Information providing apparatus, information providing method, and information providing program

Abstract

To improve usability of a data analysis tool.SOLUTION: An information providing device 10 includes an interface unit 11, a prompt generating unit 13, a text generating unit 16, and a level selecting unit 17. The level selecting unit 17 selects a level related to data analysis of a user. The prompt generating unit 13 generates a prompt for causing a language model to generate text according to the level for supporting the use of a data analysis tool, based on a request received from a terminal device on which the data analysis tool is executed and the level selected by the level selecting unit 17. The text generating unit 16 generates the text by inputting the prompt generated by the prompt generating unit 13 into the language model. The interface unit 11 causes the terminal device to display the text generated by the text generating unit 16.SELECTED DRAWING: Figure 1

Description

Technical Field

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

Background Art

[0002] In recent years, so-called no-code data analysis tools (for example, applications) may be provided (for example, refer to Patent Document 1 or Non-Patent Document 1). No-code data analysis tools are characterized in that programming skills are not required and even beginners can use them intuitively.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Non-Patent Documents

[0004]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, there is a problem that conventional data analysis tools may be difficult for users to use.

[0006] For example, even for a no-code data analysis tool, in order to master it, it is necessary to understand the concepts and methods of data analysis. When a user understands the concepts and methods related to data analysis during tool usage, they may conduct investigations through the Internet, books, etc., or by querying experts. On the other hand, if the investigation results obtained by the user do not match their skill level, the investigation may become meaningless.

[0007] Especially for beginners in data analysis, it is difficult to distinguish whether the results of the investigation they conducted on analysis methods, etc. are basic content or applied content. For this reason, the problem that it is not easy to know whether the investigation results really match their skill level can easily occur.

[0008] The same applies to the usage method of the analysis tool. For example, even if a beginner in data analysis is taught about the settings of a tool for advanced users, it will cause confusion.

Means for Solving the Problem

[0009] In order to solve the above-mentioned problems and achieve the purpose, the information providing device includes a level selection unit that selects the level of the user related to data analysis, a request received from a terminal device on which a data analysis tool is executed, and based on the level selected by the level selection unit, a sentence for assisting the use of the data analysis tool, and a prompt generation unit that generates a prompt for causing a language model to generate a sentence corresponding to the level, a sentence generation unit that generates a sentence by inputting the prompt generated by the prompt generation unit into the language model, and a display control unit that causes the terminal device to display the sentence generated by the sentence generation unit.

Effect of the Invention

[0010] According to the present invention, the usability of the data analysis tool can be improved.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Mode for Carrying Out the Invention

[0012] Hereinafter, embodiments of the information providing apparatus, information providing method, and information providing program according to the present application will be described in detail with reference to the drawings. Note that the present invention is not limited to the embodiments described below.

[0013] [First Embodiment] [Configuration of the First Embodiment] The configuration of the information providing system will be described with reference to FIG. 1. FIG. 1 is a diagram showing a configuration example of an information providing system according to the first embodiment. As shown in FIG. 1, the information providing system 1 includes an information providing apparatus 10 and a terminal apparatus 20.

[0014] The information providing apparatus 10 is a server connected to the terminal apparatus 20 via a network. The terminal apparatus 20 is a PC (Personal Computer), a tablet terminal, a smartphone, or the like.

[0015] The information providing device 10 has the functions of data analysis tools such as Node-AI described in Non-Patent Document 1. The information providing device 10 may have functions equivalent to those of the information processing device described in Patent Document 1. For example, the user executes a data analysis tool via the terminal device 20 to generate canvas data.

[0016] In Node-AI, Flow-based programming, which enables the design of a pipeline for processing data without coding, i.e., no-code, is adopted. Also, in Node-AI, by connecting the data to be processed and the functions that can execute each process, the pipeline for processing the data can be intuitively described. In the present embodiment, a function for executing a predetermined process on data is referred to as a "card". Also, data on which a pipeline is described is referred to as "canvas data".

[0017] In addition to the functions of the data analysis tool, the information providing device 10 has a function of providing the user with natural language sentences. The information providing device 10 supports the user's use of the data analysis tool by providing messages.

