Task processing support device, task processing system and program

Abstract

To assist a user in creating final input information.SOLUTION: A task processing support device supporting creation of final input information to be input to a processing model, for making the processing model execute a target task input by the user on includes: a candidate preparation part for creating one or more candidate input pieces of information according to the target task; a generation example output part for outputting a generation example of the processing model when each of the candidate input information is input to the processing model; and a providing part for providing the user with one or more of pieces of the candidate input information associated with the generated example.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a task processing assistance device, a task processing system, and a program. [Background technology]

[0002] Patent Document 1 discloses a technique for creating a prompt that includes reference information in response to a question entered by a user. Patent Document 1: Patent No. 731375 Summary of the Invention

[0003] A first aspect of the present invention provides a task processing support device that supports the creation of final input information to be input to a processing model so that a target task input by a user can be executed by the processing model. The task processing support device may include a candidate creation unit that creates one or more pieces of candidate input information based on the target task. Any of the task processing support devices may include a generation example output unit that outputs a generation example of the processing model when each piece of candidate input information is input to the processing model. Any of the task processing support devices may include a providing unit that associates one or more pieces of candidate input information with the generation example and provides them to the user.

[0004] In any of the above task processing assistance devices, the candidate creation unit may include a preprocessing unit that creates preprocessing input information based on the target task, for causing the processing model to generate related information related to the candidate input information, and inputs the preprocessing input information to the processing model.In any of the above task processing assistance devices, the candidate creation unit may include a candidate processing unit that creates one or more pieces of candidate input information based on the related information generated by the processing model.

[0005] Any of the above task processing support devices may include an input information database that stores one or more pieces of existing input information. In any of the above task processing support devices, the candidate processing unit may search the input information database for one or more pieces of existing input information corresponding to the related information, and output one or more pieces of candidate input information based on at least one piece of the searched existing input information.

[0006] In any of the above task processing support devices, the input information database may record, for each of the existing input information, output information that defines an output when the processing model executes the existing input information, among information related to the existing input information. In any of the above task processing support devices, the candidate processing unit may search for the existing input information having the output information corresponding to the related information.

[0007] In any of the above task processing assistance devices, the preprocessing unit may create the preprocessing input information for generating keywords to be included in the output information based on the target task.

[0008] In any of the above task processing support devices, the candidate creation unit may include an input information creation unit that creates input information by extracting information regarding an input to be included in the final input information from the target task.In any of the above task processing support devices, the candidate processing unit may create provisional processing input information by applying the input information to each of the existing input information searched from the input information database.In any of the above task processing support devices, the generation example output unit may obtain the generation examples corresponding to each of the existing input information by inputting the provisional processing input information to the processing model.

[0009] In any of the above task processing support devices, the candidate creation unit may include an information acquisition unit that acquires user information related to the user. In any of the above task processing support devices, the candidate processing unit may add content related to the user information to the tentative process input information.

[0010] In any of the above task processing assistance devices, the providing unit may provide the existing input information to the user as the candidate input information.

[0011] In any of the above task processing assistance devices, the providing unit may provide the tentative process input information to the user as the candidate input information.

[0012] In any of the above task processing assistance devices, the generation example output unit may generate information regarding an input to be included in the final input information in the processing model according to the target task and apply the information to the existing input information.

[0013] Any of the above task processing assistance devices may include an information acquisition unit that acquires user information related to the user. In any of the above task processing assistance devices, the generation example output unit may reflect user information in the candidate input information and cause the processing model to generate a generation example.

[0014] Any of the above task processing assistance devices may include an input information registration unit that acquires the final input information created by the user based on the candidate input information and registers the final input information in the input information database.

[0015] In any of the above task processing support devices, the input information database may record, for each of the existing input information, output information obtained by extracting information regarding an output when the existing input information is executed from information regarding the existing input information. In any of the above task processing support devices, the input information registration unit may create the output information corresponding to the final input information and register it in the input information database.

[0016] A second aspect of the present invention provides a task processing support device that supports the creation of final input information to be input to a processing model so that a target task input by a user can be executed by the processing model. The task processing support device may include a preprocessing unit that creates preprocessing input information based on the target task, causing the processing model to generate related information related to candidate input information that is a candidate for the final input information, and inputs the preprocessing input information to the processing model. Any of the task processing support devices may include a candidate processing unit that creates one or more pieces of candidate input information based on the related information generated by the processing model. Any of the task processing support devices may include a providing unit that provides one or more pieces of candidate input information to the user.

[0017] In a third aspect of the present invention, there is provided a task processing system comprising the task processing assistance device according to the first or second aspect and the processing model.

[0018] In a fourth aspect of the present invention, there is provided a program for causing a computer to function as the task processing assistance device according to the first or second aspect.

