Information processing apparatus, information processing method, and program

Abstract

To enable a person who does not necessarily have sufficient knowledge and experience related to development and operation of a machine learning model to generate an appropriate explanatory variable or a feature concerning a task of interest.SOLUTION: Prompt acquisition means acquires a prompt which is described in a natural language and includes a designation of a task. Input variable acquisition means interprets the prompt using a language model, acquires related information corresponding to the task designated by the prompt, and acquires an input variable to be used for prediction corresponding to the task designated, based on the related information. Output means outputs the input variable acquired.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] This disclosure relates to predictions using machine learning models.

Background Art

[0002] Various predictions are made using machine learning models. Generally, model designers such as data scientists select appropriate explanatory variables based on knowledge of the domain to which the model is applied and create a model to be used for prediction. However, selecting appropriate explanatory variables in model design requires sufficient knowledge and experience. Patent Document 1 discloses an explanatory variable proposal device that extracts explanatory variables effective for a target variable input by a user and presents them to the user.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] One object of the present invention is to enable a person who does not necessarily have sufficient knowledge and experience regarding the development and operation of a machine learning model to generate appropriate explanatory variables and features related to a target task.

Means for Solving the Problems

[0005] In one aspect of this disclosure, an information processing apparatus a prompt acquisition means for acquiring a prompt described in natural language and including a task specification, an input variable acquisition means for interpreting the prompt using a language model, acquiring relevant information corresponding to the task specified by the prompt, and acquiring input variables to be used for prediction corresponding to the specified task based on the relevant information, Output means for outputting the acquired input variables.

[0006] In another aspect of the present disclosure, an information processing method is executed by a computer, acquires a prompt described in natural language and including a task specification, interprets the prompt using a language model, acquires relevant information corresponding to the task specified by the prompt, acquires input variables to be used for prediction corresponding to the specified task based on the relevant information, and outputs the acquired input variables.

[0007] In still another aspect of the present disclosure, a program acquires a prompt described in natural language and including a task specification, interprets the prompt using a language model, acquires relevant information corresponding to the task specified by the prompt, acquires input variables to be used for prediction corresponding to the specified task based on the relevant information, and causes a computer to execute a process of outputting the acquired input variables.

Advantages of the Invention

[0008] According to the present disclosure, even a person who does not necessarily have sufficient knowledge and experience regarding the development and operation of a machine learning model can generate appropriate explanatory variables and feature quantities regarding a target task.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

[0010] Hereinafter, with reference to the drawings, preferred embodiments of the present disclosure will be described. <First Embodiment> [Overall Configuration] FIG. 1 shows the overall configuration of a machine learning model design support system (hereinafter, simply referred to as the “design support system”) to which the information processing apparatus according to the present disclosure is applied. The design support system 1 receives a task designation from a designer of a machine learning model (hereinafter, also referred to as a “user”), and generates input variables used by the machine learning model to execute the designated task. As shown in the figure, the design support system 1 includes a server device 10 and a terminal device 20. The server device 10 and the terminal device 20 can communicate via a wired or wireless network.

[0011] The terminal device 20 is operated by a user who designs and manages a machine learning model. The user specifies a target task for the terminal device 20. Although details will be described later, the terminal device 20 includes a natural language model that can interpret natural language. The user inputs a prompt including a task specification described in natural language to the terminal device 20. Note that a "prompt" refers to an instruction to a generative AI (Artificial Intelligence) including a natural language model and the like. The terminal device 20 receives the input of a prompt including a task specification, interprets the prompt using the natural language model, and recognizes the task specified by the user.

[0012] The server device 10 stores prediction-related information prepared in association with various tasks in a database or the like. The terminal device 20 accesses the database of the server device 10 and acquires prediction-related information related to the task specified by the user. The terminal device 20 uses the acquired prediction-related information to acquire input variables to be used in a machine learning model for executing the specified task. Here, an "input variable" is a concept indicating an explanatory variable or a feature amount used by a machine learning model to obtain a result corresponding to a specified task. For example, when the specified task is prediction of product demand by regression, the target variable is the product demand, and the explanatory variables in the prediction formula for obtaining the target variable become the input variables. Also, when the specified task is classification by a classification model, the feature amounts input to the classification model become the input variables.

