Story outline relevance evaluation method and device, electronic equipment, storage medium and program product

Abstract

This disclosure relates to a method and apparatus, electronic device, storage medium, and program product for evaluating the relevance of story synopses. The method includes: acquiring a story synopsis and basic information; displaying a first evaluation result, which is obtained by evaluating the story synopsis using at least one first intelligent agent based on at least one evaluation dimension and basic information, according to a preset evaluation process; each evaluation dimension includes multiple sub-question types and a question example for each sub-question type, the question example including a synopsis example belonging to the sub-question type and the reason why the synopsis example belongs to that sub-question type; and displaying a target evaluation result for the story synopsis determined based on the first evaluation result, the target evaluation result representing whether the relevance of the story synopsis meets the standard. Therefore, artificial intelligence technology can be used to achieve automated evaluation of the relevance of story synopses based on at least one evaluation dimension, improving the efficiency and quality of the evaluation.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence, and more particularly to a method and apparatus for evaluating the relevance of a story outline, electronic devices, storage media, and program products. Background Technology

[0002] With the rapid development of the film, television, gaming, and other cultural and entertainment industries, the demand for story creation is increasing, resulting in a rise in the number of stories but varying quality. In the traditional story creation process, a story outline must first be created, followed by the development of a more detailed story text based on that outline. The relevance of the story outline to the theme has a significant impact on the quality of the story text. Therefore, it is necessary to evaluate the relevance of the story outline to generate high-quality story texts from outlines that pass the relevance assessment. Currently, however, the relevance assessment of story outlines mainly relies on manual work. Manual evaluation of story outlines is inefficient, highly subjective, and struggles to guarantee consistency and accuracy, resulting in lower overall evaluation quality. Summary of the Invention

[0003] In view of this, this disclosure proposes a method, apparatus, electronic device, storage medium, and program product for evaluating the relevance of a story synopsis, which can utilize artificial intelligence technology to achieve automated evaluation of the relevance of a story synopsis based on at least one evaluation dimension, thereby improving the evaluation efficiency and quality of the relevance of the story synopsis.

[0004] According to one aspect of this disclosure, a method for evaluating the relevance of a story synopsis is provided, comprising: obtaining a story synopsis to be evaluated and basic information corresponding to the story synopsis; displaying a first evaluation result, the first evaluation result including: the sub-problem type to which the story synopsis belongs under any evaluation dimension and the reason for belonging to the sub-problem type; the first evaluation result is obtained by using at least one first agent to evaluate the story synopsis based on at least one evaluation dimension and the basic information, according to a preset evaluation process, with different first agents corresponding to different evaluation dimensions; the evaluation dimension includes at least one of the following: whether the story synopsis conforms to the historical context, whether the story synopsis conforms to the story type, whether the story synopsis conforms to common sense, and the way the story synopsis achieves the writing purpose; each evaluation dimension includes multiple sub-problem types and a problem example for each sub-problem type, the problem example including a synopsis example belonging to the sub-problem type and the reason why the synopsis example belongs to the sub-problem type; displaying a target evaluation result of the story synopsis determined based on the first evaluation result of the story synopsis, the target evaluation result representing whether the relevance of the story synopsis meets the standard.

[0005] In one possible implementation, the preset evaluation process includes: for any evaluation dimension, analyzing the story outline based on the analysis content corresponding to that evaluation dimension to obtain the analysis result corresponding to that evaluation dimension; determining the sub-problem type corresponding to the story outline under that evaluation dimension based on the analysis result; and determining the reason for belonging to that sub-problem type based on the problem example corresponding to that sub-problem type.

[0006] In one possible implementation, the evaluation dimension includes a first evaluation dimension, which includes: the way the story outline achieves the writing purpose; the multiple sub-problem types corresponding to the first evaluation dimension include: forcibly achieving the writing purpose and normally achieving the writing purpose; the forcibly achieving the writing purpose represents the story outline achieving the writing purpose through illogical plot development and / or character behavior; the normally achieving the writing purpose represents the story outline achieving the writing purpose through logically consistent plot development and / or character behavior; wherein, the step of using at least one first agent to evaluate the story outline based on the at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result of the story outline includes: using the first agent corresponding to the first evaluation dimension to analyze whether the plot development logic in the story outline that achieves the writing purpose can achieve the writing purpose, whether the plot development lacks key links or details, whether the character behavior lacks rationality, and whether the plot development lacks reasonable foreshadowing, to obtain a first analysis result; based on the first analysis result, determining the sub-problem type to which the story outline belongs under the first evaluation dimension; and based on the problem examples of the sub-problem type, determining the reason for belonging to that sub-problem type.

[0007] In one possible implementation, the evaluation dimension includes a second evaluation dimension, which includes whether the story outline conforms to common sense. The second evaluation dimension corresponds to multiple sub-question types, including: the story outline does not conform to common sense and it does. The step of evaluating the story outline using at least one first agent based on the at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result for the story outline, includes: using the first agent corresponding to the second evaluation dimension to analyze whether there is content in the story outline that does not conform to common sense, obtaining a second analysis result; based on the second analysis result, determining the sub-question type to which the story outline belongs under the second evaluation dimension; and based on the question examples of the sub-question type, determining the reason for belonging to that sub-question type.

[0008] In one possible implementation, the evaluation dimension includes a third evaluation dimension, which includes whether the story synopsis conforms to the historical context. The third evaluation dimension corresponds to multiple sub-question types, including: the story synopsis is completely inconsistent with the historical context, partially inconsistent with the historical context, and consistent with the historical context. The step of evaluating the story synopsis using at least one first agent based on the at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result for the story synopsis, includes: using the first agent corresponding to the third evaluation dimension to analyze whether there are elements in the story synopsis that do not conform to the historical context, obtaining a third analysis result, where the elements include elements in the story synopsis other than characters; based on the third analysis result, determining the sub-question type to which the story synopsis belongs under the third evaluation dimension; and based on the question examples of the sub-question type, determining the reason for belonging to that sub-question type.

[0009] In one possible implementation, the evaluation dimension includes a fourth evaluation dimension, which includes whether the story outline conforms to the story type; the multiple sub-question types corresponding to the fourth evaluation dimension include: the story outline does not conform to the story type and the story outline does conform to the story type; wherein, the step of using at least one first agent to evaluate the story outline based on the at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result of the story outline includes: using the first agent corresponding to the fourth evaluation dimension to analyze whether the plot in the story outline conforms to the story type, to obtain a fourth analysis result; based on the fourth analysis result, determining the sub-question type to which the story outline belongs under the fourth evaluation dimension; and based on the question examples of the sub-question type, determining the reason for belonging to that sub-question type.

[0010] In one possible implementation, determining the target evaluation result of the story outline based on the first evaluation result of the story outline includes: using a second agent to generate a second evaluation result of the story outline based on the basic information and the first evaluation result of the story outline, wherein the second evaluation result is used to indicate errors present in the first evaluation result; and determining the target evaluation result of the story outline based on the second evaluation result and the first evaluation result.

[0011] According to another aspect of this disclosure, a device for evaluating the relevance of a story synopsis is provided, comprising: an acquisition module for acquiring a story synopsis to be evaluated and basic information corresponding to the story synopsis, the basic information including backstory, story type, historical context, and writing purpose; and a first display module for displaying a first evaluation result, the first evaluation result including: the sub-problem type to which the story synopsis belongs under any evaluation dimension and the reason for belonging to the sub-problem type; the first evaluation result is obtained by evaluating the story synopsis using at least one first agent based on the at least one evaluation dimension and the basic information, according to a preset evaluation process, and different first agents can evaluate the story synopsis. One intelligent agent corresponds to different evaluation dimensions; the evaluation dimensions include at least one of the following: whether the story outline conforms to the said historical context, whether the story outline conforms to the said story type, whether the story outline conforms to human common sense, and the way the story outline achieves the said writing purpose; each evaluation dimension includes multiple sub-problem types and problem examples for each sub-problem type, the problem examples including a synopsis example belonging to the sub-problem type and the reason why the synopsis example belongs to the sub-problem type; the second display module is used to display the target evaluation result of the story outline determined according to the first evaluation result of the story outline, the target evaluation result characterizing whether the relevance of the story outline to the topic meets the standard.

[0012] According to another aspect of this disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-described method.