[0019] The above summary of the invention does not list all of the features of the present invention, and subcombinations of these features may also constitute inventions. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a block diagram showing an example of a task processing system 10 according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram illustrating an example of a target task created by a user. [Figure 3] 10 is a diagram showing an example of candidate input information created by the candidate creating unit 110. FIG. [Figure 4] 10 is a diagram showing an example of generation example request information created by the generation example output unit 120. FIG. [Figure 5] FIG. 2 is a diagram illustrating an example of a generated example generated by a processing model 210. [Figure 6] FIG. 2 is a diagram illustrating an example of the configuration of a candidate creating unit 110. [Figure 7] FIG. 10 is a diagram illustrating an example of preprocessing input information. [Figure 8] FIG. 10 is a diagram illustrating an example of processing result information. [Figure 9] FIG. 10 is a diagram illustrating another example of the configuration of the candidate creating unit 110. [Figure 10] FIG. 10 is a diagram illustrating an example of provisional process input information. [Figure 11] FIG. 2 is a diagram illustrating another example of the configuration of the task processing assistance device 100. [Figure 12] FIG. 10 is a diagram showing an example of keyword generation information to be input to a processing model 210. [Figure 13] 10 is an example of related keywords generated by the processing model 210. [Figure 14] FIG. 2 is a diagram showing an example of a structured command to be input to a processing model 210. [Figure 15] FIG. 15 is a diagram illustrating an outline of the processing of the task processing system 10 described with reference to FIGS. 1 to 14. [Figure 16] FIG. 15 is a diagram illustrating an outline of the processing of the task processing system 10 described with reference to FIGS. 1 to 14. [Figure 17] 12 illustrates an example computer 1200 in which aspects of the present invention may be embodied, in whole or in part. DETAILED DESCRIPTION OF THE INVENTION

[0021] The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the scope of the invention according to the claims. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0022] 1 is a block diagram showing an example of a task processing system 10 according to an embodiment of the present invention. The task processing system 10 supports the creation of final input information to be input to the processing model 210 so that the processing model 210 can execute a target task input by a user. The target task may include information on the procedure to be executed by the processing model 210, may include information on the output to be output by the processing model 210, or may include information on the input to be processed by the processing model 210. The target task may include document data or image data.

[0023] The processing model 210 may be created by machine learning using training data. The processing model 210 may use a publicly available general-purpose generative artificial intelligence (also called generative AI). The generative artificial intelligence in this example is created by deep learning. The generative artificial intelligence can generate outputs for a variety of requests or questions, as well as for a certain type of explanatory variables. The processing model may be a large-scale language model. The processing model 210 may be a GPT (registered trademark) (Generative Pre-trained Transformer), a BERT (Bidirectional Encoder Representations from Transformers), or another language model.

[0024] The input information is information for causing the processing model 210 to execute a predetermined task. The input information may be a so-called prompt that is input to the generation AI. The final input information is input information for causing the processing model 210 to execute the target task in this example. The final input information includes at least a portion of the content of the target task. The task processing system 10 supports the creation of the final input information, thereby supporting the processing of the task indicated by the target task.

[0025] The task processing system 10 includes a task processing assistance device 100 and a task processing device 200. The task processing device 200 has a processing model 210 that operates in response to input information. The task processing device 200 may include one or more information processing devices that operate the processing model 210. In the task processing device 200, the multiple information processing devices may cooperate to realize processing according to the processing model 210.

[0026] The task processing assistance device 100 assists a user in creating final input information. The task processing assistance device 100 includes one or more information processing devices. Multiple information processing devices may work together to function as the task processing assistance device 100. The information processing devices function as each component of the task processing assistance device 100. Each information processing device may be a computer terminal managed by a user receiving a service, a server managed by a service provider, or another computer. Each information processing device may be connected via a general-purpose network such as the Internet, or may be connected via a dedicated line. Each information processing device may be connected via a wireless line or a wired line.

[0027] The task processing support device 100 of this example includes a candidate creation unit 110, a generation example output unit 120, and a providing unit 130. The task processing support device 100 may further include an information acquisition unit 113. The candidate creation unit 110 creates one or more candidate input information based on a target task created by a user. The candidate input information is a candidate for input information that serves as the basis for creating final input information. The final input information may include at least a portion of any of the candidate input information. The task processing support device 100 of this example presents to the user a corresponding generation example generated by the processing model 210 when each candidate input information is input to the processing model 210. The user can refer to the contents of the generation example and appropriately select candidate input information that serves as the basis for creating final input information. In this way, the task processing support device 100 supports the user in creating final input information. Furthermore, by reducing the number of times that the processing model required to generate the output required for executing a task is operated through the task processing support device 100, computational resources can be used more effectively.

[0028] The candidate creation unit 110 may extract existing input information suitable for the target task from a plurality of existing input information stored in advance, and set it as candidate input information. The existing input information may be input information previously created by the user who created the target task or another user, may be input information automatically created by the information processing device, or may be input information created by the information processing device based on input information created by the user. The candidate creation unit 110 may create search criteria for searching existing input information based on the content of the target task. The candidate creation unit 110 may use words, etc. included in the target task as search keywords.

[0029] The generation example output unit 120 outputs a generation example of the processing model 210 for each piece of candidate input information when the candidate input information is input to the processing model 210. The generation example output unit 120 may acquire a generation example by actually inputting each piece of candidate input information to the processing model 210. The generation example output unit 120 of this example creates generation example request information including the candidate input information and inputs it to the processing model 210.

[0030] The information acquisition unit 113 acquires user information related to the user. The user may refer to the user who created the target task, the user to whom a generation example is provided, or another user. The information acquisition unit 113 may acquire a user ID that identifies the user, and acquire user information based on the user ID. The user information may be information that is associated with the user ID and managed in an employee database or the like. In another example, the user information may be directly input to the information acquisition unit 113. The user information includes information related to the user's attributes. The user's attributes may be the current job type, past work history, work mission, etc.