[0013] Examples of tasks specified by the user include various tasks such as prediction tasks such as prediction of product demand and prediction of power demand, and classification tasks of input information. In the following description, it is assumed that the task specified by the user is prediction of product demand in a certain store.

[0014] The terminal device 20 acquires input variables suitable for predicting the demand for a product, which is the specified task, using prediction-related information. Specifically, the server device 10 selects input variables suitable for predicting the demand for a product from among a large number of input variables included in the prediction-related information. Also, the server device 10 can, if necessary, generate new input variables that are not included in the prediction-related information. The terminal device 20 outputs the acquired input variables as an answer to the prompt input by the user. In this way, the user can obtain input variables suitable for the machine learning model for executing the specified task by inputting in natural language to the server device 10.

[0015] [Natural language model] Here, the natural language model will be explained. The natural language model is one that has learned the relationships between words in a sentence and is a model that generates related strings related to the target string from the target string. By using a natural language model that has learned sentences and texts in various contexts, it is possible to generate related strings with appropriate content related to the target string. For example, the case of using a natural language model in question and answer will be explained. In this case, the natural language model receives as input the question "What kind of country is Japan?" as the target string and generates a string such as "Japan is an island country in the Northern Hemisphere..." as an answer to the question.

[0016] The learning method of the natural language model is not particularly limited. As an example, it may be one that is learned to output at least one sentence including the input string. For a specific example, the natural language model can be GPT (Generative Pre-Training) that outputs a sentence including the input string by predicting a highly probable string following the input string. In addition to this, as the natural language model, for example, T5 (Text-to-Text Transfer Transformer), BERT (Bidirectional Encoder Representations from Transformers), RoBERTa (Robustly optimized BERT approach), ELECTRA (Efficiently Learning an Encoder that Classifies Token Replacements Accurately), etc. can also be used. In the present embodiment, a large language model may be used as the natural language model. Also, the natural language model may be one that can access the Internet or other specialized knowledge bases to obtain information.

[0017] [Hardware Configuration] (Server Device) FIG. 2 is a block diagram showing the hardware configuration of the server device 10. As shown in the figure, the server device 10 includes a processor 11, an interface (IF) 12, a ROM (Read Only Memory) 13, a RAM (Random Access Memory) 14, a database (DB) 15, and a recording medium 16. Each component is connected to each other through, for example, a bus 18.

[0018] The processor 11 is a computer such as a CPU (Central Processing Unit), and controls the entire server device 10 by executing a pre-prepared program. Specifically, as the processor 11, a CPU, GPU (Graphics Processing Unit), DSP (Digital Signal Processor), MPU (Micro Processing Unit), FPU (Floating Point number Processing Unit), PPU (Physics Processing Unit), TPU (TensorProcessingUnit), quantum processor, microcontroller, or a combination thereof can be used.

[0019] Also, the processor 11 loads the program stored in the ROM 13 and the recording medium 16 into the RAM 14, and executes each process coded in the program. The processor 11 functions as part or all of the server device 10.

[0020] The IF 12 transmits and receives data to and from an external device. Specifically, the server device 10 transmits and receives information to and from the terminal device 20 through the IF 12.

[0021] The ROM 13 stores various programs executed by the processor 11. The RAM 14 is used as a working memory during the execution of various processes by the processor 11. The DB 15 stores prediction-related information for various tasks. Note that the prediction-related information will be described later.

[0022] The recording medium 16 is a non-volatile and non-transitory recording medium such as a disk-shaped recording medium or a semiconductor memory. The recording medium 16 may be configured to be detachable from the server device 10. The recording medium 16 stores various programs executed by the processor 11.

[0023] (Terminal device) FIG. 3 is a block diagram showing the hardware configuration of the terminal device 20. The terminal device 20 is, for example, a PC or a tablet terminal. As shown in the figure, the terminal device 20 includes a processor 21, an IF 22, a ROM 23, a RAM 24, an input unit 26, and a display unit 27. Each component is connected to each other, for example, through a bus 28.

[0024] The processor 21 is a computer such as a CPU, and controls the entire terminal device 20 by executing a program prepared in advance. Note that the processor 21 may be a GPU, an FPGA, a DSP, an ASIC, or the like.

[0025] The IF 22 transmits and receives data to and from an external device. Specifically, the terminal device 20 accesses the DB 15 of the server device 10 through the IF 22 and acquires prediction-related information.