[0018] The information providing device 10 provides, in text, answers to the user's questions regarding the data analysis tool, hints on how to use the data analysis tool, actions recommended to the user, insights regarding the data, pointing out of the user's mistakes, etc. At this time, the information providing device 10 provides text according to the user's level regarding data analysis.

[0019] As shown in FIG. 1, the information providing device 10 includes an interface unit 11, a request processing unit 12, a prompt generation unit 13, a history DB 14, a language model group 15, a text generation unit 16, and a level selection unit 17.

[0020] The interface unit 11 exchanges data with the terminal device 20. For example, the interface unit 11 receives requests from the terminal device 20. Also, for example, the interface unit 11 causes the terminal device 20 to display a screen.

[0021] The interface unit 11 receives, as a request, the text of an inquiry input by the user or the operation content of the data analysis tool from the terminal device 20. The interface unit 11 transfers the received request to the request processing unit 12.

[0022] When the request is the operation content of the data analysis tool, the request processing unit 12 obtains the result for the request. The result is the canvas data returned by the data analysis tool for the operation content.

[0023] The request processing unit 12 stores the request and the result in the history DB 14. Also, the request processing unit 12 transfers the request and the result to the prompt generation unit 13 and the level selection unit 17. When there is no result, the request processing unit 12 stores the request in the history DB 14 and also transfers the request to the prompt generation unit 13 and the level selection unit 17.

[0024] The level selection unit 17 selects the level related to the user's data analysis. The level is information that represents step by step the amount of knowledge related to the user's data analysis, the proficiency in the data analysis tool, etc. Here, it is assumed that the level is represented by two levels: advanced and beginner. However, the number of levels may be three or more.

[0025] The level selection unit 17 selects the level based on the input by the user. In this case, the interface unit 11 receives the input of the level via the terminal device 20.

[0026] Also, the level selection unit 17 may automatically select the level. For example, the level selection unit 17 selects the level based on the past operation content of the user in the data analysis tool and the accuracy of the data analysis by the process generated according to the past operation content.

[0027] Based on the request (or request and result) received from the terminal device 20 on which the data analysis tool is executed, the prompt generation unit 13 generates a prompt for causing a language model to generate a sentence for assisting in the use of the data analysis tool. The prompt generation unit 13 delivers the generated prompt to the sentence generation unit 16.

[0028] Based on the request received from the terminal device on which the data analysis tool is executed and the level selected by the level selection unit 17, the prompt generation unit 13 generates a prompt for causing a language model to generate a sentence for assisting in the use of the data analysis tool, the sentence being corresponding to the level.

[0029] The sentence generation unit 16 generates a sentence by inputting the prompt generated by the prompt generation unit 13 into the language model included in the language model group 15. The sentence generation unit 16 outputs the obtained sentence to the terminal device 20 via the interface unit 11.

[0030] Also, as shown in FIG. 1, the language model group 15 includes an advanced user-oriented language model 151 and a beginner-oriented language model 152. The advanced user-oriented language model 151 is a language model corresponding to the advanced user level. The beginner-oriented language model 152 is a model corresponding to the beginner level.

[0031] The sentence generation unit 16 can generate a sentence by inputting the prompt into the language model corresponding to the level selected by the level selection unit 17 among the language models corresponding to each of the plurality of levels. The language model to be input may be specified by the prompt.

[0032] Even when the same prompt is input to the advanced user-oriented language model 151 and the beginner-oriented language model 152, the output sentences are different from each other.

[0033] The level selection unit 17 further selects a role. The prompt generation unit 13 generates a prompt for causing the language model to generate a sentence in a style according to the role. For example, the style is whether it is the desu / masu style or the de aru style, whether honorific language is used, whether interjections are used, whether it is an imperative tone, etc.

[0034] The role may be selected by the user via a screen as shown in FIG. 2. FIG. 2 is a diagram showing an example of a screen. The interface unit 11 causes the terminal device 20 to display the role selection screen 54 of FIG. 2. "Instructor", "Secretary", "University Professor" are examples of roles. The user selects one of the roles displayed on the role selection screen 54. The selected role is notified to the prompt generation unit 13 by the level selection unit 17.