[0031] The generation example output unit 120 may reflect the user information in the candidate input information and cause the processing model 210 to generate a generation example. The generation example output unit 120 may reflect the user information in the generation example request information so that a generation example suitable for the user indicated by the user information is generated. The generation example output unit 120 may add content related to the user information to the generation example request information. The generation example output unit 120 may replace at least a part of the candidate input information with the user information.

[0032] The generation examples may match the content that the processing model 210 actually outputs in response to the candidate input information, or may be partially different. The generation example output unit 120 may acquire the generation examples by inputting each piece of candidate input information into a processing model different from the processing model 210. The processing model may be a copy of the processing model 210, and may be created using the same learning data as the processing model 210, or may be created using different learning data.

[0033] The providing unit 130 provides one or more candidate input information and generation examples in association with each other to the user. The providing unit 130 may have a display device that displays the candidate input information and the generation examples to the user, and may have a communication device that transmits the candidate input information and the generation examples to the user's terminal.

[0034] By providing candidate input information and generation examples in association with each other, the user can easily understand the output of the processing model 210 when using each piece of candidate input information. This allows the user to select candidate input information according to the output they wish to obtain. Since final input information can be created based on the selected candidate input information, the user can easily create final input information according to the obtained output.

[0035] The task processing assistance device 100 may receive final input information from a user and input it to the processing model 210. The user may also input the final input information to the processing model 210 without going through the task processing assistance device 100.

[0036] 2 is a diagram showing an example of a target task created by a user. The target task in this example is expressed in sentences. The target task in this example expresses what kind of output is required from the processing model 210.

[0037] 3 is a diagram showing an example of candidate input information created by the candidate creation unit 110. The candidate input information in this example includes a process definition portion, an input information portion, and an output definition portion. In this specification, the process definition portion and the output definition portion are collectively referred to as the output information portion. The output information portion of each piece of input information includes at least one of the process definition portion and the output definition portion.

[0038] The output information section is a section that explains the requirements for the output generated by the process model 210. The process definition section is a section that indicates the content or procedure of the processing to be executed in the process model 210 to obtain the output. The process definition section may include information regarding a methodology for generating an output corresponding to a task from the process model 210. The information may be expressed, for example, as attributes of a virtual person who is thought to possess the methodology. The process definition section may include constraints on a specified process. The constraints may be expressed as points to note when performing the specified process. The points to note may be information that instructs how to use input information, which will be described later.

[0039] The output definition section indicates the specific content of the expected output. The output definition section may indicate the items included in the output, or the output format.

[0040] The input information is information about background knowledge that the processing model 210 should refer to when generating output. The input information may include, for example, information about the purpose of executing a task, information about the target on which the task is executed, and information about the entity that receives the results of the task. The input information may be information whose output changes when the input information changes, even if there is no change in the content or procedure of the processing of the processing model 210. The input information may refer to all information other than the output information.

[0041] FIG. 4 is a diagram showing an example of generation example request information created by the generation example output section 120. For each piece of candidate input information, the generation example output section 120 creates generation example request information including the candidate input information. The generation example request information may include the entire candidate input information. The candidate input information in FIG. 4 is the same as the candidate input information shown in FIG. 3, but the contents of the "# item" are omitted. Note that the generation example request information may also include hierarchical information indicating the hierarchy of the information in the generation example request information. Specific examples of hierarchical information include information indicating the start and end points of a hierarchy, and information indicating the range included in the same hierarchy.

[0042] The generation example request information may be the candidate input information itself, or may be information in which predetermined information is added to the candidate input information. The candidate input information in this example is information that has been stored in advance. Therefore, the content of the candidate input information often does not completely match the content of the target task. The generation example output unit 120 may create generation example request information that reflects the content of the target task in the candidate input information.

[0043] For example, the input information portion of the candidate input information changes depending on the specific content of the task, and therefore often does not match the target task. The generation example output unit 120 may create generation example request information that reflects the content of the target task for the input information portion of the candidate input information. Alternatively, the generation example output unit 120 may create generation example request information that includes an instruction to reflect the content of the target task for the input information portion of the candidate input information.

[0044] On the other hand, the output information portion indicates the content of the output. Therefore, by directly using the output information portion of the candidate input information, it becomes easier to imagine the content of a generation example when the candidate input information is used. The generation example output unit 120 may use the output information portion of the candidate input information directly to cause the processing model 210 to generate a generation example.

[0045] The generation example request information in this example additionally includes first instruction information, first procedure information, target task, and background information. These pieces of information may be created by the generation example output unit 120. The first instruction information is an instruction for causing the processing model 210 to execute candidate input information. The first procedure information indicates a procedure that the processing model 210 should execute in order to cause the processing model 210 to generate a generation example. The contents of the first instruction information and the first procedure information may be set in advance in the generation example output unit 120.

[0046] The first procedure information may include a procedure for dividing the candidate input information into an output definition portion, a process definition portion, and an input information portion. The first procedure information may include a procedure for supplementing the input information portion based on the target task and background information. The first procedure information may include a procedure for executing the candidate input information with the supplemented input information portion.

[0047] The background information is an example of the user information described above. As described above, the background information includes information about the attributes of the user to whom the generation example is provided. The user's attributes may be the current occupation, past work history, work mission, etc. The generation example output unit 120 may specify that the content of the background information is to be reflected in the input information portion of the candidate input information. In the example of FIG. 4, an instruction to that effect is included in the first procedure information. By including the background information in the generation example request information, it is possible to obtain an appropriate generation example according to the user's attributes.

