Role evaluation method and device, electronic equipment and storage medium

Abstract

The invention relates to a role evaluation method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a drama text related to a target role in a script document; extracting a plurality of sub-plot texts corresponding to the target role in multiple dimensions from the plot text; inputting the plurality of sub-plot texts into a multi-dimensional analysis model, so that the multi-dimensional analysis model outputs an evaluation value corresponding to each dimension, and obtaining an evaluation value set of the target role; and based on the evaluation value set, determining a comprehensive evaluation score of the target role in the script document. According to the method, a multi-dimensional role shaping exploding potential analysis prediction model is established, a script role effect can be evaluated and analyzed to a certain extent in a project making preparation stage, and a more accurate direction is provided for optimizing a character shaping evaluation dimension in script evaluation to better meet market requirements and providing a potential commercial value of project broadcasting.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a role evaluation method and device, electronic equipment and storage medium. BACKGROUND

[0002] In the field of project development, the shaping of a role is a key factor in determining whether a work can attract user retention and have commercial value. Generally, by analyzing whether a role in a script can cause "emotional resonance" and "relationship tension" and other key elements, the popularity of the role among the target audience can be predicted.

[0003] However, the current evaluation method combines "manual evaluation and user data analysis", and marks the role attributes by counting keywords, relying on the evaluation experience of experts to judge the integrity of the role development. In this process, the evaluation is easily influenced by personal preferences, and there is a risk of missing key turning points in role shaping. At the same time, the market prediction dimension is relatively single, and it is difficult to capture changes in audience aesthetic preferences. In addition, due to the lack of a quantitative analysis model for "role combination relationships", traditional data analysis methods can only count interaction frequencies, but cannot evaluate and understand changes in emotional intensity; leading to a lack of data support for strategy development in role shaping and plot advancement, making it difficult to accurately position and optimize role settings.

[0004] Therefore, there is an urgent need to develop a role evaluation method, device, electronic equipment and storage medium to solve one or more of the above problems. SUMMARY

[0005] In view of this, in order to solve the above technical problems or part of the technical problems, the embodiments of the present application provide a role evaluation method, device, electronic equipment and storage medium.

[0006] In a first aspect, the present application provides a role evaluation method, the method comprising: obtaining a plot text related to a target role in a script document; extracting a plurality of sub-plot texts corresponding to the target role in multiple dimensions from the plot text; inputting the plurality of sub-plot texts into a multi-dimensional analysis model, so that the multi-dimensional analysis model outputs an evaluation value corresponding to each dimension, and obtains an evaluation value set of the target role; determining a comprehensive evaluation score of the target role in the script document based on the evaluation value set.

[0007] In one possible implementation, the multiple dimensions include role setting, role combination relationship, and role growth trajectory. The extracting a plurality of sub-plot texts corresponding to the target role in multiple dimensions from the plot text comprises: dividing the plot text into multiple plot paragraphs according to a plot scenario; For any plot paragraph, input the plot paragraph into a large language model to make the large language model output description text corresponding to each dimension of the plot paragraph, to obtain character description text, character combination description text, and character state text of the plot paragraph; According to the time sequence of the plot text, the character description texts corresponding to the multiple plot paragraphs are combined to obtain a character profile; According to the time sequence of the plot text, the character combination description texts corresponding to the multiple plot paragraphs are combined to obtain a character combination relationship; According to the time sequence of the plot text, the character state texts corresponding to the multiple plot paragraphs are combined to obtain a character growth trajectory; Based on the character profile, the character combination relationship, and the character growth trajectory, a plurality of sub-plot texts of the target character are generated.

[0008] In one possible implementation, the inputting of the multiple sub-plot texts into the multi-dimensional analysis model to make the multi-dimensional analysis model output an evaluation value corresponding to each dimension to obtain an evaluation value set of the target character includes: Obtaining a character text related to the target character in the script document; Determining the character features and shaping goals of the target character according to the character text; Inputting the character features, shaping goals, and multiple sub-plot texts into a multi-dimensional analysis model to make the multi-dimensional analysis model output an evaluation value corresponding to each dimension in combination with the character features and shaping goals, to obtain an evaluation value set of the target character.