[0013] According to another aspect of this disclosure, a non-volatile computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the above-described method.

[0014] According to another aspect of this disclosure, a computer program product is provided, including a computer program or a non-volatile computer-readable storage medium carrying the computer program, wherein the computer program, when executed by a processor, implements the steps of the above-described method.

[0015] According to various aspects of this disclosure, by demonstrating the first evaluation result obtained by using at least one first intelligent agent to evaluate a story outline based on at least one evaluation dimension and basic information, and following a preset evaluation process, and then demonstrating the target evaluation result based on the first evaluation result of the story outline to determine whether the story outline's relevance to the topic meets the standard, it is possible to achieve automated evaluation of the relevance of a story outline using artificial intelligence technology based on at least one evaluation dimension and basic information, as well as a preset evaluation process. The evaluation dimensions used for relevance evaluation, such as how the story outline achieves its writing purpose, whether the story outline conforms to the historical context, whether the story outline conforms to the story type, and whether the story outline conforms to common sense, are all to examine whether the story outline is consistent with the constraints indicated by the preset basic information and common sense. This, combined with artificial intelligence technology, enables the rapid and objective discovery of potential relevance issues in the story outline, which can greatly improve the evaluation efficiency and quality of the story outline, reduce the burden of manual evaluation, and thus contribute to improving the quality of the story outline.

[0016] Other features and aspects of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0017] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this disclosure together with the specification and serve to explain the principles of this disclosure.

[0018] Figure 1 A flowchart illustrating a story outline relevance assessment method according to an embodiment of the present disclosure is shown.

[0019] Figure 2 A schematic diagram of a story outline relevance assessment framework according to an embodiment of the present disclosure is shown.

[0020] Figure 3 A block diagram of a story outline relevance assessment device according to an embodiment of the present disclosure is shown.

[0021] Figure 4 A block diagram of an electronic device 1900 according to an embodiment of the present disclosure is shown. Detailed Implementation

[0022] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0023] As used herein, the terms “comprising,” “including,” “having,” or variations thereof are open-ended and include one or more of the stated features, integrals, elements, steps, components, or functions, but do not exclude the presence or addition of one or more other features, integrals, elements, steps, components, functions, or groups thereof.

[0024] When an element is referred to as “connected,” “coupled,” “responding,” or a variation thereof relative to another element, it may be directly connected, coupled, or responding to another element, or there may be an intermediate element present.

[0025] Although the terms first, second, third, etc., may be used herein to describe various elements / operations, these elements / operations should not be limited by these terms. These terms are only used to distinguish one element / operation from another. Therefore, without departing from the teachings of the inventive concept, a first element / operation in some embodiments may be referred to as a second element / operation in other embodiments.

[0026] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0027] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0028] As mentioned above, with the development of the film and entertainment industry, the number of stories has increased, but their quality varies. Traditional manual evaluation of story synopsis relevance is inefficient and highly subjective. Automated evaluation of story synopsis relevance is therefore essential. For story production companies, it allows for rapid screening of synopsis, reducing costs and increasing objectivity; for creators, it provides timely feedback to help them improve their synopsis; and from an industry trend perspective, it aligns with the direction of digital and intelligent development.

[0029] In view of this, the present disclosure proposes a method for evaluating the relevance of a story synopsis, which can use artificial intelligence technology to automatically evaluate the relevance of a story synopsis based on at least one evaluation dimension. This enables the rapid and objective discovery of potential relevance issues in the story synopsis, which can greatly improve evaluation efficiency and quality, reduce the burden of manual evaluation, and improve the quality of the story synopsis.

[0030] The story outline relevance assessment method of this disclosure can be deployed on various terminal devices through software or hardware modifications. The terminal devices involved in this disclosure can refer to devices with wireless and / or wired connection functions. Wireless connection means that they can connect to other devices via Wi-Fi, Bluetooth, or other wireless methods. The terminal devices involved in this disclosure can also communicate with other devices via wired connections. The terminal devices involved in this disclosure can be touchscreen, non-touchscreen, or screenless. Touchscreen devices can be controlled by clicking or swiping on the display screen using fingers or styluses. Non-touchscreen devices can connect to input devices such as mice, keyboards, or touch panels to control the terminal device. Screenless devices can be, for example, screenless Bluetooth speakers. For example, the terminal devices in this application can include, but are not limited to, user equipment (UE), mobile devices, user terminals, terminals, handheld devices, tablet computers, laptops, PDAs, and computing devices.

[0031] The story synopsis relevance evaluation method of this disclosure can also be deployed on a server. This server can be located in the cloud or locally, and can be a physical device or a virtual device, such as a virtual machine or container, with wireless communication capabilities. These wireless communication capabilities can be configured in the server's chip (system) or other components. It can refer to a device with wireless connectivity, meaning it can connect to other servers or terminal devices via wireless connections such as Wi-Fi or Bluetooth. The server involved in this disclosure can also have wired communication capabilities. For example, the server in this disclosure can be located in the cloud, communicating with terminal devices, receiving story synopses and corresponding basic information sent by the terminal devices, and using the story synopsis relevance evaluation method deployed on the server to generate a target evaluation result for the story synopsis based on the story synopsis and basic information. This result is then returned to the terminal device for display to the user.

[0032] Figure 1 A flowchart illustrating a method for evaluating the relevance of a story outline according to an embodiment of this disclosure is shown. Figure 1 As shown, the method includes steps S11 to S13.

[0033] In step S11, the story outline to be evaluated and the basic information corresponding to the story outline are obtained.

[0034] A story synopsis can be understood as a concise summary of the main characters, core content, and story direction of a story. It helps to outline the main plot, clarify the creative direction, and introduce the general content of the story.

[0035] The basic information in a story synopsis includes the backstory, story type, historical setting, and writing purpose. The backstory can be understood as the information preceding the main plot points, or the preceding context. Story type can include, for example, urban, period, workplace, espionage, or political intrigue genres. Historical setting can include the year and location of the story, such as England in 1919. The writing purpose indicates the direction of the story in the synopsis; for example, it can include one or more of the following: characters to be included in the synopsis, relationships between characters, characters' emotions, key plot points, the ending, and the nature of the story (e.g., happy or sad).

[0036] In step S12, a first evaluation result is displayed. The first evaluation result includes: the sub-problem type to which the story outline belongs under any evaluation dimension and the reason for belonging to the sub-problem type. The first evaluation result is obtained by evaluating the story outline using at least one first agent based on at least one evaluation dimension and basic information, according to a preset evaluation process. Different first agents correspond to different evaluation dimensions.

[0037] The evaluation dimensions include at least one of the following: whether the story synopsis conforms to the historical context, whether the story synopsis conforms to the story genre, whether the story synopsis conforms to common sense, and how the story synopsis achieves its writing purpose. Assessing whether the story synopsis conforms to the historical context is equivalent to assessing whether there is any content in the story synopsis that is inconsistent with the historical context. For example, if the story synopsis is set in the steam age, but mentions smartphones and the internet, then the story synopsis is completely inconsistent with the historical context. Another example is if the story synopsis is set in the 1950s, but describes an ordinary family owning a private plane. While airplanes existed in the 1950s, it is still unreasonable for an ordinary family to own a private plane; in this case, the story synopsis is partially inconsistent with the historical context. Assessing whether the story synopsis conforms to the story genre is equivalent to judging whether the plot of the story synopsis conforms to the story genre. For example, if the story genre is thriller, but the plot described in the story synopsis has a cheerful and lighthearted atmosphere without any thriller elements, then the story synopsis is not consistent with the story genre.