[0048] Fig. 5 is a diagram showing an example of a generation example generated by the processing model 210. This generation example is an example when the generation example request information shown in Fig. 4 is input to the processing model 210. The generation example output unit 120 creates generation example request information for each piece of candidate input information, and acquires a generation example for each piece of candidate input information.

[0049] The providing unit 130 may provide the candidate input information shown in Fig. 3 and the generation example shown in Fig. 5 in association with each other to the user. In another example, the providing unit 130 may provide the generation example request information shown in Fig. 4 and the generation example shown in Fig. 5 in association with each other to the user. In either case, the candidate input information and the corresponding generation example are presented to the user. This can assist the user in creating final input information.

[0050] 6 is a diagram showing an example of the configuration of the candidate creation unit 110. The candidate creation unit 110 of this example generates related information for creating candidate input information using a processing model 210. The related information may be information for searching for candidate input information.

[0051] The candidate creation unit 110 in this example includes a preprocessing unit 112, a candidate processing unit 114, and an input information database 116. The preprocessing unit 112 creates preprocessing input information based on a target task and inputs the preprocessing input information to the processing model 210. Here, the preprocessing input information is information that causes the processing model 210 to generate related information used to create candidate input information. The preprocessing unit 112 may create one piece of preprocessing input information for one target task. The preprocessing input information may be a prompt that causes the processing model 210 to generate search conditions for searching for candidate input information suitable for the target task.

[0052] The processing model 210 generates related information based on preprocessing input information. The processing model 210 of this example outputs processing result information including the related information. The preprocessing unit 112 of this example extracts the related information from the processing result information and outputs it to the candidate processing unit 114.

[0053] The candidate processing unit 114 creates one or more candidate input information based on the related information generated by the processing model. In this example, the candidate processing unit 114 searches the input information database 116 for one or more existing input information corresponding to the related information, and outputs one or more candidate input information based on at least one of the searched existing input information. The candidate processing unit 114 may output candidate input information that includes at least a portion of the content included in the existing input information. For example, the candidate processing unit 114 may output candidate input information that includes the output information portion of the existing input information. The candidate processing unit 114 may use the existing input information as the candidate input information as is.

[0054] The input information database 116 records one or more pieces of existing input information. In this example, the input information database 116 is provided in the task processing assistance device 100, but the input information database 116 may be provided outside the task processing assistance device 100.

[0055] The candidate processing unit 114 outputs one or more pieces of candidate input information extracted from the input information database 116 to the generation example output unit 120. The operation of the generation example output unit 120 is similar to the examples described with reference to FIGS.

[0056] The input information database 116 may record output information for each piece of existing input information, among information related to the existing input information. Here, output information refers to information included in the input information that defines the output to be generated by the processing model. The input information database 116 may classify and record the contents of the existing input information into a process definition part, an output definition part, and an input information part.

[0057] The candidate processing unit 114 may search for existing input information having output information corresponding to the related information. The candidate processing unit 114 may search the process definition portion or the output definition portion using keywords included in the related information. This makes it possible to present to the user a generation example including an output corresponding to the target task. In this case, the preprocessing unit 112 may create preprocessing input information for generating keywords to be included in the output information based on the target task information. The preprocessing unit 112 may include in the preprocessing input information an indication that keywords to be included in the output information will be generated.

[0058] FIG. 7 is a diagram showing an example of preprocessing input information. The preprocessing input information in this example includes second instruction information, second procedure information, and a target task. The second instruction information is an instruction for causing the processing model 210 to generate related information. In the example of FIG. 7, "keywords" correspond to the related information. The second procedure information indicates the procedure that the processing model 210 should execute in order to cause the processing model 210 to generate related information. The contents of the second instruction information and the second procedure information may be set in advance in the preprocessing unit 112.

[0059] The second procedure information may include a procedure for extracting a deliverable (output) required by the target task. In this example, the deliverable is a lean canvas made of insulation material with a predetermined thermal conductivity, as shown in FIG. 5. The second procedure information may include a procedure for identifying a profession that is good at making the deliverable or good at teaching how to make it. The second procedure information may include a procedure for extracting reference information or intermediate deliverables that people in that profession need when making the deliverable. The second procedure information may include a procedure for creating related information based on the reference information or intermediate deliverables. The related information may include words contained in the reference information or words that indicate the intermediate deliverables.

[0060] FIG. 8 is a diagram showing an example of processing result information. The processing result information includes related information. The processing result information may also include information other than related information. The processing result information in this example includes information about the processing content and the intermediate product. The processing content may include at least a portion of the content of the preprocessing input information. The processing content may include at least a portion of the content of the second procedure information of the preprocessing input information. The information about the intermediate product indicates the content of the intermediate product generated in accordance with the preprocessing input information.

[0061] The candidate processing unit 114 extracts candidate input information based on the related information, and the content of the processing result information may be provided to the user together with the extracted candidate input information.

[0062] 9 is a diagram showing another example configuration of the candidate creation unit 110. The candidate creation unit 110 of this example includes at least one of an input information creation unit 111 and an information acquisition unit 113. The configuration other than the input information creation unit 111 and the information acquisition unit 113 is the same as any of the examples described in this specification. When the information acquisition unit 113 is provided in the candidate creation unit 110, unlike the example of FIG. 1, user information does not need to be input to the generation example output unit 120.