[0009] In one possible implementation, before the inputting of the multiple sub-plot texts into the multi-dimensional analysis model, the method further includes: Judging whether the current role evaluation standard is updated; In the case that the current role evaluation standard is updated, adjusting the parameters of the multi-dimensional analysis model according to the update of the role evaluation standard.

[0010] In one possible implementation, the multi-dimensional analysis model is established as follows: Collecting historical plot texts and role evaluation data corresponding to the historical plot texts, the role evaluation data including evaluation index values of the role in each dimension and a comprehensive evaluation score of the role; Obtaining multiple sub-plot texts corresponding to the character profile, the character combination relationship, and the character growth trajectory in the historical plot text, and labeling the respective evaluation index values in the multiple sub-plot texts; The evaluation value set of the target role, the comprehensive evaluation score of the target role in the script document, and the plurality of sub-plot texts are taken as training samples, and a preset learning model is trained by using the training samples to obtain a multi-dimensional analysis model.

[0011] In one possible implementation, the determining, based on the evaluation value set, of the comprehensive evaluation score of the target role in the script document comprises: obtaining a weight of each role shaping dimension; calculating a weighted sum of the evaluation value of each dimension in the evaluation value set according to the weight to obtain the comprehensive evaluation score of the target role in the script document.

[0012] In one possible implementation, the method further comprises: in a case where the comprehensive evaluation score does not reach a preset passing score, formulating a corresponding optimization strategy according to the evaluation value of each role shaping dimension, and displaying the optimization strategy and the comprehensive evaluation score after the optimization strategy is executed.

[0013] In a second aspect, the present application provides a role evaluation device, the device comprising: an obtaining module configured to obtain a plot text related to a target role in a script document; an extracting module configured to extract a plurality of sub-plot texts corresponding to the target role in a plurality of dimensions from the plot text; a multi-dimensional analysis module configured to input the plurality of sub-plot texts into a multi-dimensional analysis model to enable the multi-dimensional analysis model to output an evaluation value corresponding to each dimension to obtain an evaluation value set of the target role; a comprehensive evaluation module configured to determine a comprehensive evaluation score of the target role in the script document based on the evaluation value set.

[0014] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the role evaluation method of any one of the embodiments of the first aspect when executing the computer program.

[0015] In a fourth aspect, the present application further provides a computer storage medium having a computer program stored thereon, wherein the computer program is executable by a processor to implement the steps of the role evaluation method of any one of the embodiments of the first aspect.

[0016] Compared with the prior art, the above technical solutions provided by the embodiments of the present application have the following advantages: the method provided by the embodiments of the present application establishes a multi-dimensional role shaping blockbuster potential analysis prediction model, can perform certain evaluation analysis on the script role effect in the project preparation stage, optimizes the evaluation dimension of the character shaping in the script evaluation to be more in line with the market demand, and provides a more accurate direction for the potential commercial value of the project broadcast. BRIEF DESCRIPTION OF DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and serve to explain the principles of the present application together with the specification.

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced as follows: obviously, for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0019] One or more embodiments are exemplarily illustrated by pictures in the drawings corresponding thereto, and these exemplary illustrations do not constitute a limitation on the embodiments, elements with the same reference numerals in the drawings represent similar elements, unless otherwise specified, the drawings do not constitute a proportional limitation.

[0020] Figure 1 A flowchart of a role evaluation method provided by the embodiments of the present application is shown in the figure; Figure 2 A flowchart of another role evaluation method provided by the embodiments of the present application is shown in the figure; Figure 3 A flowchart of still another role evaluation method provided by the embodiments of the present application is shown in the figure; Figure 4 A flowchart of a role evaluation method provided by the embodiments of the present application is shown in the figure; Figure 5 A structure diagram of a role evaluation device provided by the embodiments of the present application is shown in the figure; Figure 6 A structure diagram of an electronic device provided by the embodiments of the present application is shown in the figure. DETAILED DESCRIPTION

[0021] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of protection of the present application.