[0035] Here, the language models included in the language model group 15 are large language models (LLMs). The language model may be a neural network. Also, the language model is, for example, ChatGPT (registered trademark) (reference URL: https: / / openai.com / blog / chatgpt).

[0036] The language model executes inference based on the input prompt and outputs a natural language sentence obtained by the inference.

[0037] Here, the data analysis tool generates a series of processes (pipelines) related to data analysis using a machine learning model according to the user's operation. The machine learning model for performing data analysis is called an analysis model. For example, the data analysis tool outputs an executable file that executes a series of processes.

[0038] The process can be divided into three: preprocessing of data, learning of the analysis model using the data (updating of parameters), and evaluation of the learned analysis model.

[0039] The language model may include language models tuned for each of the three processes. For example, the language model group 15 includes a language model tuned for preprocessing, a language model tuned for learning, and a language model tuned for evaluation.

[0040] For example, the language model tuned for preprocessing learns a prompt based on the operation content specifying the preprocessing for the data used in learning and the advice in the text by veteran users for the operation content. The language model tuned for preprocessing can generate more appropriate text regarding preprocessing compared to other language models.

[0041] Note that the way of dividing the processes of the data analysis tool is not limited to the above, and it may be further subdivided. And the language model group 15 includes language models tuned for each of the subdivided processes. For example, the language model group 15 may include language models tuned for each of the cards in the data analysis tool.

[0042] Also, the language model group 15 may include language models that are subdivided by process and correspond to each of the multiple levels. For example, the language model group 15 includes six language models: a beginner-level language model tuned for preprocessing, a beginner-level language model tuned for learning, a beginner-level language model tuned for evaluation, an advanced-level language model tuned for preprocessing, an advanced-level language model tuned for learning, and an advanced-level language model tuned for evaluation.

[0043] The text generation unit 16 selects an appropriate language model based on the prompt and generates text using the selected language model. For example, when the keyword "preprocessing" is included in the prompt and the user's level is a beginner, the text generation unit 16 selects a beginner-level language model that has been tuned according to the preprocessing.

[0044] Hereinafter, the operation of the information providing apparatus 10 will be described in detail by dividing it into the case where the request is the text of an inquiry input by the user and the case where the request is the operation content of the data analysis tool.

[0045] (1. When the request is the text of an inquiry input by the user) As shown in FIG. 3, on the screen (user interface) of the data analysis tool, canvas data created according to the operation content is displayed. FIG. 3 is a diagram showing an example of the screen. The screen is displayed by the terminal device 20 according to the control of the information providing apparatus 10.

[0046] The canvas data represents a series of processes (pipeline) related to data analysis using the analysis model. First, in the canvas data, the data and functions specified by the user's operation are arranged as cards.

[0047] In the example of FIG. 3, card 511, card 512, and card 513 are arranged. Card 511 represents the data input to the analysis model. Card 512 and card 513 represent the preprocessing of the data. Also, comments regarding the cards are displayed in areas 521 and 522.

[0048] The chat screen 53 is a screen for displaying the text generated by the information providing apparatus 10. The chat screen 53 may be displayed as a pop-up overlapping the screen for displaying the canvas data. For example, the interface unit 11 functions as a display control unit that causes the terminal device 20 to display the canvas data and the chat screen 53.

[0049] In box 5311, the text of the inquiry input by the user is displayed. Note that the text of the inquiry is input via input field 533. Icon 532 is an image indicating the pseudo-speaker (bot) when outputting the text generated by information providing device 10 as a statement in the chat. Also, in box 5312, the text generated by information providing device 10 is displayed.

[0050] In the example of FIG. 3, the user inputs the text "Please tell me the difference between lasso and ridge." In response, information providing device 10 generates and outputs the text "What the hell, don't you know the difference between lasso and ridge!? Have you been studying properly, you bastard!? First of all, lasso is an abbreviation for Least Absolute Shrinkage and Selection Operator, and it is a regularization method that selects features by approaching the coefficient to 0. On the other hand, ridge is a method that uses L2 regularization to prevent overfitting. That is, lasso is used when you want to remove specific features, and ridge is suitable when you want to stabilize the entire model. Do you understand, you idiot!?"