[0063] 4, input information according to the target task is supplied to the processing model 210 to generate a generation example. In this example, the candidate creation unit 110 creates input information according to the target task and applies it to each piece of candidate input information. This example also makes it possible to obtain a generation example when input information according to the target task is applied to candidate input information.

[0064] The input information creation unit 111 creates input information by extracting information about inputs to be included in final input information from the target task. The input information is, for example, solution information, industry information, etc. in the candidate input information shown in FIG. 3, but is not limited to these. The input information creation unit 111 may create the input information from words included in the target task. For example, the input information creation unit 111 may have a table previously set up that associates words included in the target task with input information. The input information creation unit 111 may generate input information from the target task using a language model such as the processing model 210.

[0065] 1 to 8, the candidate processing unit 114 searches the input information database 116 for existing input information based on related information. The candidate processing unit 114 creates provisional processing input information by applying the input information from the input information creation unit 111 to each piece of searched existing input information. The generation example output unit 120 inputs the provisional processing input information into the processing model 210, thereby acquiring generation examples corresponding to each piece of existing input information.

[0066] The candidate processing unit 114 may embed input information in each piece of candidate input information. The candidate processing unit 114 may notify the input information creation unit 111 of what input information is used in each piece of candidate input information. The input information creation unit 111 may create input information in accordance with the content notified by the candidate processing unit 114. The candidate processing unit 114 may apply input information of the same item to multiple pieces of candidate input information, or may apply input information to be used in each piece of candidate input information for each piece of candidate input information.

[0067] The information acquisition unit 113 acquires user information related to the user, similar to the information acquisition unit 113 shown in Fig. 1. However, the information acquisition unit 113 in this example inputs the user information to the candidate processing unit 114.

[0068] The candidate processing unit 114 adds content related to the user information to the provisional process input information. The candidate processing unit 114 creates provisional process input information by applying the user information from the information acquisition unit 113 to each piece of searched candidate input information. The candidate processing unit 114 may embed the user information in each piece of candidate input information. The generation example output unit 120 inputs the provisional process input information to the processing model 210, thereby acquiring generation examples corresponding to each piece of candidate input information.

[0069] In the example shown in Fig. 4, user information (background information) according to the target task is included in the generation example request information. In this example, user information is embedded in each piece of candidate input information to create provisional processing information. This example also makes it possible to obtain a generation example when user information according to the user is applied to the candidate input information.

[0070] The providing unit 130 (see FIG. 1) of this example may provide the user with provisional processing input information including at least one of input information and user information as candidate input information. This allows more accurate candidate input information to be presented to the user. The providing unit 130 may also provide the user with existing input information in which neither input information nor user information is embedded.

[0071] 10 is a diagram showing an example of provisional process input information. The provisional process input information in this example includes a process definition portion, an output definition portion, and an input information portion. The process definition portion and the output definition portion are the same as the process definition portion and the output definition portion of the candidate input information. The input information portion of the provisional process input information is the portion of the candidate input information supplemented with user information, etc.

[0072] Fig. 11 is a diagram showing another example of the configuration of the task processing support device 100. The providing unit 130 is omitted from Fig. 11. The task processing support device 100 of this example includes an input information registration unit 115. The configuration other than the input information registration unit 115 is the same as any of the examples described in this specification.

[0073] The input information registration unit 115 acquires the final input information created by the user based on the candidate input information, and registers it as new existing input information in the input information database 116. The input information registration unit 115 may modify the final input information so that it matches the format required by the input information database 116, may delete part of the content of the final input information, or may add information to the final input information.

[0074] For example, if the input information database 116 records, for each piece of existing input information, output information extracted from information about the existing input information, which is information about the output when the existing input information is executed, the input information registration unit 115 creates output information corresponding to the final input information and registers it in the input information database 116. In this example, the input information database 116 classifies the content of each piece of existing input information into a process definition portion, an output definition portion, and an input information portion. In this case, the input information registration unit 115 classifies the final input information into a process definition portion, an output definition portion, and an input information portion and registers it in the input information database 116. If the input information database 116 records existing input information from which at least a portion of the input information portion has been omitted, the input information registration unit 115 may also register the final input information with at least a portion of the input information omitted. The classification of the content of the final input information may be performed by the input information registration unit 115 based on a preset algorithm, or may be performed by a language model such as the processing model 210.

[0075] The input information registration unit 115 may create related keywords indicating fields, industries, occupations, etc. related to the final input information and register them in the input information database 116 in association with the final input information. The related keywords may be registered as part of the input information portion of the final input information. The candidate processing unit 114 can efficiently acquire candidate input information by searching for candidate input information based on the related keywords. The creation of the related keywords may be performed by the input information registration unit 115 based on a preset algorithm, or may be performed by a language model such as the processing model 210.

[0076] 12 is a diagram showing an example of keyword generation information to be input to the processing model 210. The keyword generation information in this example is a prompt that generates related keywords for target information. The target information is final input information, but the target information in the example of FIG. 12 is different from the final input information in the examples of FIGS. 1 to 11.

[0077] The keyword generation information in this example includes third instruction information and target information. As described above, the target information includes at least a portion of the final input information to be registered in the input information database 116. The third instruction information is an instruction for causing the processing model 210 to generate related keywords related to the target information. The third instruction information may include information specifying the output format of the processing model 210. The content of the third instruction information may be set in advance in the input information registration unit 115.