[0022] The following disclosure provides many different embodiments or examples for implementing different structures of the present application. For the purpose of simplicity, the components and arrangements of the various examples are described in the following detailed description. These are, of course, merely examples and are not intended to limit the application. Also, the present application can be implemented in different examples having different combinations of the disclosed features. It is intended that the application be defined by the scope of the appended claims and their equivalents. It is also to be understood that the written description and examples provided herein are intended to be illustrative of the application and not in limited. Changes can be made by one of ordinary skill in the art to the details of the application without departing from the application.

[0023] In order to solve the technical problems in the prior art that the evaluation experience of experts is relied on to judge the integrity of role development, which is easily affected by personal preferences, there is a risk of missing key turning points of character shaping, the dimension of market prediction is relatively single, it is difficult to capture changes in audience aesthetic preferences, there is a lack of quantitative analysis model for "role combination relationship", and traditional data analysis methods can only count interaction frequency and cannot evaluate and understand changes in emotional concentration; leading to the technical problems that the strategy making in role shaping and plot advancing lacks data support, and it is difficult to accurately position and optimize the role setting, the present application provides a role evaluation method, device, electronic equipment and storage medium, which can establish a multi-dimensional role shaping blockbuster potential analysis prediction model, realize certain evaluation and analysis of the script role effect in the project preparation stage, optimize the evaluation dimension of character shaping in the script evaluation to be more in line with market demand, and provide a more accurate direction for the potential commercial value of future project broadcast.

[0024] Figure 1 A flowchart of a role evaluation method provided by the embodiments of the present application is shown in FIG. Figure 1 The method specifically includes the following steps. S101, obtaining a plot text related to a target role in a script document; In the present embodiment, the plot text related to the target role in the script document is obtained In processing the script document, the plot text content related to the specific target role is extracted, and this process involves a detailed analysis and understanding of the document to accurately identify the text directly related to the target role.

[0025] For example, various text analysis techniques such as keyword matching, semantic analysis, and context understanding can be used to effectively filter out the plot passages closely related to the target character from a large amount of text data, thereby providing strong data support for subsequent plot analysis or character behavior research.

[0026] S102、from the plot text corresponding to the target character in multiple dimensions Extract multiple sub-plot texts; In this embodiment, the evaluation of the character generally requires star rating of the character's personality, character combination relationship, and character growth trajectory in the related plot. To this end, multiple sub-plot texts corresponding to the target character in multiple dimensions need to be extracted from the plot text. The sub-plot text records the behavior, personality characteristics, and interaction between the target character and other characters in different situations. Through in-depth analysis of these sub-plot texts, the motivation, emotional changes, and growth trajectory of the target character can be obtained.

[0027] For example, for the character combination relationship dimension, the sub-plot text records the interaction between the target character and other characters, such as cooperation, conflict, emotional communication, etc. The interaction reflects the combination relationship, influence, and importance of the character in the plot. For the character growth trajectory dimension, the sub-plot text records the state changes of the target character in chronological order, including the growth, transformation, or challenges of the character, and then the events together constitute the growth history of the character.

[0028] S103, input multiple sub-plot texts into a multi-dimensional analysis model, so that the multi-dimensional analysis model outputs an evaluation value corresponding to each dimension, and obtains an evaluation value set of the target character; The multi-dimensional analysis model can analyze the plot from multiple angles to more comprehensively evaluate the performance of the target character. This model can process complex plot data, extract key information, and generate corresponding evaluation values for each dimension based on pre-set evaluation standards. These evaluation values reflect the performance of the target character in different aspects, such as the attractiveness of the character's personality, the depth of the character's combination relationship, and the rationality of the character's growth trajectory, etc.

[0029] In this embodiment, by inputting multiple sub-plot texts into a multi-dimensional analysis model, the multi-dimensional model conducts in-depth analysis and evaluation of each dimension, thereby obtaining an evaluation value corresponding to each dimension.

[0030] Specifically, the evaluation process of the multi-dimensional analysis model involves a detailed processing and understanding of the text data, which can accurately capture the key information and subtle differences in the text, and output a series of evaluation values representing the performance and characteristics of the target role in different dimensions. In this way, a comprehensive evaluation value set including the evaluation of each dimension can be obtained, which can reflect the overall performance of the target role in each dimension and provide strong data support for subsequent decision-making and analysis.