[0026] The ROM 23 stores various programs executed by the processor 21. The RAM 24 is used as a work memory during the execution of various processes by the processor 21.

[0027] The recording medium 25 is a non-volatile and non-transitory recording medium such as a disk-shaped recording medium or a semiconductor memory. The recording medium 25 may be configured to be detachable from the terminal device 20. The recording medium 25 stores various programs executed by the processor 21.

[0028] The input unit 26 is an input device such as a keyboard, a mouse, or a touch panel. The user inputs a prompt including task specification by operating the input unit 26. The display unit 27 is a display or the like that performs display based on the control of the processor 21. The answer to the prompt input by the user is displayed on the display unit 27 and presented to the user.

[0029] [Functional Configuration] FIG. 4 is a block diagram showing the functional configurations of the server device 10 and the terminal device 20. The server device 10 includes a prediction formula DB 151, an explanatory variable DB 152, and a processing method DB 153 in the database 15. On the other hand, the terminal device 20 includes an input variable proposal unit 28 in addition to the aforementioned input unit 26 and display unit 27.

[0030] As described above, the database 15 stores prediction-related information. The prediction-related information includes a prediction formula, an explanatory variable, and a processing method. The prediction formula DB 151 stores the prediction formula, the explanatory variable DB 152 stores the explanatory variable, and the processing method DB 153 stores the processing method.

[0031] FIG. 5 shows an example of the prediction formula data stored in the prediction formula DB 151. The prediction formula DB 151 stores prediction formula data indicating the prediction formula corresponding to each task for each task. In the example of FIG. 5, the prediction formula DB 151 stores a plurality of prediction formula data used for the task "product demand prediction". Specifically, the prediction formula data includes a prediction formula ID, an explanatory variable, a prediction formula, and accuracy. The prediction formula ID is identification information for each prediction formula. The explanatory variable indicates the variable used in each prediction formula. The prediction formula is a mathematical expression of each prediction formula. The accuracy indicates the accuracy of the prediction by each prediction formula.

[0032] FIG. 6 shows an example of the explanatory variable data stored in the explanatory variable DB 152. The explanatory variable DB 152 stores explanatory variable data indicating the explanatory variables used for the prediction corresponding to each task for each task. In the example of FIG. 6, the explanatory variable DB 152 stores a plurality of explanatory variable data used for the task "product demand prediction". Specifically, the explanatory variable data includes the relationship between the explanatory variable and the target variable, and a numerical example. The relationship with the target variable is information indicating the relationship between each explanatory variable and the target variable of the task (product demand in this example). The numerical example shows an example of the numerical value that each explanatory variable can take.

[0033] FIG. 7 shows an example of the processing method data stored in the processing method DB153. The processing method DB153 stores, for each task, the processing method data indicating the processing method of the explanatory variables used for the prediction corresponding to that task. In the prediction formula, variables obtained by processing each explanatory variable in a predetermined method may be used. For example, for the explanatory variable "sales", variables can be defined from different perspectives such as "daily sales" and "average sales in the past 7 days" according to the period, and used for prediction. Variables processed for use in prediction are thus called "processed variables" or "features". The processing method DB153 indicates the processing method corresponding to each explanatory variable. That is, the processing method data indicates how each explanatory variable is processed and used in the prediction model. Examples of the processing method of the explanatory variables include a method of defining arithmetic operations between a plurality of explanatory variables and a method of calculating a predetermined statistical value such as a moving average.

[0034] In the example of FIG. 7, the processing method DB153 stores the processing method data of a plurality of explanatory variables used for the task of "product demand prediction". Specifically, the processing method data includes an explanatory variable, a processing method, and a numerical example of the feature amount. The processing method indicates how each explanatory variable is processed and used. The numerical example of the feature amount indicates the numerical example of the feature amount obtained by processing the explanatory variable.

[0035] Returning to FIG. 4, in the terminal device 20, the input unit 26 receives the prompt input by the user and outputs it to the input variable proposal unit 28. The input variable proposal unit 28 interprets the input prompt using a natural language model and recognizes the task specified by the user. Further, by interpreting the prompt, the input variable proposal unit 28 recognizes additional information other than the task specification, for example, conditions for obtaining input variables. Next, the input variable proposal unit 28 accesses the database 15 of the server device 10, refers to each of the DBs 151 to 153, and obtains input variables suitable for the prediction model corresponding to the specified task. Then, the input variable proposal unit 28 displays the obtained input variables on the display unit 27. Thereby, the user can obtain an answer to the input prompt. That is, the user can know input variables suitable for the model that executes the task by inputting a prompt that specifies the task in natural language.