[0038] Among them, assessing how the story outline achieves its writing purpose is equivalent to assessing whether the story outline forcibly achieves its writing purpose, or in other words, assessing whether there are obvious logical breaks in the plot development, resulting in the lack of rationality or feasibility in achieving the writing purpose, and requiring reliance on illogical plot development or character behavior to achieve it; among them, forcibly achieving the writing purpose can include at least one of the following: the plot development logic of the story outline cannot achieve the writing purpose, the plot development lacks key links or details, the character behavior lacks rationality, and the plot development lacks reasonable foreshadowing. In this context, "the plot development logic in the story synopsis fails to achieve the writing purpose" can be understood as the plot in the story synopsis logically failing to achieve the writing purpose. It is as if the author forcibly made the story achieve the writing purpose, or even if the plot development proceeds smoothly, it is not logically possible to achieve the writing purpose. It is just that the author ultimately forced the story to achieve the writing purpose. For example, the writing purpose is to design a plot so that A stops investigating B. However, the story synopsis describes "B using a photo of A and C eating to threaten A, and A stopped the investigation after seeing the photo." This is logically untenable, and the author forcibly made the plot achieve the writing purpose. Assessing whether the plot development lacks key elements or details can be understood as evaluating whether the story synopsis lacks crucial logical elements or details, making the plot development unconvincing and the achievement of the writing purpose seem abrupt and far-fetched. For example, if the writing purpose is to design a plot to help Xiao A escape from capture, but the story synopsis describes "When Xiao A was surrounded, a group of monkeys suddenly appeared, causing chaos, allowing Xiao A to escape in the confusion," this plot lacks necessary supporting elements or details, making the achievement of the writing purpose seem very far-fetched. Similarly, the lack of comprehension in character behavior can be understood as the character being given a cheat code, that is, giving the character an extra advantage, granting them extraordinary abilities, resources, or background. For example, if the writing purpose is to design a plot to help Xiao A escape from capture, but the story synopsis describes "During Xiao A's escape, he accidentally discovered an abandoned laboratory with a device that allows for instantaneous movement," this is flawed."Little A activates the device and instantly teleports to another location, escaping pursuit." This is essentially giving Little A a cheat code, making her escape seem overly simple and illogical—in other words, the character's actions lack rationality. The lack of reasonable foreshadowing in plot development can be understood as a deus ex machina, where an unexpected character, event, ability, or object suddenly appears to resolve the conflict or advance the plot when the story is stuck and difficult to develop, without any prior setup. For example, the writing objective might be to design a plot where Little A and Little B escape from the abyss; however, the synopsis describes, "Just when Little A and Little B felt hopeless about escaping the abyss, a deity descended from the sky and rescued them." Since the deity's sudden appearance is completely unforeshadowed, it's an external force forcibly introduced to solve the protagonists' predicament—this is an example of plot development lacking reasonable foreshadowing and forcibly achieving the writing objective.

[0039] Assessing whether a story synopsis conforms to common sense is equivalent to assessing whether it contains content that contradicts common sense. For example, if a story synopsis describes "a character being able to speak after death," it is considered to contain content that contradicts common sense. Assessing whether a story synopsis conforms to common sense is equivalent to assessing whether the story synopsis conforms to the constraint of common sense. In other words, common sense can also serve as constraint information for assessing whether the story plot is relevant to the theme, thus evaluating the synopsis's relevance to the theme.

[0040] In practical applications, each evaluation dimension can include multiple sub-problem types and problem examples for each sub-problem type. Each problem example can include a summary example belonging to the sub-problem type and the reason why the summary example belongs to that sub-problem type. The problem example corresponding to each sub-problem type can also include basic information corresponding to the summary example. By providing multiple sub-problem types under each evaluation dimension and problem examples for each sub-problem type, the first agent can better understand the content evaluated by different evaluation dimensions. That is, it can analyze and understand the evaluation criteria of the evaluation dimension and why the summary example is divided into its respective sub-problem type, so as to enable the first agent to give more accurate evaluation results according to the evaluation dimension.

[0041] Specifically, the evaluation dimensions can include: a first evaluation dimension, a second evaluation dimension, a third evaluation dimension, and a fourth evaluation dimension. The first evaluation dimension includes: how the story synopsis achieves its writing purpose; the second evaluation dimension includes: whether the story synopsis conforms to common sense; the third evaluation dimension includes: whether the story synopsis conforms to the historical context; and the fourth evaluation dimension includes: whether the story synopsis conforms to the story genre. The first evaluation dimension corresponds to several sub-question types, including: forcibly achieving the writing purpose and achieving the writing purpose normally; forcibly achieving the writing purpose indicates that the story synopsis achieves its purpose through illogical plot development and / or character behavior; achieving the writing purpose normally indicates that the story synopsis achieves its purpose through logically consistent plot development and / or character behavior. The second evaluation dimension corresponds to several sub-question types, including: the story synopsis does not conform to common sense and conforms to common sense. The third evaluation dimension corresponds to several sub-question types, including: the story synopsis is completely inconsistent with the historical context, partially inconsistent with the historical context, and consistent with the historical context. The fourth evaluation dimension corresponds to several sub-question types, including: the story synopsis does not conform to the story genre and conforms to the story genre.

[0042] The synopsis examples corresponding to sub-problem types can be understood as exemplary story synopses of that sub-problem type. The reason why the synopsis example belongs to that sub-problem type can be understood as the reason why the exemplary story synopsis belongs to that sub-problem type. For example, if the synopsis example describes "Just when A and B felt hopeless about escaping the abyss, a deity descended from the sky and rescued A and B from the abyss," the corresponding sub-problem type is: forcibly achieving the writing purpose, specifically, the plot development lacks reasonable foreshadowing. The reason is: the sudden appearance of the deity is completely unforeshadowed; it is an external force forcibly introduced to solve the protagonist's predicament, thus forcibly achieving the writing purpose without reasonable foreshadowing. It should be understood that users can customize the sub-problem types included in each evaluation dimension according to their actual needs, as well as the synopsis examples corresponding to each sub-problem type, the reasons why the synopsis example belongs to that sub-problem type, and the basic information corresponding to the synopsis example.

[0043] In practical applications, each evaluation dimension may also include a score set for each sub-problem type. For example, in the first evaluation dimension, the score for a story outline that forcibly achieves the writing purpose may be set to "-1 point," and the score for a story outline that normally achieves the writing purpose may be set to "0 points." In the second evaluation dimension, the score for a story outline that does not conform to common sense may be set to "-1 point," and the score for one that conforms to common sense may be set to "1 point." In the third evaluation dimension, the score for a story outline that is completely inconsistent with the historical context may be set to "-1 point," the score for one that is partially inconsistent with the historical context may be set to "0 points," and the score for one that conforms to the historical context may be set to "1 point." In the fourth evaluation dimension, the score for a story outline that does not conform to the story type may be set to "-1 point," and the score for one that conforms to the story type may be set to "1 point." This embodiment of the present disclosure does not impose any limitations on this.

[0044] It should be understood that the sub-problem types under the above evaluation dimensions are some possible implementation methods provided by the embodiments of this disclosure. In fact, those skilled in the art can also customize the types of sub-problem types and corresponding problem examples under each evaluation dimension according to actual needs. The story outline conforming to the story type in the fourth evaluation dimension can be further subdivided into: the story outline partially conforming to the story type and completely conforming to the story type. This disclosure does not limit this.

[0045] In recent years, Natural Language Processing (NLP) technology has made significant progress, especially with the development of large-scale NLP models based on deep learning (such as GPT and BERT models), which have demonstrated powerful capabilities in natural language generation and understanding. These models, trained on large amounts of text data, are able to understand and generate text. Therefore, in one possible implementation, a first agent can be constructed and trained using existing large-scale NLP models to evaluate story summaries based on various evaluation dimensions.

[0046] It should be understood that the first agent corresponding to each evaluation dimension can adopt a large language model of a type known in the art (such as a general large language model like qwen 2.0). Those skilled in the art can use model training methods known in the art to fine-tune the pre-trained general large language model using a dataset to obtain the first agent corresponding to each evaluation dimension. This disclosure does not limit this. For example, the training process for the first agent corresponding to the first evaluation dimension may include: constructing a dataset including multiple sample data (i.e., multiple sample story outlines and corresponding sample basic information) and sample labels for each sample data (i.e., sample evaluation results for each sample story outline under the first evaluation dimension). The sample evaluation results may include, for example, the sub-problem type to which the sample story outline belongs, the corresponding reasons, and the score; then, the sample data and the text content of the first evaluation dimension, as well as related instructions, can be input into a pre-trained general-purpose large language model based on the dataset. The related instructions may indicate a preset evaluation process. For example, the related instructions may include "Please evaluate the input sample story outline based on the input evaluation dimension and sample basic information, and according to the given evaluation process, and output the sub-problem type to which the sample story outline belongs and the reasons for belonging to the sub-problem type"); to obtain the predicted evaluation results output by the general-purpose large language model, and calculate the loss based on the difference between the predicted evaluation results and the sample labels of the sample data (i.e., the difference in sub-problem types, the difference in reasons, etc.), and adjust the parameters of the general-purpose large language model based on the loss to obtain the first agent corresponding to the first evaluation dimension.

