding instruction manual to a third server different from the first server; The third server generates partial instruction manual files for the program division data using a text generation means which is a pre-trained artificial intelligence means based on the program division data, ABI division information, and instruction information for generating instruction manuals corresponding thereto, and transmits the partial instruction manual files to the first server in sequence; An instruction manual generation system characterized in that the first server is configured to synthesize multiple partial instruction manual files received sequentially into a single overall instruction manual file, and to associate the synthesized overall instruction manual file with program data and register it in a database.

3. The instruction manual generation system described in claim 2, characterized in that when the first server sends command information to the second server to generate instruction manuals corresponding to the program division data and ABI division information, it determines which of the functions of the program division data and ABI division information is an initialization function, a function function, or a recording function, and based on this determination, selects command information to generate instruction manuals for each template prepared in advance according to the initialization function, function function, or recording function, which is the function of the program division data and ABI division information, and sends command information to the second server to generate the selected instruction manuals.

4. The instruction manual generation system described in claim 2 or claim 3, characterized in that when any program data is uploaded from the user terminal to a first server, the first server accesses a fourth server different from the first server based on the reference information input from the user terminal or the contents of the uploaded program data or ABI information, to obtain related information related to the contents of the program data or ABI information, and adds it to command information to generate instruction manuals.

5. The first server determines whether or not the capacity of the program data is equal to or larger than a predetermined capacity before transmitting the program data, the ABI information, and the command information for generating the instruction manual to the second server; The instruction manual generation system described in claim 1, characterized in that when the capacity is equal to or greater than the specified capacity, the first server divides the program data into multiple parts and transmits a first program part data containing the first part of the entire program data to the second server as program data.

6. The instruction manual generation system described in claim 5, characterized in that when the first server receives a program instruction manual file from the second server, it inserts subsequent program portion data that was not divided and sent to the second server into the received program instruction manual file to create a supplemental instruction manual file, and registers the supplemental instruction manual file in a database as an instruction manual file in association with the program data.

7. A program for an instruction manual generation system for generating an instruction manual file for an arbitrary program, an upload determination step of determining whether any program data has been uploaded from the user terminal to the first server; an ABI information generating step in which the first server generates ABI information, which is specification / standard information that guarantees program compatibility at a binary level, based on the program data when the program data is uploaded; a prompt transmission step in which the first server transmits the program data, the ABI information, and command information for generating an instruction manual to a second server different from the first server; an instruction manual generating step in which the second server generates an instruction manual file for the program using a pre-trained artificial intelligence means based on the received program data, ABI information, and instruction manual generating command information; an instruction manual transmission step in which the second server transmits the generated instruction manual file of the program to the first server; a step of registering an instruction manual file received by the first server in a database in association with program data, the instruction manual file being registered by the first server in a database;

8. A program for an instruction manual generation system for generating an instruction manual file for an arbitrary program, an upload determination step of determining whether any program data has been uploaded from the user terminal to a first server; an ABI information generating step in which the first server generates ABI information, which is specification / standard information that guarantees program compatibility at a binary level, based on the program data; a division / storage step in which the first server divides the program data and the ABI information into sections for each function, which is the content of the program data, and stores the divided program division data and the ABI division information in a database; A division information transmission step in which the first server sequentially transmits the divided program division data and ABI division information to a second server different from the first server; a vector coordinate system information conversion step in which the second server converts the sequentially received program partition data and ABI partition information into vector coordinate system program partition data and vector coordinate system ABI partition information, which are vector coordinate system data, respectively, using a vector coordinate conversion means which is a pre-trained artificial intelligence means; a vector coordinate system information transmitting step in which the second server sequentially transmits the vector coordinate system program partition data and the vector coordinate system ABI partition information to the first server; a map recording and storage step in which the first server records the vector coordinate system program partition data and the vector coordinate system ABI partition information sequentially received in a spatial map and stores the program partition data and the ABI partition information in a database in association with each other; a first prompt transmission step in which the first server sequentially transmits to a second server command information for generating an instruction manual corresponding to the program division data and the ABI division information; a prompt vector coordinate system information conversion step in which the second server converts the sequentially received instruction information for generating an instruction manual into vector coordinate system instruction information, which is vector coordinate system data, by using a vector coordinate conversion means which is a pre-trained artificial intelligence means; a vector coordinate system prompt sending step in which the second server sequentially sends the vector coordinate system command information to the first server; a second prompt transmission step in which the first server refers to a space map based on the sequentially received vector coordinate system command information, selects program partition data and ABI partition information corresponding to the vector coordinate system program partition data and vector coordinate system ABI partition information that are most closely related to the vector of the vector coordinate system command information, and sequentially transmits the program partition data, the ABI partition information, and command information for generating the corresponding instruction manual to a third server different from the first server; an instruction manual generating step in which the third server generates a partial instruction manual file for the program division data by using a text generating means which is a pre-trained artificial intelligence means, based on the program division data, ABI division information, and instruction information for generating instruction manuals corresponding thereto, which are sequentially received; an instruction manual transmission step in which the third server sequentially transmits the generated partial instruction manual file to the first server; A program for an instruction manual generation system, characterized in that the first server has an instruction manual synthesis / registration step of synthesizing a plurality of partial instruction manual files received sequentially into one overall instruction manual file, and associating the synthesized overall instruction manual file with program data and registering it in a database.

Images