[0036] Specifically, the input variable proposal unit 28 refers to the prediction formula DB 151, selects one or more prediction formulas used for the specified task, and obtains the explanatory variables used in the prediction formula as input variables. Further, the input variable proposal unit 28 refers to the explanatory variable DB 152 to obtain the relationship between the explanatory variable and the task and numerical examples of each explanatory variable, and may present them to the user. Further, when the obtained prediction formula includes a feature amount obtained by processing the explanatory variable, the input variable proposal unit 28 may refer to the processing method DB 153 and present to the user the explanation (processing method) of the feature amount, numerical examples of the feature amount, and the like. Thereby, the user can know the reason and basis for which each input variable is proposed.

[0037] The input variable proposal unit 28 may generate a new prediction formula not included in the existing prediction formula data. For example, as shown in FIG. 8, the input variable proposal unit 28 may generate a new explanatory variable X5 using the existing explanatory variables X1 to X4 and generate a new prediction formula 5 using the explanatory variable X5. In this case, the explanatory variable X5 may be calculated, for example, by arithmetic operations using two or more of the explanatory variables X1 to X4.

[0038] In addition, the input variable proposal unit 28 may generate a new processing method that is not included in the existing processing method data. For example, as shown in FIG. 9, the input variable proposal unit 28 may generate a feature amount "number of consecutive holiday days" based on the explanatory variable "holiday / special day". Further, the input variable proposal unit 28 may generate a feature amount "average sales per person" based on the explanatory variables "sales" and "number of customers". Further, the input variable proposal unit 28 may generate a feature amount "event scale" based on the explanatory variable "nearby events / activities".

[0039] In this way, by generating new prediction formulas, explanatory variables, feature amounts, etc. based on existing prediction formulas, explanatory variables, feature amounts, etc., it becomes possible to propose more appropriate input variables in various situations. Further, by registering the new prediction formulas, feature amounts, etc. generated by the input variable proposal unit 28 in the plurality of terminal devices 20 in the DB 15 of the server device 10, the latest prediction-related information can be shared among the users of the plurality of terminal devices 20.

[0040] [Operation example] Next, an operation example of the support system 1 will be described. FIGS. 10(A) to 10(D) show examples of prompts input by the user.

[0041] Basically, if the user specifies at least a task in the input prompt, the user can receive a proposal of input variables suitable for the task. In the example of FIG. 10(A), the user specifies the task "product demand prediction" and requests a proposal of input variables. In this case, the input variable proposal unit 28 can output an appropriate number of prediction formulas, the input variables used therein, the relevance of each input variable to the target variable, etc. When presenting a plurality of prediction formulas, the input variable proposal unit 28 may sort and output them by accuracy, etc. Further, when proposing a plurality of input variables, the input variable proposal unit 28 may sort and output them by the relationship with the target variable, the degree of correlation with the target variable, the actual availability of the data of the input variable, etc.

[0042] The user can describe additional information such as conditions regarding the proposal of input variables in the input prompt. For example, as shown in the example of FIG. 10(B), the user may present candidates for input variables. In this case, the input variable proposal unit 28 may output whether each candidate for the input variable presented by the user is appropriate, along with the reasons and bases therefor. Further, the input variable proposal unit 28 may propose alternative input variables that are more appropriate than the candidates presented by the user.

[0043] Also, instead of presenting candidates for input variables, the user may present the input variables actually being used and request a proposal for additional input variables to be used.

[0044] The user may also specify matters to be considered as conditions regarding the proposal of input variables. For example, in the example of FIG. 10(C), for the task of predicting product demand, the user has added the condition of "increasing the prediction accuracy during consecutive holidays". In this case, the input variable proposal unit 28 may refer to prediction-related information and make a proposal that emphasizes explanatory variables and features particularly related to consecutive holidays. For example, as shown in FIG. 7, assume that in the processing method DB153, the prompt of FIG. 10(C) is input when the feature "number of consecutive holiday days" does not exist for the explanatory variable "holiday / special day". In this case, based on the prompt of FIG. 10(C), the input variable proposal unit 28 may generate a new feature "number of consecutive holiday days" for the explanatory variable "holiday / special day" as shown in FIG. 9 and propose it as an input variable.