[0047] It should be understood that the training process of the first agent corresponding to the first evaluation dimension can be referenced to achieve the training of the first agents corresponding to the second, third, and fourth evaluation dimensions, respectively. The difference lies in that, when training the first agent corresponding to the first evaluation dimension, the sample label of the sample data is the sample evaluation result of each sample story outline under the first evaluation dimension; when training the first agent corresponding to the second evaluation dimension, the sample label of the sample data is the sample evaluation result of each sample story outline under the second evaluation dimension; when training the first agent corresponding to the third evaluation dimension, the sample label of the sample data is the sample evaluation result of each sample story outline under the third evaluation dimension; and when training the first agent corresponding to the fourth evaluation dimension, the sample label of the sample data is the sample evaluation result of each sample story outline under the fourth evaluation dimension. The training process of the first agents corresponding to the first, second, third, and fourth evaluation dimensions will not be elaborated further in this embodiment. The first agents of each of the above evaluation dimensions can analyze the text content of their respective input evaluation dimensions and understand the reasons why the outline examples in each sub-question type of the input evaluation dimension are classified into their respective sub-question types, which helps the first agent to give more accurate evaluation results according to the evaluation dimensions.

[0048] In one possible implementation, the preset evaluation process may include: analyzing the story outline based on the analysis content corresponding to any evaluation dimension to obtain the analysis result corresponding to that evaluation dimension; determining the sub-problem type corresponding to the story outline under that evaluation dimension based on the analysis result; and determining the reason belonging to that sub-problem type based on the problem example corresponding to that sub-problem type. In practical applications, the above evaluation process can be instructed in the command (i.e., prompt) input to the first agent to control the first agent to perform the evaluation work according to the above evaluation process. Alternatively, the first agent can be divided into three evaluation sub-modules, each of which can be constructed using a large language model. Then, each evaluation sub-module can be set to execute the steps of the above evaluation process. That is, the first evaluation sub-module can be used to analyze the story outline based on the analysis content corresponding to any evaluation dimension to obtain the analysis result corresponding to that evaluation dimension; the second evaluation sub-module can be used to determine the sub-problem type corresponding to the story outline under that evaluation dimension based on the analysis result; and the third evaluation sub-module can be used to determine the reason belonging to that sub-problem type based on the problem example corresponding to that sub-problem type. This disclosure embodiment does not limit this. The following sections describe the evaluation process for each evaluation dimension.

[0049] Based on the first agent corresponding to the first evaluation dimension that has been trained, the above-mentioned method of using at least one first agent to evaluate the story outline according to at least one evaluation dimension and basic information, based on a preset evaluation process, to obtain the first evaluation result of the story outline, may include:

[0050] Using the first intelligent agent corresponding to the first evaluation dimension, the analysis of the plot development logic in the story outline to achieve the writing purpose is conducted to determine whether the plot development lacks key links or details, whether the characters' behavior lacks rationality, and whether the plot development lacks reasonable foreshadowing, thus obtaining the first analysis result.

[0051] Based on the first analysis results, determine the sub-problem type to which the story outline belongs under the first evaluation dimension;

[0052] Based on the problem examples of the sub-problem type, determine the reasons that belong to that sub-problem type.

[0053] The analysis content corresponding to the first evaluation dimension includes: analyzing whether the plot development logic in the story outline achieves the writing purpose, whether the plot development lacks key links or details, whether the characters' behavior lacks rationality, and whether the plot development lacks reasonable foreshadowing; the evaluation process corresponding to the first evaluation dimension may include: analyzing whether the plot development logic in the story outline achieves the writing purpose, whether the plot development lacks key links or details, whether the characters' behavior lacks rationality, and whether the plot development lacks reasonable foreshadowing, to obtain the first analysis result; based on the first analysis result, determining the sub-problem type to which the story outline belongs under the first evaluation dimension; and based on the problem examples of the sub-problem type, determining the reasons for belonging to that sub-problem type.

[0054] The first evaluation dimension can include at least two sub-evaluation dimensions. These sub-evaluation dimensions can include: whether the plot development logic in the story synopsis achieves the writing objective; whether the plot development lacks key elements or details; whether the characters' behavior lacks rationality; and whether the plot development lacks reasonable foreshadowing. In other words, the analysis of the plot development logic in the story synopsis to achieve the writing objective, whether the plot development lacks key elements or details, whether the characters' behavior lacks rationality, and whether the plot development lacks reasonable foreshadowing, yielding the first analysis result, can be understood as using the first agent corresponding to the first evaluation dimension to analyze each sub-evaluation dimension separately, obtaining the second evaluation result.

[0055] In practical applications, the text content of the first evaluation dimension (i.e., multiple sub-question types and question examples for each sub-question type), the story outline, and corresponding basic information, along with relevant instructions (such as "Based on the input basic information, and following the preset evaluation process and evaluation dimensions, evaluate the input story outline, and output the sub-question type to which the story outline belongs and the reason for belonging to that sub-question type"), can be input into the first agent corresponding to the first evaluation dimension. This first agent can analyze the performance of the input story outline in whether the plot forcibly achieves the writing purpose to determine if the story outline does so. For example, it can analyze whether there are obvious logical loopholes in the plot that cause character behavior or event development to deviate from common sense or basic rules, thus logically failing to achieve the writing purpose; analyze whether character behavior lacks reasonable motivation, resulting in unclear or insufficient motivation for a certain behavior; and analyze the plot development... The analysis method involves examining whether the plot development logic in the story outline achieves the writing purpose, whether the plot development lacks key links or details, whether the characters' behavior lacks rationality, and whether the plot development lacks reasonable foreshadowing. If the first analysis result indicates that the story outline has any of the following problems—that the plot development logic fails to achieve the writing purpose, that the plot development lacks key links or details, that the character behavior lacks rationality, and that the plot development lacks reasonable foreshadowing—then the sub-problem type of the story outline is considered to be the forced achievement of the writing purpose. If the first analysis result indicates that the story outline has no of the above forced achievement problems, then the sub-problem type of the story outline is considered to be the normal achievement of the writing purpose. In this case, the story outline achieving the writing purpose can be understood as the overall logic of the story outline being able to achieve the writing purpose. At this time, there may be a few unnatural or deliberate elements, or some coincidences and plots that lack explanation or missing some details. However, if the overall logic of the story outline can achieve the writing purpose, then the story outline can still be considered to have achieved the writing purpose.

[0056] Specifically, when the first analysis result indicates that the story synopsis exhibits at least one forced achievement problem—such as the plot development logic failing to achieve the writing purpose, the plot development lacking key links or details, the character's behavior lacking rationality, or the plot development lacking reasonable foreshadowing—the reasons for the occurrence of at least one forced achievement problem in the story synopsis can be analyzed based on the problem examples corresponding to the at least one forced achievement problem. It should be understood that each forced achievement problem can have its own corresponding problem examples. Conversely, if the first analysis result indicates that none of the above forced achievement problems occur, the reasons for the story synopsis achieving the writing purpose normally can be analyzed based on the problem examples corresponding to the sub-problem type of the story synopsis achieving the writing purpose normally. Analyzing the reasons for the story synopsis forcibly achieving or normally achieving the writing purpose can be understood as analyzing why the story synopsis is similar to the synopsis examples in the problem examples of the corresponding sub-problem types, thereby obtaining the sub-problem type to which the story synopsis belongs under the first evaluation dimension and the reasons for belonging to that sub-problem type. As mentioned above, the first evaluation dimension can also include scores corresponding to each sub-problem type. Therefore, the first agent corresponding to the first evaluation dimension can output the sub-problem type to which the story outline belongs under the first evaluation dimension, as well as the reason and score for belonging to that sub-problem type. For example, if the story outline has any forced achievement problem, the score can be output as "-1". If the story outline achieves the writing purpose normally, the score can be output as "0".

[0057] Based on the first agent corresponding to the trained second evaluation dimension, the above-mentioned evaluation of the story outline using at least one first agent according to at least one evaluation dimension and basic information, based on a preset evaluation process, to obtain a first evaluation result of the story outline, may include:

[0058] The first agent corresponding to the second evaluation dimension is used to analyze whether there is any content in the story outline that does not conform to human common sense, and the second analysis result is obtained.