[0031] S104, based on the evaluation value set, determining the comprehensive evaluation score of the target role in the script document.

[0032] In this embodiment, when evaluating the target role, the analysis results of the pre-obtained evaluation value set are quantified into specific numerical values to obtain a series of evaluation scores, and then these scores are comprehensively processed through weighted average or other statistical methods to finally determine the comprehensive evaluation score of the target role in the entire script document. The comprehensive evaluation score reflects the overall performance and value of the role in the script, and has important guiding significance for the optimization of subsequent scripts and the shaping of the role.

[0033] The role evaluation method provided by the embodiment of the application first extracts the plot text related to the target role from the script document, then selects the sub-plot text corresponding to the target role in multiple dimensions from the plot text, then inputs the sub-plot text into a multi-dimensional analysis model, so that the model can output the evaluation value corresponding to each dimension, thereby obtaining the evaluation value set of the target role, and finally determines the comprehensive evaluation score of the target role in the script document based on the evaluation value set. By accurately extracting the plot text related to the target role and combining the multi-dimensional analysis model, a comprehensive and objective evaluation of the role performance can be achieved, which not only helps script creators better understand and shape the role and improve the script quality, but also provides a basis for subsequent plot adjustment and role development.

[0034] In an optional scheme of the embodiment of the application, the multi-dimensions include role setting, role combination relationship, and role growth trajectory. It should be noted that in this embodiment, the role shaping dimensions are taken as an example of role setting, role combination relationship, and role growth trajectory; in specific implementation, the role shaping dimensions can be increased or reduced according to actual needs to achieve the purpose of comprehensive evaluation of the target role.

[0035] For example, dimensions such as role psychological changes and emotional entanglements between the role and other roles can also be included to more carefully analyze the inner world and emotional state of the role.

[0036] For example, for specific types of scripts, such as science fiction or historical dramas, additional evaluation dimensions related to the script's background or historical context can be added to more accurately assess the character's performance in specific situations.

[0037] It should be noted that when dealing with different types of scripts, the in-depth analysis and evaluation methods for each character may differ. For example, in a female script, the process of analyzing and evaluating the female lead and male lead characters may differ in terms of the criteria used to evaluate different dimensions due to the diversity of the evaluation objects.

[0038] For example, the multi-dimensional analysis model may produce the following three results when evaluating the female character's setup: First setup evaluation value: The corresponding character setup is consistent with the plot logic and world building, and is logically self-consistent from beginning to end. The backstory is detailed, such as being reborn, being a chosen one, etc., and the pressure points and talents are fully demonstrated, providing sufficient context for the audience. The character can independently achieve self-worth and goals.

[0039] Second setup evaluation value: The corresponding character setup is consistent with the plot logic and world building, but there are inconsistencies in the setup. The character's backstory provides the necessary conditions for the audience's perception, but the logic of the background presentation is not thorough.

[0040] Third setup evaluation value: The corresponding character setup does not conform to the plot logic or world building. The character's backstory lacks sufficient pressure points, and the potential for subsequent audience perception is insufficient.

[0041] For example, the multi-dimensional analysis model may produce the following three results when evaluating the female character's combination relationship (this example uses the combination of the female lead and male lead as an example): First combination evaluation value: The corresponding combination meets the contrast setup. The combination characters have a sense of destiny, and the double-strong protagonists grow together in mutual redemption. There are emotional ups and downs, and the emotional intensity is strong.

[0042] Second combination evaluation value: The corresponding combination setup has contrast, but it is a more conventional contrast setup, lacking innovation compared to previous episodes. The combination characters have a sense of destiny, and the double-strong protagonists support each other, but their joint growth is not well portrayed. The plot is formulaic, and the emotional intensity is not significant.

[0043] Third combination evaluation value: The corresponding combination setup lacks contrast and has similar personalities. The combination characters do not conform to the double-strong protagonists growing together in mutual redemption. There are no emotional ups and downs, and the combination emotional line is not complex.