[0045] Also, in the example of FIG. 10(D), the user has attached the condition of "considering the influence by events" to the task of predicting commodity demand. In this case, the input variable proposal unit 28 may refer to the prediction-related information and make a proposal that emphasizes explanatory variables and features particularly related to events. For example, as shown in FIG. 7, in the processing method DB153, assume that the prompt of FIG. 10(D) is input in a state where the feature "event scale" does not exist for the explanatory variable "adjacent events / occasions". In this case, based on the prompt of FIG. 10(D), the input variable proposal unit 28 may generate a new feature "event scale" for the explanatory variable "adjacent events / occasions" as shown in FIG. 9 and propose it as an input variable.

[0046] [Input Variable Proposal Process] Next, the flow of the input variable proposal process will be described. FIG. 11 is a flowchart of the input variable proposal process. This process is realized by the processor 21 shown in FIG. 3 executing a pre-prepared program and operating as the element shown in FIG. 4.

[0047] First, the terminal device 20 receives the prompt generated by the user (step S51). Next, the terminal device 20 interprets the prompt using the natural language model and acquires the prediction-related information regarding the specified task from the DB15 of the server device 10 (step S52). Next, when the prompt includes conditions, the terminal device 20 acquires explanatory variables, features, etc. suitable for the specified task as input variables in consideration of those conditions (step S53). At this time, the terminal device 20 generates a new prediction formula, explanatory variable, feature, etc. as necessary. Then, the terminal device 20 displays the acquired input variables on the display unit 27 (step S54). Thus, the input variable proposal process ends.

[0048] [Modification Example] Next, a modification example of the above embodiment will be described. The following modification examples can be applied in appropriate combinations.

[0049] (Modification Example 1) In the above embodiment, the input variable proposal unit 28 including the natural language model is provided in the terminal device 20. Instead, one input variable proposal unit may be provided in the server device 10, and the input variable proposal unit may propose the input variable. In this case, the terminal device 20 transmits the prompt input by the user to the server device 10. In the server device 10, the input variable proposal unit may access the DB 15, refer to the prediction-related information, obtain the input variable suitable for the task specified by the user, and transmit it to the terminal device 20.

[0050] (Modification Example 2) The prediction formula DB 151, the explanatory variable DB 152, and the processing method DB 153 stored in the DB 15 of the server device 10 may store data described in natural language. In this case, the input variable proposal unit proposes an appropriate input variable by using the prompt described in natural language by the user and the prediction-related information described in natural language.

[0051] <Second Embodiment> FIG. 12 is a block diagram showing the functional configuration of the information processing apparatus according to the second embodiment. The information processing apparatus 70 includes a prompt acquisition unit 71, an input variable acquisition unit 72, and an output unit 73.

[0052] FIG. 13 is a flowchart of the process by the information processing apparatus according to the second embodiment. The prompt acquisition unit 71 acquires a prompt described in natural language and including a task specification (step S71). The input variable acquisition unit 72 interprets the prompt using a language model, acquires related information corresponding to the task specified by the prompt, and acquires an input variable to be used for prediction corresponding to the specified task based on the related information (step S72). The output unit 73 outputs the acquired input variable (step S73).

[0053] According to the information processing apparatus 70 of the second embodiment, even a person who does not necessarily have sufficient knowledge and experience regarding the development and operation of the machine learning model can obtain appropriate explanatory variables and feature amounts regarding the target task.

[0054] Some or all of the above embodiments may be described as follows in the appended claims, but are not limited thereto.

[0055] (Appendix 1) A prompt acquisition means for acquiring a prompt described in natural language and including a task specification, An input variable acquisition means for interpreting the prompt using a language model, acquiring related information corresponding to the task specified by the prompt, and acquiring input variables to be used for prediction corresponding to the specified task based on the related information, An output means for outputting the acquired input variables, An information processing apparatus comprising the above.

[0056] (Appendix 2) The information processing apparatus according to Appendix 1, wherein the input variable acquisition means generates a new input variable not included in the related information.