[0059] Based on the second analysis results, the sub-problem type to which the story outline belongs under the second evaluation dimension is determined;

[0060] Based on the problem examples of the sub-problem type, determine the reasons that belong to that sub-problem type.

[0061] The analysis content corresponding to the second evaluation dimension includes: analyzing whether there is any content in the story outline that does not conform to common sense; the evaluation process corresponding to the second evaluation dimension includes: analyzing whether there is any content in the story outline that does not conform to common sense, and obtaining the second analysis result; based on the second analysis result, determining the sub-problem type to which the story outline belongs under the second evaluation dimension; and based on the problem examples of the sub-problem type, determining the reasons for belonging to that sub-problem type.

[0062] In practical applications, the text content of the second evaluation dimension (i.e., multiple sub-question types and question examples for each sub-question type), the story outline, the corresponding basic information, and related instructions (such as "Please evaluate the input story outline based on the input basic information and according to the preset evaluation process and evaluation dimensions, and output the sub-question type to which the story outline belongs and the reason for belonging to that sub-question type") can be input to the first intelligent agent corresponding to the second evaluation dimension. This first intelligent agent can analyze whether the character information story outline contains content that does not conform to human common sense (i.e., contrary to human common sense) based on a general human common sense database. If the second analysis result indicates that the character behavior in the story outline contains content that does not conform to human common sense, then the sub-question type to which the story outline belongs under the second evaluation dimension is considered to be... If the story synopsis doesn't conform to common sense, we can analyze the reasons for this inconsistency based on the corresponding problem examples. Conversely, if the second analysis indicates that the characters' behaviors in the story synopsis don't contain anything that contradicts common sense, then the sub-problem type to which the story synopsis belongs under the second evaluation dimension can be considered to conform to common sense. In this case, we can analyze the reasons for the story synopsis conforming to common sense based on the corresponding problem examples. Analyzing the reasons for whether the story synopsis conforms to or contradicts common sense can be understood as analyzing why the story synopsis is similar to the synopsis examples in the corresponding sub-problem type, thus obtaining the sub-problem type to which the story synopsis belongs under the second evaluation dimension and the reasons for belonging to that sub-problem type. As mentioned above, the second evaluation dimension can also include scores corresponding to each sub-problem type. Therefore, the first agent corresponding to the second evaluation dimension can output the sub-problem type to which the story synopsis belongs under the second evaluation dimension, the reasons for belonging to that sub-problem type, and the score.

[0063] Based on the first agent corresponding to the trained third evaluation dimension, the above-mentioned method of using at least one first agent to evaluate the story outline according to at least one evaluation dimension and basic information, based on a preset evaluation process, to obtain the first evaluation result of the story outline, may include:

[0064] The first intelligent agent corresponding to the third evaluation dimension is used to analyze whether there are elements in the story outline that do not conform to the historical context, and the third analysis result is obtained. The elements include other elements in the story outline besides the characters.

[0065] Based on the results of the third analysis, the sub-problem type to which the story outline belongs under the third evaluation dimension is determined;

[0066] Based on the problem examples of the sub-problem type, determine the reasons that belong to that sub-problem type.

[0067] The analysis content corresponding to the third evaluation dimension includes: analyzing whether there are elements in the story synopsis that are inconsistent with the historical context; the evaluation process corresponding to the third evaluation dimension includes: analyzing whether there are elements in the story synopsis that are inconsistent with the historical context, obtaining the third analysis result, wherein the elements include other elements in the story synopsis besides the characters; based on the third analysis result, determining the sub-problem type to which the story synopsis belongs under the third evaluation dimension; and based on the problem examples of the sub-problem type, determining the reasons for belonging to that sub-problem type.

[0068] In practical applications, the text content of the third evaluation dimension (i.e., multiple sub-question types and question examples for each sub-question type), the story outline, the corresponding basic information, and related instructions (such as "Please evaluate the input story outline based on the input basic information and according to the preset evaluation process and evaluation dimensions, and output the sub-question type to which the story outline belongs and the reason for belonging to that sub-question type") can be input into the first agent corresponding to the third evaluation dimension. The first agent can first analyze the story outline to determine the elements appearing in the story outline (elements may include various items, equipment, facilities, environment, buildings, daily necessities, etc., except for the characteristics of characters such as names, genders, personalities, and social relationships). Then, based on the historical background indicated by the basic information, it can analyze whether these elements are consistent with the historical background of the story outline to find elements that are inconsistent with the historical background. If the third analysis result indicates that all elements in the story outline are consistent with the historical background of the story outline, it can be understood that the plot setting of the story outline is completely consistent with the given era and background, all elements and details are realistically possible at that time, and are consistent with the social norms, technological level, and cultural customs of that time. If the third analysis result indicates that all elements in the story synopsis are inconsistent with the historical context, then the story synopsis is considered to be completely inconsistent with the historical context. If the third analysis result indicates that none of the elements in the story synopsis are consistent with the historical context, that is, the plot setting of the story synopsis is impossible in the given era and background, seriously violating the historical facts and scientific common sense of the given historical context, then the story synopsis is considered to be completely inconsistent with the historical context. If the third analysis result indicates that some elements in the story synopsis are inconsistent with the historical context, or that some elements are not valid in the given specific scenario, then the story synopsis is considered to be partially inconsistent with the historical context. Then, based on the third analysis result, the sub-problem type corresponding to the third analysis result can be found, and based on the problem examples of the sub-problem type corresponding to the third analysis result, the reasons why the story synopsis is completely inconsistent with the historical context, partially inconsistent with the historical context, and consistent with the historical context can be analyzed. Among them, analyzing the reasons why the story synopsis is completely inconsistent with the historical context, partially inconsistent with the historical context, and consistent with the historical context can be understood as analyzing the reasons why the story synopsis is similar to the synopsis examples in the problem examples of the corresponding sub-problem type, thereby obtaining the sub-problem type to which the story synopsis belongs under the third evaluation dimension and the reasons why it belongs to that sub-problem type. As mentioned above, the third evaluation dimension can also include scores corresponding to each sub-problem type. Therefore, the first agent corresponding to the third evaluation dimension can also output the sub-problem type to which the story outline belongs under the third evaluation dimension, as well as the reason and score for belonging to that sub-problem type.

[0069] Based on the first agent corresponding to the trained fourth evaluation dimension, the above-mentioned evaluation of the story outline using at least one first agent according to at least one evaluation dimension and basic information, based on a preset evaluation process, to obtain the first evaluation result of the story outline, may include:

[0070] The first agent corresponding to the fourth evaluation dimension analyzes whether the plot in the story outline conforms to the story type, and obtains the fourth analysis result;

[0071] Based on the results of the fourth analysis, the sub-problem type to which the story outline belongs under the fourth evaluation dimension is determined;

[0072] Based on the problem examples of the sub-problem type, determine the reasons that belong to that sub-problem type.

[0073] In practical applications, the text content of the fourth evaluation dimension (i.e., multiple sub-question types and question examples for each sub-question type), the story outline, the corresponding basic information, and related instructions (such as "Please evaluate the input story outline based on the input basic information and according to the preset evaluation process and evaluation dimensions, and output the sub-question type to which the story outline belongs and the reason for belonging to that sub-question type") can be input to the first agent corresponding to the fourth evaluation dimension. The first agent can first analyze the story plot of the story outline to determine the story type of the story outline, and then determine whether the analyzed story type is consistent with the story type in the basic information, thereby realizing the analysis of whether the story outline conforms to the story type indicated by the basic information; if the fourth analysis result indicates that the story outline does not conform to the story type, then the story outline is evaluated in the fourth evaluation. The sub-problem type under the fourth evaluation dimension is that the story outline does not conform to the story type. Then, based on the problem examples corresponding to the story outline not conforming to the story type in the fourth evaluation dimension, the reasons for the story outline not conforming to the story type can be analyzed. Conversely, if the fourth analysis result indicates that the story outline conforms to the story type, then the sub-problem type of the story outline under the fourth evaluation dimension is that the story outline conforms to the story type. Then, based on the problem examples corresponding to the story outline conforming to the story type under the fourth evaluation dimension, the reasons for the story outline conforming to the story type can be analyzed. Analyzing the reasons why the story outline does not conform to or does not conform to the story type can be understood as analyzing why the story outline is similar to the outline examples in the problem examples of the corresponding sub-problem types, thereby obtaining the sub-problem type to which the story outline belongs under the fourth evaluation dimension and the reasons for belonging to that sub-problem type. As mentioned above, the fourth evaluation dimension can also include scores corresponding to each sub-problem type. Therefore, the first agent corresponding to the fourth evaluation dimension can output the sub-problem type to which the story outline belongs under the fourth evaluation dimension, the reasons for belonging to that sub-problem type, and the score.