[0078] Fig. 13 shows an example of related keywords generated by the processing model 210. The processing model 210 in this example generates the keyword information shown in Fig. 13 in accordance with the keyword generation information shown in Fig. 12. The keyword information includes related keywords in the output format specified by the keyword generation information. The input information registration unit 115 may register the related keywords indicated in the keyword information in the input information database 116 in association with the final input information.

[0079] 14 is a diagram showing an example of a structuring command to be input to the processing model 210. The structuring command in this example is a prompt that divides the target information into three parts: an output definition part, a process definition part, and an input definition part. The target information is the final input information, but the target information in the example of FIG. 14 is different from the final input information in the examples of FIGS. 1 to 13.

[0080] The structuring command in this example includes fourth instruction information and target information. As described above, the target information includes at least a portion of the final input information to be registered in the input information database 116. The fourth instruction information is an instruction for causing the processing model 210 to divide the target information into three parts: an output definition part, a process definition part, and an input definition part. The fourth instruction information may include information specifying the output format of the processing model 210. The content of the fourth instruction information may be preset in the input information registration unit 115.

[0081] The processing model 210 in this example classifies the contents of the target information into three parts, an output definition part, a process definition part, and an input definition part, and outputs them based on the structuring command shown in Fig. 14. The input information registration unit 115 may register the contents output by the processing model 210 in the input information database 116. Through this processing, the final input information can be registered in a format that is easy to search and cite.

[0082] 15 and 16 are diagrams illustrating an overview of the processing of the task processing system 10 described in FIGS. 1 to 14. FIGS. 15 and 16 illustrate an overview of the processing in the user, task processing assistance device 100, and processing model 210. The task processing system 10 may perform part of the processing described in FIGS. 15 and 16, or may perform additional processing.

[0083] In step S902, the user inputs a target task to the task processing assistance device 100. In step S902, the user may input the target task and a user ID to the task processing assistance device 100.

[0084] In step S904, the task processing assistance device 100 creates preprocessing input information including the target task, as shown in Fig. 7. The preprocessing input information is a prompt for causing the processing model 210 to generate related information for searching for existing input information related to the target task. The task processing assistance device 100 inputs the preprocessing input information to the processing model 210.

[0085] In step S906, the processing model 210 generates processing result information including related information as shown in FIG. 8 in accordance with the preprocessing input information. In step S908, the task processing support device 100 searches for existing input information based on the related information. The task processing support device 100 creates one or more pieces of candidate input information using the one or more pieces of searched existing input information. When multiple pieces of existing input information with similar content are searched, the task processing support device 100 may select one piece of existing input information and use it as the candidate input information. The task processing support device 100 may create generation example request information to which user information is added for each piece of candidate input information, as shown in FIG. 4.

[0086] In step S910, the processing model 210 outputs a generation example based on the generation example request information. In step S912, the task processing support device 100 collects generation examples corresponding to each piece of candidate input information. The task processing support device 100 associates each piece of candidate input information with the generation example and provides them to the user.

[0087] In step S912, the user refers to the provided candidate input information and generation examples to determine whether there is any candidate input information that the user wants to reference when creating the final input information. If there is any candidate input information that the user wants to reference, the process described in Fig. 16 is performed. If there is no candidate input information that the user wants to reference, the user may modify the target task, user information, etc., and repeat the process from S902, or may end the support process by the task processing support device 100.

[0088] In step S916, the user creates final input information by referencing the candidate input information. The user refers to the generated example and modifies the candidate input information as appropriate. This process can assist the user in creating the final input information. In step S918, the final input information created by the user is input to the task processing assistance device 100. The task processing assistance device 100 transmits the final input information to the processing model 210 and also saves the final input information.

[0089] In step S920, the processing model 210 executes the final input information to generate an answer. In step S922, the task processing support device 100 receives the answer from the processing model 210. The task processing support device 100 presents the answer to the user and saves the answer.

[0090] In step S924, the user determines whether the target task has been completed based on the presented answer. If the target task has been completed, the process proceeds to step S926 and subsequent steps. If the target task has not been completed, the user may select other candidate input information and repeat the process from step S916, or may end the assistance process by the task processing assistance device 100.

[0091] In step S926, the task processing assistance device 100 generates a structuring command as shown in Fig. 14 for classifying the final input information created by the user into parts. In step S926, the task processing assistance device 100 may further generate keyword generation information as shown in Fig. 12 for generating related keywords for the final input information created by the user. The keyword generation information may be included in the structuring command.

[0092] In step S928, the processing model 210 classifies the final input information into three parts: an output definition part, a process definition part, and an input definition part based on the structuring instructions. The processing model 210 may generate related keywords based on the keyword generation information.

[0093] In step S930, the task processing assistance device 100 registers the structured final input information and related keywords in the input information database 116. Through this process, the final input information created by the user can be registered as new existing input information.

[0094] The task processing system 10 described in Figures 1 to 16 may be realized by installing a program in one or more computers. The task processing assistance device 100 described in Figures 1 to 16 may be realized by installing a program in one or more computers. The processing model 210 described in Figures 1 to 16 may be realized by installing a program in one or more computers. These programs may be recorded on computer-readable media.