[0051] The prompt generation unit 13 receives, as a request, the text input via terminal device 20 by the user who uses the data analysis tool. The prompt generation unit 13 generates a prompt based on the text input by the user. The prompt generation unit 13 generates a prompt including the following elements. (1) Instruction: A specific task or instruction to be executed by the language model (2) Context: External information or additional context to lead the model to a better response (3) Input data: The request (4) Output indicator: The type or format of the output

[0052] In the example of FIG. 3, the request is the text input by the user. For example, when text is input into the input field 533 of the chat screen 53, the request processing unit 12 determines that the request is the text input by the user, and notifies the determination result to the prompt generation unit 13.

[0053] When the request is the text input by the user, the prompt generation unit 13 sets element (1) as "generation of a sentence for answering the user". The prompt generation unit 13 may determine that the request is the text input by the user when the user's speech displayed on the chat screen 53 (for example, the text in the box 5311) matches the request.

[0054] Also, the prompt generation unit 13 determines element (2) according to the following procedure. First, the prompt generation unit 13 extracts terms related to predetermined data analysis from the text input by the user. Assume that "lasso" and "ridge" are terms related to predetermined data analysis. Then, the prompt generation unit 13 adds the fact that "lasso" and "ridge" are terms related to data analysis as external information or additional context to element (2).

[0055] Thereby, for example, it is possible to avoid the language model generating a sentence answering the difference as mere character strings of "lasso" and "ridge", or generating a sentence explaining the difference in meaning listed in the dictionary, and to prompt the language model to generate a sentence explaining the difference in data analysis. That is, element (2) makes it easier to provide the information required by the user.

[0056] Element (3) is the text itself input by the user. However, the prompt generation unit 13 may correct obvious misnotations (such as spelling mistakes, typos, etc.) and then set the text as element (3).

[0057] Further, the prompt generation unit 13 sets the element (4) as, for example, "a natural language sentence". Further, the prompt generation unit 13 may set the element (4) including a language such as "an English sentence" or "a Japanese sentence". For example, the prompt generation unit 13 sets the element (4) so that it becomes a sentence in the same language as the language set for the data analysis tool situation or the language of the text already displayed on the chat screen 53.

[0058] In the example of FIG. 3, the prompt generation unit 13 generates a prompt as follows, for example. (1) Instruction: Generate a sentence for the user to answer (2) Context: "lasso" and "ridge" are data analysis terms (3) Input data: "Please tell me the difference between lasso and ridge" (4) Output indicator: A Japanese sentence

[0059] As described herein, when the request is the text of an inquiry input by the user, the information providing apparatus 10 may generate a sentence without considering the level.

[0060] (2. When the request is the operation content of the data analysis tool) An example in which the request is the operation content of the data analysis tool will be described with reference to FIG. 4. FIG. 4 is a diagram showing an example of a screen. Canvas data is shown on the screen of FIG. 4.

[0061] In the example of FIG. 4, a card 514 is arranged. The card 514 represents that the analysis model is a "linear model". The analysis model may be paraphrased as an algorithm for analyzing data.

[0062] Sentences generated by the information providing apparatus 10 are displayed in the boxes 5313 and 5314 of the chat screen 53. In the example of FIG. 4, according to the user's operation content and the selected level, the information providing apparatus 10 generates a sentence and outputs the generated sentence.

[0063] The level selection unit 17 selects a level. For example, it selects the level specified by the user. For example, the level is either an advanced user or a beginner.

[0064] The prompt generation unit 13 receives the operation content for the data analysis tool as a request. For example, the prompt generation unit 13 generates a prompt in response to an operation content (request) such as "specify that the analysis model is a linear model". The elements included in the prompt are as described above.

[0065] When the request is the operation content of the user, the prompt generation unit 13 sets element (1) as "generation of an advice sentence according to the operation content". If the speech on the user side displayed on the chat screen 53 (for example, the text in the box 5311) does not match the request, the prompt generation unit 13 may determine that the request is the operation content of the user.