[0074] In step S13, the target evaluation result of the story outline, determined based on the first evaluation result of the story outline, is displayed. The target evaluation result characterizes whether the story outline meets the requirements for relevance to the topic.

[0075] Optionally, if the first evaluation result of the story synopsis indicates that the story synopsis forcibly achieves the writing purpose, does not conform to common sense, does not conform to or partially conforms to the historical context, or does not conform to any of the seed problem types in the story genre, that is, if the score under some evaluation dimensions in the first evaluation result of the story synopsis is not "1", then the story synopsis is considered to be unsatisfactory in terms of relevance to the topic. Conversely, if the first evaluation result of the story synopsis indicates that the story synopsis normally achieves the writing purpose, conforms to common sense, conforms to the historical context, and conforms to the story genre, that is, if the score under all evaluation dimensions in the first evaluation result of the story synopsis is "1", then the story synopsis is considered to be satisfactory in terms of relevance to the topic.

[0076] Considering that the first agents may make incorrect assessments under the aforementioned evaluation dimensions, to improve the accuracy of the story synopsis's relevance evaluation, in one possible implementation, after the first agents complete the evaluations for each of the aforementioned evaluation dimensions, a second agent can be used to perform a self-checking process. That is, the second agent can be used to comprehensively evaluate the entire story synopsis and the evaluation results of each first agent to identify and point out potential errors in the evaluation results of each first agent, thereby checking the evaluation results of the story synopsis and improving the accuracy of the story synopsis evaluation. Specifically, determining the target evaluation result of the story synopsis based on the first evaluation result can include:

[0077] The second agent generates a second evaluation result for the story outline based on the story outline, basic information, and the first evaluation result of the story outline. The second evaluation result is used to indicate the errors in the first evaluation result.

[0078] Based on the results of the second and first assessments, the target assessment results for the story outline are determined.

[0079] The second agent can also employ a large language model of a known type in the field (such as the general-purpose large language model qwen 2.0), and fine-tune the pre-trained general-purpose large language model using a dataset. For example, the training process of the second agent may include: constructing a dataset comprising multiple sample story outlines and corresponding basic sample information, sample evaluation results for each sample story outline, and sample labels (the sample labels indicate whether the sample evaluation results contain errors, and if so, the errors). The sample evaluation results may include negative and positive sample evaluation results. Negative sample evaluation results can be understood as evaluation results with errors, and positive sample evaluation results can be understood as evaluation results without errors. Evaluation results with errors may include, for example, an incorrect sub-problem type and / or incorrect cause in the output sample story outline. For instance, the story outline should not conform to the historical context, but the evaluation result is consistent with the historical context, or the reason for conforming to the historical context is incorrect. Then… Based on the aforementioned dataset, a sample story synopsis, corresponding basic sample information, sample evaluation results, and related instructions (such as "Based on the input basic sample information and sample story synopsis, determine whether there are errors in the input sample evaluation results, and if there are errors in the input sample evaluation results, output the errors in the sample evaluation results") can be input into a pre-trained general-purpose large language model. This yields a prediction result output by the general-purpose large language model (this prediction result includes whether there are errors in the sample evaluation results, and if there are errors in the input sample evaluation results, the errors present in the sample evaluation results). The loss is calculated based on the difference (or similarity, etc.) between the prediction result and the sample labels of the sample story synopsis, and the parameters of the general-purpose large language model are adjusted based on this loss to obtain a trained second agent. This embodiment of the present disclosure does not limit the training process of the aforementioned second agent.

[0080] In practical applications, the story outline, basic information, and first evaluation result can be input into a trained second agent to obtain the second evaluation result of the story outline output by the second agent. It should be understood that when there are no errors in the first evaluation result, the second evaluation result can indicate that there are no errors in the first evaluation result. At this time, the target evaluation result of the story outline is determined based on the second evaluation result and the first evaluation result, which is equivalent to determining the target evaluation result of the story outline based on the first evaluation result. When an error exists in the first evaluation result, the second evaluation result can indicate the error in the first evaluation result. In this case, determining the target evaluation result of the story outline based on the second evaluation result and the first evaluation result may include: correcting the first evaluation result based on the second evaluation result, that is, correcting the erroneous sub-problem types and / or causes of errors in the first evaluation result, to obtain a corrected first evaluation result. For example, an automated script can be set up to detect the first and second evaluation results. If it is detected that the second evaluation result indicates a sub-problem type and / or cause different from the first evaluation result for any evaluation dimension, then the first evaluation result is corrected based on the second evaluation result, or in other words, the erroneous sub-problem types and / or causes in the first evaluation dimension are modified to the sub-problem types and / or causes given by the first evaluation dimension; or the first evaluation result can be manually modified based on the second evaluation result, which is not limited in this embodiment of the disclosure; then, the target evaluation result of the story outline is determined based on the corrected first evaluation result. This process can refer to the above implementation method of determining the target evaluation result of the story outline based on the first evaluation result of the story outline, which will not be elaborated here. By utilizing the second intelligent agent to comprehensively evaluate the story content of the story outline across various evaluation dimensions based on the aforementioned first evaluation results, and identifying and pointing out possible errors in the first evaluation results, the accuracy of the story outline's relevance to the topic can be improved. The second intelligent agent plays a role in correcting and rectifying errors.

[0081] Based on the topic relevance assessment method provided in the embodiments of this disclosure, the embodiments of this disclosure also provide Figure 2 The diagram illustrates a framework for assessing the relevance of a story synopsis to the topic, as follows: Figure 2As shown, the evaluation framework includes: an input module for inputting basic information C of the story outline and the story outline itself; and a problem discovery module for outputting the first evaluation result of the story outline using the first agent of each evaluation dimension. Specifically, it analyzes the story outline, searches for special actions (i.e., content related to human common sense), and analyzes whether they conform to human common sense to determine if the story outline conforms to human common sense. It also analyzes whether the story outline conforms to the current spatiotemporal context (i.e., the historical context) by searching for special elements (i.e., elements other than characters) and analyzing whether these elements conform to the current spatiotemporal context. Finally, it analyzes the story outline and its story type to determine whether the story type in the story outline matches the story type in the basic information C, thus determining whether the story outline conforms to the current type (i.e., whether it conforms to the basic information C). The system includes: (1) the story type indicated in the basic information; and (2) identifying error types by defining judgment criteria (i.e., analyzing whether the story outline forcibly achieves the writing purpose, and the sub-problem types of forcibly achieving the writing purpose by defining judgment criteria for whether the writing purpose is forcibly achieved). The sub-problem types of forcibly achieving the writing purpose include: the author forcibly achieving the logical purpose (i.e., the plot development logic of the story outline cannot achieve the writing purpose), the protagonist having a cheat code (i.e., the character's behavior lacks rationality), deus ex machina (i.e., the plot development lacks reasonable foreshadowing), and the existence of key logical loopholes and specific details (i.e., the plot development lacks key links or details). The self-checking module is used to use the second intelligent agent to check whether there are errors in the first evaluation result based on the story outline and basic information, so as to correct the first evaluation result and obtain a more accurate evaluation that is relevant to the topic.

[0082] In practical applications, after obtaining the target evaluation results of the story outline, the creator of the story outline can also modify and adjust the story outline based on the reasons for the sub-problem types that appear in the first evaluation results if the target evaluation results indicate that the story outline does not meet the requirements for relevance to the topic. This can help the creator to create high-quality story outlines efficiently.