[0095] Various embodiments of the present invention may be described with reference to flowcharts and block diagrams, where the blocks may represent (1) stages of a process in which operations are performed or (2) sections of an apparatus responsible for performing the operations. Particular stages and sections may be implemented by dedicated circuitry, programmable circuitry provided with computer-readable instructions stored on a computer-readable medium, and / or a processor provided with computer-readable instructions stored on a computer-readable medium. Dedicated circuitry may include digital and / or analog hardware circuitry, and may include integrated circuits (ICs) and / or discrete circuits. Programmable circuitry may include reconfigurable hardware circuitry, including logical AND, OR, XOR, NAND, NOR, and other logical operations, flip-flops, registers, memory elements such as field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), and the like.

[0096] A computer-readable medium may include any tangible device capable of storing instructions that are executed by an appropriate device, such that the computer-readable medium having instructions stored thereon comprises an article of manufacture containing instructions that can be executed to create means for performing the operations specified in the flowcharts or block diagrams. Examples of computer-readable media may include electronic, magnetic, optical, electromagnetic, and semiconductor storage media. More specific examples of computer-readable media may include floppy disks, diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), compact disc read-only memory (CD-ROM), digital versatile disc (DVD), Blu-ray (RTM) disc, memory stick, integrated circuit card, and the like.

[0097] The computer readable instructions may include either assembler instructions, Instruction Set Architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, or source or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk®, JAVA®, C++, etc., and conventional procedural programming languages ​​such as the “C” programming language or similar programming languages.

[0098] The computer-readable instructions may be provided to a processor or programmable circuit of a programmable data processing device, such as a general-purpose computer, a special-purpose computer, or another computer, either locally or via a wide-area network (WAN) such as a local area network (LAN) or the Internet, which executes the computer-readable instructions to create means for performing the operations specified in the flowcharts or block diagrams. Here, the computer may be a personal computer (PC), a tablet computer, a smartphone, a workstation, a server computer, a general-purpose computer, a special-purpose computer, or the like, or may be a computer system in which multiple computers are connected. Such a computer system in which multiple computers are connected is also called a distributed computing system, and is a broad definition of computers. In a distributed computing system, multiple computers collectively execute a program by each executing a portion of the program and passing data between computers as needed during program execution.

[0099] Examples of processors include a computer processor, a central processing unit (CPU), a processing unit, a microprocessor, a digital signal processor, a controller, a microcontroller, etc. A computer may have one or more processors. In a multiprocessor system with multiple processors, each processor executes a portion of a program and passes data between processors as needed during program execution, allowing the multiple processors to collectively execute a program. For example, in multitasking, each of the multiple processors may execute a portion of each task in small chunks by switching tasks at time slice intervals. In this case, which portion of a program each processor executes changes dynamically. Which portion of a program each of the multiple processors executes may also be statically determined by multiprocessor-aware programming.

[0100] 17 illustrates an example of a computer 1200 in which aspects of the present invention may be embodied, in whole or in part. Programs installed on the computer 1200 may cause the computer 1200 to function as or perform operations associated with an apparatus or one or more sections of the apparatus according to embodiments of the present invention, and / or to perform a process or steps of a process according to embodiments of the present invention. Such programs may be executed by the CPU 1212 to cause the computer 1200 to perform specific operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

[0101] A computer 1200 according to this embodiment includes a CPU 1212, a RAM 1214, a graphics controller 1216, and a display device 1218, which are interconnected by a host controller 1210. The computer 1200 also includes input / output units such as a communication interface 1222, a storage device 1224 such as a hard disk drive, a DVD-ROM drive 1226, and an IC card drive, which are connected to the host controller 1210 via an input / output controller 1220. The computer also includes legacy input / output units such as a ROM 1230 and a keyboard 1242, which are connected to the input / output controller 1220 via an input / output chip 1240.

[0102] The CPU 1212 operates according to programs stored in the ROM 1230 and the RAM 1214, thereby controlling each unit. The graphics controller 1216 acquires image data generated by the CPU 1212 into a frame buffer or the like provided in the RAM 1214 or into the graphics controller 1216 itself, and causes the image data to be displayed on the display device 1218.

[0103] The communication interface 1222 communicates with other electronic devices via a network. The storage device 1224 stores programs and data used by the CPU 1212 in the computer 1200. The DVD-ROM drive 1226 reads programs or data from a DVD-ROM 1227 and provides the programs or data to the storage device 1224 via the RAM 1214. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0104] The ROM 1230 stores therein a boot program or the like that is executed by the computer 1200 upon activation, and / or programs that depend on the hardware of the computer 1200. The input / output chip 1240 may also connect various input / output units to the input / output controller 1220 via a parallel port, a serial port, a keyboard port, a mouse port, etc.

[0105] The programs are provided by a computer-readable medium such as a DVD-ROM 1227 or an IC card. The programs are read from the computer-readable medium, installed in the storage device 1224, RAM 1214, or ROM 1230, which are also examples of computer-readable media, and executed by the CPU 1212. Information processing described in these programs is read by the computer 1200, and causes cooperation between the programs and the various types of hardware resources described above. An apparatus or a method may be configured by implementing information manipulation or processing in accordance with the use of the computer 1200.

[0106] For example, when communication is performed between the computer 1200 and an external device, the CPU 1212 may execute a communication program loaded into the RAM 1214 and instruct the communication interface 1222 to perform communication processing based on the processing described in the communication program. Under the control of the CPU 1212, the communication interface 1222 reads transmission data stored in a transmission buffer processing area provided in the RAM 1214, the storage device 1224, the DVD-ROM 1227, or a recording medium such as an IC card, and transmits the read transmission data to the network, or writes reception data received from the network to a reception buffer processing area or the like provided on the recording medium.