[0066] Also, the prompt generation unit 13 determines element (2) according to the following procedure. First, the prompt generation unit 13 determines whether a predetermined operation content has been performed before the target operation content. For example, when the operation content of the request is the specification of the analysis model, the prompt generation unit 13 determines whether the preprocessing has already been specified. The prompt generation unit 13 determines this determination result as element (2). Also, element (3) is the operation content itself.

[0067] Also, the prompt generation unit 13 sets element (4) as something like "a sentence in natural language". Also, the prompt generation unit 13 may set element (4) including a language such as "a sentence in English" or "a sentence in Japanese".

[0068] Also, the prompt generation unit 13 adds the level selected by the level selection unit 17 to element (4). Here, it is assumed that the level selection unit 17 has selected "beginner" as the level.

[0069] In the example of FIG. 4, the prompt generation unit 13 generates a prompt as follows, for example. (1) Instruction: Generate an advice sentence according to the operation content (2) Context: Pretreatment not specified (3) Input data: Specify "linear model" for the analysis model (4) Output indicator: Japanese sentence, level: beginner

[0070] In this way, in the data analysis tool where the prompt generation unit 13 generates a series of processes related to data analysis, the prompt generation unit 13 receives, as a request, the operation content of specifying an algorithm for analyzing data, and generates a prompt instructing to generate an advice sentence according to the operation content. Further, the sentence generation unit 16 inputs the prompt into the language model, and generates a sentence indicating an operation to be performed after the operation content (for example, specification of pretreatment), and generates a sentence according to the level. Also, the style of the sentence changes according to the selected role.

[0071] When there is one language model included in the language model group 15, the sentence generation unit 16 can input the level specified in element (4) (for example, "level: beginner") as it is into the language model as part of the prompt.

[0072] On the other hand, as shown in FIG. 1, when the language model group 15 includes an advanced user language model 151 and a beginner language model 152, the language model into which the prompt is input may be determined according to the level specified in element (4).

[0073] Here, the difference between the sentence generated by the language model when the level is "advanced user" (hereinafter, the advanced user-oriented sentence) and the sentence generated by the language model when the level is "beginner" (hereinafter, the beginner-oriented sentence) will be explained. The difference explained here is caused by the inference process by the language model.

[0074] For example, the text for advanced users contains less information than the text for beginners. For example, assume that the text shown in box 5313 of FIG. 4 is for beginners. When a prompt in which the "Level: beginner" in element (4) of the prompt in the case of FIG. 4 is replaced with "Level: advanced" is input to the language model, a text in which the part "And then, the next thing to do is the preprocessing." is omitted is generated from the text shown in box 5313.

[0075] The text for advanced users can be said to be a text with a higher level of difficulty than the text for beginners. That is, the level of difficulty of the text is determined by the level.

[0076] Also, for example, the text for beginners tends to move the user's emotions more strongly than the text for advanced users.

[0077] For example, the prompt generation unit 13 generates a prompt for causing the language model to generate a text that moves the user's emotions more strongly as the level selected by the level selection unit 17 is lower. For example, the prompt generation unit 13 adds the phrase "emotionally" to element (4) of the prompt together with "Level: beginner".

[0078] For example, the language model generates a text that moves the user's emotions strongly by using interjections, expressing the operations on the user's operation content in an exaggerated manner, etc. For example, when praising an operation that specifies a certain preprocessing, a text with a plain impression such as "I think that preprocessing is good" is added in the text for advanced users, and a text with an exaggerated impression such as "I think that preprocessing is very good!" is added in the text for beginners.

[0079] The level selection unit 17 may automatically select a level using at least one of the request and result obtained from the request processing unit 12 and the past requests and results obtained from the history DB 14.

[0080] The operation content is, for example, a specified analysis model (linear model, Lasso, Ridge, etc.), preprocessing, set parameters, etc. Also, the results are, for example, whether the process execution was successful, accuracy indicators (e.g., RMSE (Root Mean Squared Error) of a linear model), etc. Note that the operation content and results are expressed as canvas data.

[0081] For example, if the accuracy indicator shown in the result obtained from the request processing unit 12 is greater than or equal to the threshold value, the advanced level is selected by the level selection unit 17, and if the accuracy indicator is less than the threshold value, the beginner level is selected.