[0083] According to the evaluation method of this disclosure, by demonstrating the first evaluation result obtained by using at least one first intelligent agent to evaluate the story outline according to at least one evaluation dimension and basic information and in accordance with a preset evaluation process, and then demonstrating the target evaluation result based on the first evaluation result of the story outline to determine whether the relevance of the story outline meets the standard, it is possible to realize the automated evaluation of the relevance of the story outline by using artificial intelligence technology based on at least one evaluation dimension and basic information and a preset evaluation process. The evaluation dimensions used for the relevance evaluation, such as the way the story outline achieves the writing purpose, whether the story outline is in line with the historical context, whether the story outline is in line with the story type, and whether the story outline is in line with common sense, are all to examine whether the story outline is consistent with the constraints indicated by the preset basic information. In this way, combined with artificial intelligence technology, it is possible to quickly and objectively discover possible relevance problems in the story outline. This can greatly improve the evaluation efficiency and quality of the story outline, reduce the burden of manual evaluation, and thus help improve the quality of the story outline.

[0084] According to the evaluation method of this disclosure, a comprehensive evaluation of the story outline is conducted from multiple dimensions, including the way the writing purpose is achieved, common sense, spatiotemporal background, and story type. This constructs a relatively complete and detailed evaluation framework, capable of identifying potential problems in various aspects of the story outline. Each dimension has a detailed workflow and clear judgment criteria, making the evaluation process more operable and accurate. By comparing the story outline with problem examples of given sub-problem types across multiple evaluation dimensions, the first agent can easily explain the reasons for the similarity between the story outline and the corresponding sub-problem type problem examples, providing a reference basis for each evaluation dimension and enhancing the objectivity and reliability of the topic relevance evaluation. The second agent conducts a comprehensive evaluation based on the multi-dimensional evaluation results, discovering and pointing out errors in the first evaluation results, assuming the role of error correction, aiming to improve the quality and topic relevance of the story outline, emphasizing the practical role of content optimization.

[0085] In evaluating the relevance of a story synopsis using the relevance evaluation method of this disclosure, each agent demonstrated excellent performance, achieving an accuracy of 0.87. This accuracy was obtained through rigorous testing on 31 carefully selected data points, meaning that the agent correctly judged and processed 87% of these data points. The recall rate was 0.82, indicating that the agent successfully identified relevant content at a high rate and performed well in finding and extracting key information. The F1 score, considering both accuracy and recall, was 0.84, reflecting a good balance between precision and comprehensiveness, and overall stable and reliable performance.

[0086] Figure 3 This diagram shows a block diagram of a story outline relevance assessment device according to an embodiment of the present disclosure, such as... Figure 3 As shown, the device includes:

[0087] The acquisition module 301 is used to acquire the story outline to be evaluated and the basic information corresponding to the story outline;

[0088] The first display module 302 is used to display the first evaluation result, which includes: the sub-problem type to which the story outline belongs under any evaluation dimension and the reason why it belongs to the sub-problem type; the first evaluation result is obtained by using at least one first agent to evaluate the story outline based on at least one evaluation dimension and the basic information, according to a preset evaluation process, with different first agents corresponding to different evaluation dimensions; the evaluation dimension includes at least one of the following: whether the story outline conforms to the historical background, whether the story outline conforms to the story type, whether the story outline conforms to common sense, and the way the story outline achieves the writing purpose; each evaluation dimension includes multiple sub-problem types and a problem example for each sub-problem type, and the problem example includes a synopsis example belonging to the sub-problem type and the reason why the synopsis example belongs to the sub-problem type;

[0089] The second display module 303 is used to display the target evaluation result of the story outline determined based on the first evaluation result of the story outline, wherein the target evaluation result indicates whether the story outline meets the requirements for relevance to the topic.

[0090] In one possible implementation, the preset evaluation process includes: for any evaluation dimension, analyzing the story outline based on the analysis content corresponding to that evaluation dimension to obtain the analysis result corresponding to that evaluation dimension; determining the sub-problem type corresponding to the story outline under that evaluation dimension based on the analysis result; and determining the reason for belonging to that sub-problem type based on the problem example corresponding to that sub-problem type.

[0091] In one possible implementation, the evaluation dimension includes a first evaluation dimension, which includes: the way the story outline achieves the writing purpose; the multiple sub-problem types corresponding to the first evaluation dimension include: forcibly achieving the writing purpose and normally achieving the writing purpose; the forcibly achieving the writing purpose represents the story outline achieving the writing purpose through illogical plot development and / or character behavior; the normally achieving the writing purpose represents the story outline achieving the writing purpose through logically consistent plot development and / or character behavior; wherein, the step of using at least one first agent to evaluate the story outline based on the at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result of the story outline includes: using the first agent corresponding to the first evaluation dimension to analyze whether the plot development logic in the story outline that achieves the writing purpose can achieve the writing purpose, whether the plot development lacks key links or details, whether the character behavior lacks rationality, and whether the plot development lacks reasonable foreshadowing, to obtain a first analysis result; based on the first analysis result, determining the sub-problem type to which the story outline belongs under the first evaluation dimension; and based on the problem examples of the sub-problem type, determining the reason for belonging to that sub-problem type.

[0092] In one possible implementation, the evaluation dimension includes a second evaluation dimension, which includes whether the story outline conforms to common sense. The second evaluation dimension corresponds to multiple sub-question types, including: the story outline does not conform to common sense and it does. The step of evaluating the story outline using at least one first agent based on the at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result for the story outline, includes: using the first agent corresponding to the second evaluation dimension to analyze whether there is content in the story outline that does not conform to common sense, obtaining a second analysis result; based on the second analysis result, determining the sub-question type to which the story outline belongs under the second evaluation dimension; and based on the question examples of the sub-question type, determining the reason for belonging to that sub-question type.

[0093] In one possible implementation, the evaluation dimension includes a third evaluation dimension, which includes whether the story synopsis conforms to the historical context. The third evaluation dimension corresponds to multiple sub-question types, including: the story synopsis is completely inconsistent with the historical context, partially inconsistent with the historical context, and consistent with the historical context. The step of evaluating the story synopsis using at least one first agent based on the at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result for the story synopsis, includes: using the first agent corresponding to the third evaluation dimension to analyze whether there are elements in the story synopsis that do not conform to the historical context, obtaining a third analysis result, where the elements include elements in the story synopsis other than characters; based on the third analysis result, determining the sub-question type to which the story synopsis belongs under the third evaluation dimension; and based on the question examples of the sub-question type, determining the reason for belonging to that sub-question type.

[0094] In one possible implementation, the evaluation dimension includes a fourth evaluation dimension, which includes whether the story outline conforms to the story type; the multiple sub-question types corresponding to the fourth evaluation dimension include: the story outline does not conform to the story type and the story outline does conform to the story type; wherein, the step of using at least one first agent to evaluate the story outline based on the at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result of the story outline includes: using the first agent corresponding to the fourth evaluation dimension to analyze whether the plot in the story outline conforms to the story type, to obtain a fourth analysis result; based on the fourth analysis result, determining the sub-question type to which the story outline belongs under the fourth evaluation dimension; and based on the question examples of the sub-question type, determining the reason for belonging to that sub-question type.

[0095] In one possible implementation, determining the target evaluation result of the story outline based on the first evaluation result of the story outline includes: using a second agent to generate a second evaluation result of the story outline based on the basic information and the first evaluation result of the story outline, wherein the second evaluation result is used to indicate errors present in the first evaluation result; and determining the target evaluation result of the story outline based on the second evaluation result and the first evaluation result.

[0096] According to the evaluation method of this disclosure, by demonstrating the first evaluation result obtained by using at least one first intelligent agent to evaluate the story outline according to at least one evaluation dimension and basic information and in accordance with a preset evaluation process, and then demonstrating the target evaluation result based on the first evaluation result of the story outline to determine whether the relevance of the story outline meets the standard, it is possible to realize the automated evaluation of the relevance of the story outline based on at least one evaluation dimension and basic information and a preset evaluation process using artificial intelligence technology. The evaluation dimensions used for the relevance evaluation, such as the way the story outline achieves the writing purpose, whether the story outline is in line with the historical context, whether the story outline is in line with the story type, and whether the story is in line with common sense, are all to examine whether the story outline is consistent with the constraints indicated by the preset basic information. In this way, combined with artificial intelligence technology, it is possible to quickly and objectively discover possible relevance problems in the story outline. This can greatly improve the evaluation efficiency and quality of the story outline, reduce the burden of manual evaluation, and thus help improve the quality of the story outline.