[0107] The CPU 1212 may read all or a necessary portion of a file or database stored on an external recording medium such as the storage device 1224, the DVD-ROM drive 1226 (DVD-ROM 1227), an IC card, etc. into the RAM 1214, and perform various types of processing on the data on the RAM 1214. The CPU 1212 then writes back the processed data to the external recording medium.

[0108] Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and may undergo information processing. CPU 1212 may perform various types of processing on data read from RAM 1214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search / replacement, etc., as described throughout this disclosure and specified by the instruction sequences of the programs, and write the results back to RAM 1214. CPU 1212 may also search for information in a file, database, etc. on the recording medium. For example, if multiple entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored on the recording medium, CPU 1212 may search for an entry that matches a condition specified by the attribute value of the first attribute from among the multiple entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0109] The above-described programs or software modules may be stored in a computer-readable medium on or near the computer 1200. A recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can also be used as a computer-readable medium, thereby providing the programs to the computer 1200 via the network.

[0110] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included within the technical scope of the present invention.

[0111] It should be noted that the execution order of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before," "prior to," etc., and that the processes can be performed in any order unless the output of a previous process is used in a subsequent process. Even if the operational flow in the claims, specifications, and drawings is described using "first," "next," etc. for convenience, this does not mean that the processes must be performed in this order. [Explanation of symbols]

[0112] 10. Task processing system, 100. Task processing support device, 110. Candidate creation unit, 111. Input information creation unit, 112. Preprocessing unit, 113. Information acquisition unit, 114. Candidate processing unit, 115. Input information registration unit, 116. Input information database, 120. Generation example output unit, 130. Provision unit, 200. Task processing device, 210. Processing model

Claims

1. 1. A task processing support device that supports creation of final input information to be input to a processing model in order to cause the processing model to execute a target task input by a user, comprising: a candidate generation unit that generates one or more pieces of candidate input information based on the target task; a generation example output unit that outputs a generation example of the processing model when each of the candidate input information is input to the processing model; a providing unit that associates one or more pieces of candidate input information with the generation examples and provides them to the user; A task processing support device comprising:

2. The candidate creation unit a preprocessing unit that generates preprocessing input information for causing the processing model to generate related information related to the candidate input information based on the target task and inputs the preprocessing input information to the processing model; a candidate processing unit that generates one or more pieces of candidate input information based on the related information generated by the processing model; The task processing support device according to claim 1 , further comprising:

3. further comprising an input information database that records one or more pieces of existing input information; The candidate processing unit searches the input information database for one or more pieces of existing input information corresponding to the related information, and outputs one or more pieces of candidate input information based on the searched at least one piece of existing input information. The task processing support device according to claim 2 .

4. the input information database records, for each of the existing input information, output information that defines an output when the processing model executes the existing input information, among information regarding the existing input information; The candidate processing unit searches for the existing input information having the output information corresponding to the related information. The task processing support device according to claim 3 .

5. The preprocessing unit creates the preprocessing input information for generating keywords to be included in the output information based on the target task. The task processing support device according to claim 4 .

6. the candidate creation unit further includes an input information creation unit that creates input information by extracting information on an input to be included in the final input information from the target task; the candidate processing unit creates provisional processing input information by applying the input information to each of the existing input information searched from the input information database; The generation example output unit inputs the provisional processing input information into the processing model to obtain the generation examples corresponding to the respective existing input information. The task processing support device according to claim 4 .

7. the candidate creation unit further includes an information acquisition unit that acquires user information related to the user; The candidate processing unit adds content related to the user information to the provisional process input information. The task processing support device according to claim 6 .

8. The providing unit provides the existing input information to the user as the candidate input information. The task processing support device according to claim 6 .

9. The providing unit provides the provisional process input information to the user as the candidate input information. The task processing support device according to claim 6 .

10. The generation example output unit generates information about an input to be included in the final input information in accordance with the target task, and applies the information to the existing input information in the processing model. The task processing support device according to claim 3 .

11. further comprising an information acquisition unit that acquires user information related to the user; The generation example output unit reflects user information in the candidate input information and causes the processing model to generate a generation example. The task processing support device according to claim 3 .

12. The input information registration unit further includes: an input information registration unit that acquires the final input information created by the user based on the candidate input information and registers the final input information in the input information database; The task processing support device according to any one of claims 3 to 11.

13. the input information database records, for each of the existing input information, output information obtained by extracting information about an output when the existing input information is executed from information about the existing input information; The input information registration unit creates the output information corresponding to the final input information and registers it in the input information database. The task processing support device according to claim 12.

14. 1. A task processing support device that supports creation of final input information to be input to a processing model, in order to cause the processing model to execute a target task input by a user, comprising: a preprocessing unit that generates preprocessing input information based on the target task, for causing the processing model to generate related information related to candidate input information that is a candidate for the final input information, and inputs the preprocessing input information to the processing model; a candidate processing unit that generates one or more pieces of candidate input information based on the related information generated by the processing model; a providing unit for providing one or more pieces of candidate input information to the user; A task processing support device comprising:

15. a task processing support device according to claim 1 or 14; The processing model; A task processing system comprising:

16. A program for causing a computer to function as the task processing support device according to claim 1 or 14.

Images