[0082] Also, the level selection unit 17 may calculate the success rate at the time of process execution, the average value of the accuracy indicator, etc. from the canvas data of a predetermined number (e.g., 10) of past users obtained from the history DB 14, and select the level based on the calculation result.

[0083] Also, the prompt generation unit 13 may generate a prompt by referring to past requests and results obtained from the history DB 14. For example, the prompt generation unit 13 selects the canvas data with the highest evaluation value from the canvas data that shows the evaluation value of the accuracy of the finally trained analysis model, which is the canvas data when the "linear model" was specified for the analysis model in the past. Then, the prompt generation unit 13 generates a prompt that prompts the generation of a sentence that recommends the preprocessing specified in the selected canvas data.

[0084] Furthermore, the data content to be analyzed may be stored in the history DB 14. For example, the request processing unit 12 stores the data to be analyzed in the history DB 14 together with the request and the result. In the history DB 14, data in a predetermined format such as CSV format or table format may be stored as it is, or a part of the data may be stored, or statistical quantities calculated from the data may be stored.

[0085] The prompt generation unit 13 can generate a prompt by referring not only to the past requests and results obtained from the history DB 14 but also to the past data to be analyzed obtained from the history DB 14. Thereby, for example, the prompt generation unit 13 can use the data to be analyzed as a material for determining preprocessing with a high recommendation degree. For example, when the characteristics of the data to be analyzed acquired by the request processing unit 12 are similar to the characteristics of the data to be analyzed that has been "outlier interpolation" as preprocessing in the past, the prompt generation unit 13 generates a prompt so that a sentence recommending "outlier interpolation" as preprocessing is likely to be generated. Note that the characteristics of the data may be statistical quantities such as variance and average.

[0086] For example, as follows, the prompt generation unit 13 describes in the element (2) of the prompt that the recommended degree of the preprocessing specified in the selected canvas data is high. (2) Context: Preprocessing not specified (however, "outlier interpolation" is recommended from the history and the data content to be analyzed)

[0087] In this case, for example, a sentence with the sentence "The recommended preprocessing is outlier interpolation." added to the sentence shown in the box 5313 is generated by the language model group 15.

[0088] (Modification example) FIG. 5 is a diagram showing a modification example of the information providing system according to the first embodiment. The information providing apparatus 10a of the information providing system 1a shown in FIG. 5 is different from the information providing apparatus 10 in that it has a level selection model 18.

[0089] The level selection unit 17 selects a level using the level selection model 18. The level selection model 18 is a learned machine learning model that estimates and outputs a level as a classification or regression problem based on the input operation content and result.

[0090] [Flow of processing in the first embodiment] Using FIG. 6, the processing flow of the information providing apparatus 10 will be described. FIG. 6 is a flowchart showing the processing flow of the information providing system according to the first embodiment.

[0091] First, the request processing unit 12 receives the operation content or the inquiry content as a request via the interface unit 11 (step S101). Next, the request processing unit 12 stores the request content and the result in the history DB 14 (step S102).

[0092] Here, the prompt generation unit 13 acquires the past request content and result from the history DB (step S103).

[0093] The level selection unit 17 selects a level based on the request content and the result (step S104). The level selection unit 17 may select a level based on the user's input.

[0094] Subsequently, the prompt generation unit 13 generates a prompt based on the request content and the result according to the selected level (step S105).

[0095] Then, the text generation unit 16 inputs the prompt into the language model (large language model) included in the language model group 15 and generates text (step S106). The text generation unit 16 outputs the generated text via the interface unit 11 (step S107).

[0096] [Effects of the First Embodiment] As described so far, the level selection unit 17 selects the level regarding the user's data analysis. The prompt generation unit 13 generates a prompt for generating a sentence according to the level, which is a sentence for assisting the use of the data analysis tool based on the request received from the terminal device on which the data analysis tool is executed and the level selected by the level selection unit 17, and inputs the prompt into the language model. The sentence generation unit 16 generates a sentence by inputting the prompt generated by the prompt generation unit 13 into the language model. The interface unit 11 causes the terminal device to display the sentence generated by the sentence generation unit 16.