[0057] (Appendix 3) The related information is described in natural language, The information processing apparatus according to Appendix 1, wherein the related information includes a prediction formula corresponding to the specified task, input variables used for prediction corresponding to the specified task, and a processing method of the input variables.

[0058] (Appendix 4) The information processing apparatus according to Appendix 3, wherein the input variable acquisition means generates a new input variable from a plurality of existing input variables included in the related information based on the prompt and the related information.

[0059] (Appendix 5) The information processing apparatus according to Appendix 4, wherein the input variable acquisition means generates a new prediction formula using the new input variable.

[0060] (Appendix 6) The input variable acquisition means is an information processing apparatus according to Supplementary Note 3 that generates a new processing method for the input variable and a new input variable obtained by the processing method based on the prompt and the related information.

[0061] (Supplementary Note 7) The related information is stored in a storage unit, the input variable acquisition means acquires the related information from the storage unit, The input variable acquisition means is an information processing apparatus according to Supplementary Note 6 that stores the generated new processing method and new input variable in the storage unit.

[0062] (Supplementary Note 8) The prompt includes conditions related to the task, the input variable acquisition means is an information processing apparatus according to Supplementary Note 1 that acquires an input variable that satisfies the conditions.

[0063] (Supplementary Note 9) Executed by a computer, acquires a prompt described in natural language and including a task specification, interprets the prompt using a language model, acquires related information corresponding to the task specified by the prompt, acquires an input variable to be used for prediction corresponding to the specified task based on the related information, An information processing method for outputting the acquired input variable.

[0064] (Supplementary Note 10) acquires a prompt described in natural language and including a task specification, interprets the prompt using a language model, acquires related information corresponding to the task specified by the prompt, acquires an input variable to be used for prediction corresponding to the specified task based on the related information, A program for causing a computer to execute a process of outputting the acquired input variable.

[0065] The present disclosure has been described with reference to the embodiments and examples, but the present disclosure is not limited to the above embodiments and examples. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure.

Explanation of Signs

[0066] 10 Server device 20 Terminal device 21 Processor 26 Input unit 27 Display unit 28 Input variable proposal unit 15 Database (DB) 151 Prediction formula DB 152 Explanatory variable DB 153 Processing method DB

Claims

1. prompt acquisition means for acquiring a prompt described in natural language and including a task specification; input variable acquisition means for interpreting the prompt using a language model, acquiring relevant information corresponding to the task specified by the prompt, and acquiring input variables to be used for prediction corresponding to the specified task based on the relevant information; output means for outputting the acquired input variables; An information processing apparatus comprising:

2. The information processing apparatus according to claim 1, wherein the input variable acquisition means generates new input variables not included in the relevant information.

3. The relevant information is described in natural language, The information processing apparatus according to claim 1, wherein the relevant information includes a prediction formula corresponding to the specified task, input variables used for prediction corresponding to the specified task, and a method for processing the input variables.

4. The information processing apparatus according to claim 3, wherein the input variable acquisition means generates new input variables from a plurality of existing input variables included in the relevant information based on the prompt and the relevant information.

5. The information processing apparatus according to claim 4, wherein the input variable acquisition means generates a new prediction formula using the new input variables.

6. The information processing apparatus according to claim 3, wherein the input variable acquisition means generates a new processing method for the input variables and new input variables obtained by the processing method based on the prompt and the relevant information.

7. The relevant information is stored in a storage unit, The input variable acquisition means acquires the relevant information from the storage unit, The information processing apparatus according to claim 6, wherein the input variable acquisition means stores the generated new processing method and new input variables in the storage unit.

8. The prompt includes conditions related to the task, The information processing apparatus according to claim 1, wherein the input variable acquisition means acquires input variables that satisfy the conditions.

9. Executed by a computer, acquiring a prompt described in natural language and including a task specification, interpreting the prompt using a language model, acquiring relevant information corresponding to the task specified by the prompt, and acquiring input variables to be used for prediction corresponding to the specified task based on the relevant information, An information processing method for outputting the acquired input variables.

10. acquiring a prompt described in natural language and including a task specification, Interpret the prompt using a language model, obtain relevant information corresponding to the task specified by the prompt, and obtain input variables used for prediction corresponding to the specified task based on the relevant information. A program that causes a computer to execute a process of outputting the obtained input variables.

Images