[0097] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0098] This disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0099] This disclosure also provides a non-volatile computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0100] This disclosure also provides a computer program product, including a computer program or a non-volatile computer-readable storage medium carrying the computer program, wherein the computer program, when executed by a processor, implements the steps of the above method.

[0101] Figure 4 A block diagram of an electronic device 1900 according to an embodiment of the present disclosure is shown. For example, the electronic device 1900 may be provided as a server or a terminal device. (Refer to...) Figure 4 The electronic device 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions, such as application programs, that can be executed by the processing component 1922. The application programs stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1922 is configured to execute instructions to perform the methods described above.

[0102] Electronic device 1900 may also include a power supply component 1926 configured to perform power management of electronic device 1900, a wired or wireless network interface 1950 configured to connect electronic device 1900 to a network, and an input / output interface 1958 (I / O interface). Electronic device 1900 can operate on an operating system, such as Windows Server, stored in memory 1932. TM Mac OS X TM Unix TM Linux TM FreeBSD TM Or similar.

[0103] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions that can be executed by a processing component 1922 of an electronic device 1900 to perform the above-described method.

[0104] Computer-readable storage media can be tangible devices capable of holding and storing programs / instructions used by instruction execution devices. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0105] The computer program (or computer-readable program instructions) described herein can be downloaded from a computer-readable storage medium to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage medium in the respective computing / processing device.

[0106] The computer program (or computer program instructions) used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuits, such as programmable logic circuits, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), are personalized by utilizing state information of computer-readable program instructions. These electronic circuits can execute computer-readable program instructions to implement various aspects of this disclosure.

[0107] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0108] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0109] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0110] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0111] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for evaluating the relevance of a story synopsis to the topic, characterized in that, include: Obtain the story synopsis to be evaluated and the basic information corresponding to the story synopsis, including the backstory, story type, historical background, and writing purpose; The first evaluation result is displayed, which includes: the sub-problem type to which the story outline belongs under any evaluation dimension and the reason why it belongs to the sub-problem type; the first evaluation result is obtained by evaluating the story outline using at least one first agent based on at least one evaluation dimension and the basic information, according to a preset evaluation process, with different first agents corresponding to different evaluation dimensions; the evaluation dimension includes at least one of the following: whether the story outline conforms to the said historical background, whether the story outline conforms to the said story type, whether the story outline conforms to common sense, and the way the story outline achieves the said writing purpose; each evaluation dimension includes multiple sub-problem types and a question example for each sub-problem type, and the question example includes a synopsis example belonging to the sub-problem type and the reason why the synopsis example belongs to that sub-problem type; The target evaluation result of the story outline, determined based on the first evaluation result, is displayed, and the target evaluation result characterizes whether the story outline meets the requirements for relevance to the topic.

2. The method according to claim 1, characterized in that, The preset evaluation process includes: For any evaluation dimension, the story outline is analyzed based on the analysis content corresponding to that evaluation dimension to obtain the analysis results corresponding to that evaluation dimension; Based on the analysis results, the sub-problem type corresponding to the story outline under this evaluation dimension is determined; Based on the problem examples corresponding to this sub-problem type, determine the reason why it belongs to this sub-problem type.

3. The method according to claim 1 or 2, characterized in that, The evaluation dimensions include a first evaluation dimension, which includes: the way in which the story synopsis achieves the writing purpose; the multiple sub-problem types corresponding to the first evaluation dimension include: forcibly achieving the writing purpose and normally achieving the writing purpose; the forcibly achieving the writing purpose indicates that the story synopsis achieves the writing purpose through illogical plot development and / or character behavior; the normally achieving the writing purpose indicates that the story synopsis achieves the writing purpose through logically consistent plot development and / or character behavior. The step of using at least one first agent to evaluate the story outline based on at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result of the story outline, includes: Using the first intelligent agent corresponding to the first evaluation dimension, analyze whether the plot development logic in the story outline can achieve the writing purpose, whether the plot development lacks key links or details, whether the characters' behavior lacks rationality, and whether the plot development lacks reasonable foreshadowing, to obtain the first analysis result; Based on the first analysis results, determine the sub-problem type to which the story outline belongs under the first evaluation dimension; Based on the problem examples of the sub-problem type, determine the reasons that belong to that sub-problem type.

4. The method according to claim 1 or 2, characterized in that, The evaluation dimension includes a second evaluation dimension, which includes whether the story synopsis conforms to common sense. The multiple sub-question types corresponding to the second evaluation dimension include: the story synopsis does not conform to common sense and the story synopsis does conform to common sense. The step of using at least one first intelligent agent to evaluate the story outline based on at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result of the story outline, includes: The first agent corresponding to the second evaluation dimension is used to analyze whether there is any content in the story outline that does not conform to human common sense, and a second analysis result is obtained. Based on the second analysis results, the sub-problem type to which the story outline belongs under the second evaluation dimension is determined; Based on the problem examples of the sub-problem type, determine the reasons that belong to that sub-problem type.

5. The method according to claim 1 or 2, characterized in that, The evaluation dimension includes a third evaluation dimension, which includes whether the story synopsis conforms to the historical context; wherein, the multiple sub-question types corresponding to the third evaluation dimension include: the story synopsis does not conform to the historical context at all, partially does not conform to the historical context, and conforms to the historical context; The step of using at least one first intelligent agent to evaluate the story outline based on at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result of the story outline, includes: The first agent corresponding to the third evaluation dimension is used to analyze whether there are any elements in the story outline that do not conform to the historical background, and a third analysis result is obtained. The elements include other elements in the story outline besides the characters. Based on the results of the third analysis, the sub-problem type to which the story outline belongs under the third evaluation dimension is determined; Based on the problem examples of the sub-problem type, determine the reasons that belong to that sub-problem type.

6. The method according to claim 1 or 2, characterized in that, The evaluation dimension includes a fourth evaluation dimension, which includes: whether the story synopsis conforms to the story type; the multiple sub-question types corresponding to the fourth evaluation dimension include: the story synopsis does not conform to the story type and the story synopsis does conform to the story type; The step of using at least one first intelligent agent to evaluate the story outline based on at least one evaluation dimension and the basic information, according to a preset evaluation process, to obtain a first evaluation result of the story outline, includes: The first agent corresponding to the fourth evaluation dimension is used to analyze whether the plot in the story outline conforms to the story type, and the fourth analysis result is obtained. Based on the results of the fourth analysis, the sub-problem type to which the story outline belongs under the fourth evaluation dimension is determined; Based on the problem examples of the sub-problem type, determine the reasons that belong to that sub-problem type.

7. The method according to any one of claims 1 to 6, characterized in that, The step of determining the target evaluation result of the story outline based on the first evaluation result of the story outline includes: The second agent generates a second evaluation result for the story outline based on the basic information and the first evaluation result of the story outline. The second evaluation result is used to indicate the errors present in the first evaluation result. Based on the second evaluation result and the first evaluation result, the target evaluation result of the story outline is determined.

8. A device for evaluating the relevance of a story synopsis to the topic, characterized in that, include: The acquisition module is used to retrieve the story outline to be evaluated and the basic information corresponding to the story outline. The basic information includes the backstory, story type, historical background and writing purpose. The first display module is used to display the first evaluation result, which includes: the sub-problem type to which the story outline belongs under any evaluation dimension and the reason why it belongs to the sub-problem type; the first evaluation result is obtained by using at least one first agent to evaluate the story outline based on at least one evaluation dimension and the basic information, according to a preset evaluation process, with different first agents corresponding to different evaluation dimensions; the evaluation dimension includes at least one of the following: whether the story outline conforms to the historical background, whether the story outline conforms to the story type, whether the story outline conforms to common sense, and the way the story outline achieves the writing purpose; each evaluation dimension includes multiple sub-problem types and a problem example for each sub-problem type, the problem example including a synopsis example belonging to the sub-problem type and the reason why the synopsis example belongs to that sub-problem type; The second display module is used to display the target evaluation result of the story outline, which is determined based on the first evaluation result of the story outline. The target evaluation result indicates whether the story outline meets the requirements for relevance to the topic.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

10. A non-volatile computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

11. A computer program product comprising a computer program, or a non-volatile computer-readable storage medium carrying a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images