[0097] As a result, the user can obtain knowledge regarding the use of the data analysis tool from the sentence according to his / her own level regarding data analysis. As a result, the usability of the data analysis tool is improved.

[0098] The sentence generation unit 16 generates a sentence by inputting the prompt into the language model corresponding to the level selected by the level selection unit 17 among the language models corresponding to each of the plurality of levels.

[0099] The level selection unit 17 further selects a role. The prompt generation unit 13 generates a prompt for generating a sentence in a style according to the role in the language model. Thereby, the style of the provided sentence can be changed according to the user's preference.

[0100] The prompt generation unit 13 generates a prompt for generating a sentence that strongly moves the user's emotion in the language model as the level selected by the level selection unit 17 is lower. Thereby, the motivation of beginner users can be enhanced.

[0101] The level selection unit 17 selects a level based on the past operation content of the user in the data analysis tool and the accuracy of data analysis by the process generated according to the past operation content. Thereby, a level more suitable for the user can be selected.

[0102] [System Configuration, etc.] Moreover, each component of each illustrated device is functionally conceptual and does not necessarily have to be physically configured as illustrated. That is, the specific forms of distribution and integration of each device are not limited to those illustrated, and all or part of them can be functionally or physically distributed or integrated in any unit according to various loads, usage situations, etc. Furthermore, each processing function performed by each device can be realized in whole or in any part by a CPU and a program analyzed and executed by the CPU, or can be realized as hardware by wired logic. Note that the program may be executed not only by a CPU but also by other processors such as a GPU.

[0103] Also, among the processes described in the embodiments, all or part of the processes described as being automatically performed can be manually performed, or all or part of the processes described as being manually performed can be automatically performed by a known method. In addition, regarding the processing procedures, control procedures, specific names, and information including various data and parameters shown in the above documents and drawings, they can be arbitrarily changed unless otherwise specified.

[0104] [Program] As one embodiment, the information providing apparatus 10 can be implemented by installing an information providing program for executing the above information processing as package software or online software in a desired computer. For example, by causing the information processing apparatus to execute the above information providing program, the information processing apparatus can function as the information providing apparatus 10. The information processing apparatus mentioned here includes desktop or notebook personal computers. In addition, other information processing apparatuses include mobile communication terminals such as tablet terminals, smartphones, mobile phones, and PHS (Personal Handy-phone System), and furthermore, slate terminals such as PDAs (Personal Digital Assistants) are included in this category.

[0105] In addition, the information providing apparatus 10 can also be implemented as a server that uses the terminal device 20 used by the user as a client and provides the above-described services related to the information processing to the client. For example, the server is implemented as a server device that provides an information processing service that takes a request as an input and outputs a sentence. In this case, the server may be implemented as a Web server, or may be implemented as a cloud that provides the above-described services related to the information processing by outsourcing.

[0106] FIG. 7 is a diagram showing a configuration example of a computer that executes an information providing program. The computer 1000 includes, for example, a memory 1010 and a CPU 1020. The computer 1000 also includes a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These components are connected by a bus 1080.

[0107] The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM (Random Access Memory) 1012. The ROM 1011 stores a boot program such as a BIOS (Basic Input Output System), for example. The hard disk drive interface 1030 is connected to the hard disk drive 1090. The disk drive interface 1040 is connected to the disk drive 1100. A removable storage medium such as a magnetic disk or an optical disk is inserted into the disk drive 1100, for example. The serial port interface 1050 is connected to, for example, a mouse 1110 and a keyboard 1120. The video adapter 1060 is connected to, for example, a display 1130.

[0108] The hard disk drive 1090 stores, for example, an OS (Operating System) 1091, application programs 1092, program modules 1093, and program data 1094. That is, the programs defining each process of the information providing apparatus 10 are implemented as program modules 1093 in which computer-executable code is described. The program modules 1093 are stored, for example, in the hard disk drive 1090. For example, program modules 1093 for executing processes similar to the functional configuration in the information providing apparatus 10 are stored in the hard disk drive 1090. Note that the hard disk drive 1090 may be replaced by an SSD (Solid State Drive).

[0109] Also, the setting data used in the processes of the above-described embodiments is stored as program data 1094, for example, in the memory 1010 or the hard disk drive 1090. Then, the CPU 1020 reads out the program modules 1093 and the program data 1094 stored in the memory 1010 or the hard disk drive 1090 into the RAM 1012 as necessary, and executes the processes of the above-described embodiments.

[0110] Note that the program modules 1093 and the program data 1094 are not limited to being stored in the hard disk drive 1090, and may be stored, for example, in a removable storage medium and read by the CPU 1020 via a disk drive 1100 or the like. Alternatively, the program modules 1093 and the program data 1094 may be stored in another computer connected via a network (such as a LAN (Local Area Network) or a WAN (Wide Area Network)). Then, the program modules 1093 and the program data 1094 may be read by the CPU 1020 from the other computer via the network interface 1070.

Explanation of Signs

[0111] 1 Information Provision System 10 Information Provision Device 11 Interface Unit 12 Request Processing Unit 13 Prompt Generation Unit 14 History DB 15 Language Model Group 16 Sentence Generation Unit 53 Chat Screen 511, 512, 513, 514 Cards 521, 522 Areas 532 Icon 533 Input Field 5311, 5312, 5313, 5314 Boxes

Claims

1. A level selection unit that selects a level related to user data analysis, A request received from a terminal device on which a data analysis tool that generates a series of processes related to data analysis is executed, the request being an operation content that designates an algorithm for analyzing data, and based on the level selected by the level selection unit, a sentence indicating an operation to be performed after the operation content, and a prompt generation unit that generates a prompt for causing a language model to generate a sentence corresponding to the level, A sentence generation unit that generates a sentence by inputting the prompt generated by the prompt generation unit into the language model, A display control unit that causes the terminal device to display the sentence generated by the sentence generation unit, having The level selection unit selects a level based on the operation content previously specified by the user in the data analysis tool, the success or failure of data analysis processing and the accuracy index according to the process generated according to the operation content. An information providing device characterized by the above.

2. The sentence generation unit generates a sentence by inputting the prompt into the language model corresponding to the level selected by the level selection unit among the language models corresponding to each of the plurality of levels. The information providing device according to claim 1, characterized in that.

3. The level selection unit further selects a role, The prompt generation unit generates a prompt for causing the language model to generate a sentence in a style corresponding to the role. The information providing device according to claim 1, characterized in that.

4. The prompt generation unit generates a prompt for causing the language model to generate a sentence that more strongly moves the user's emotions as the level selected by the level selection unit is lower. The information providing device according to claim 1, characterized in that.

5. An information providing method executed by a computer, comprising: a level selection step of selecting a level related to user data analysis; a request received from a terminal device on which a data analysis tool that generates a series of processes related to data analysis is executed, the request being an operation content for specifying an algorithm for analyzing data, and based on the level selected in the level selection step, a sentence indicating an operation to be performed after the operation content, and a prompt generation step of generating a prompt for causing a language model to generate a sentence corresponding to the level; a sentence generation step of generating a sentence by inputting the prompt generated in the prompt generation step into the language model; a display control step of causing the terminal device to display the sentence generated in the sentence generation step; and the level selection step selects a level based on the operation content previously specified by the user in the data analysis tool, the success or failure of data analysis processing and the accuracy index according to the process generated according to the operation content. An information providing method characterized by the above.

6. a level selection step of selecting a level related to user data analysis; a request received from a terminal device on which a data analysis tool that generates a series of processes related to data analysis is executed, the request being an operation content for specifying an algorithm for analyzing data, and based on the level selected in the level selection step, a sentence indicating an operation to be performed after the operation content, and a prompt generation step of generating a prompt for causing a language model to generate a sentence corresponding to the level; a sentence generation step of generating a sentence by inputting the prompt generated in the prompt generation step into the language model; A display control step of causing the terminal device to display the text generated by the text generation step, causing a computer to execute, The level selection step selects a level based on the operation content specified by the user in the data analysis tool in the past, the success or failure of execution of data analysis processing by a process generated according to the operation content, and an accuracy index. An information providing program characterized by the above.

Images