[0044] For example, the multi-dimensional analysis model may produce the following three results when evaluating the female character's growth trajectory: The first growth evaluation value corresponds to the significant change in the character's personality as the plot develops, reflecting the character's gradual growth and independence, and gradually achieving balance in the pursuit of personal goals and emotions, achieving a win-win situation.

[0045] The second growth evaluation value corresponds to the change in the character's personality as the plot develops, and the character has grown, but has not balanced the pursuit of personal goals and emotions or lost some independence.

[0046] The third growth evaluation value corresponds to the lack of change in the character's personality from beginning to end, remaining flat.

[0047] For example, the multi-dimensional analysis model may have the following three results when analyzing the male character's evaluation value: The first character evaluation value corresponds to the character setting that conforms to the era background and type theme setting; the character has a contrast; respects women and guides and helps women grow.

[0048] The second character evaluation value corresponds to the character setting that conforms to the plot logic and world view construction, but the setting has contradictions before and after; the character lacks contrast; respects women, but helps the female protagonist grow less.

[0049] The second character evaluation value corresponds to the character setting that does not conform to the plot logic or world view construction; the character has no contrast.

[0050] Among them, the multi-dimensional analysis model for analyzing the relationship between male characters (this example takes the combination of the male lead and the female lead as an example) may have the following three results:

[0051] For example, the multi-dimensional analysis model may have the following three results when analyzing the male character's growth trajectory: The first growth evaluation value corresponds to the character having a significant personality change and personal growth, and the emotional relationship with the female lead is constantly increasing.

[0052] The second growth evaluation value corresponds to the character having a personality change, but the contrast change in the emotional relationship with the female lead is not great.

[0053] The third growth evaluation value corresponds to the character being flat, with no personality change or always being self-centered.

[0054] Correspondingly, the following three results may occur when evaluating female characters: The first evaluation result: Character setting is not recommended: female characters are set flat and stereotyped; the CP relationship lacks freshness; the character changes are abrupt.

[0055] Second evaluation result: character setting is recommended, but needs to be adjusted: female character setting breaks away from male perspective, but independent characterization is insufficient; the destiny or contrast setting of the combination relationship needs to be strengthened; the character transformation is logical, but the foreshadowing is insufficient.

[0056] Third evaluation result: character setting is good, with high potential for a blockbuster: female character setting is reasonable; combination setting has strong contrast; character arc is complete, transformation is fully foreshadowed, and theme is echoed.

[0057] Correspondingly, the following three results may occur when evaluating the male character: First evaluation result: character setting is not recommended: male character setting is flat and stereotyped; combination relationship lacks freshness; character transformation is abrupt.

[0058] Second evaluation result: character setting is recommended, but needs to be adjusted: male character setting meets the needs of female independence; the destiny or contrast setting of the CP relationship needs to be strengthened; the character transformation is logical, but the foreshadowing is insufficient.

[0059] Third evaluation result: character setting is good, with high potential for a blockbuster: male character setting is reasonable; CP setting has strong contrast; character arc is complete, transformation is fully foreshadowed, and theme is echoed.

[0060] Taking the female protagonist of the work "Eternal XXX" as an example, the following character evaluation analysis table is obtained, as shown in Table 1: Table 1

[0061] Figure 2 Another flowchart of a role evaluation method provided by an embodiment of the present application is shown in FIG. 2. Figure 2 The method specifically includes the following steps: S201, dividing the plot text into multiple plot paragraphs according to the script plot; In this embodiment, by analyzing the plot text, the key plot nodes of the plot are determined, and the plot text is divided into multiple plot paragraphs according to the script plot. For example, the plot text can be divided into an opening paragraph, a development paragraph, a climax paragraph, and an ending paragraph, etc., each of which corresponds to an important plot or turning point in the script.

[0062] For example, the key events at the beginning of the story are receiving a mysterious task, suddenly encountering an accident, etc., which drive the plot development; the turning points that promote the plot development are discovering important clues, meeting key characters, etc.; the conflict escalation points are the first face-to-face confrontation between the protagonist and the villain, significant disagreements within the team, etc.; the climax part is the final battle, the truth being revealed, etc.; the ending is the protagonist successfully achieving the goal, the story ending in tragedy, etc.

[0063] S202, for any plot paragraph, input the plot paragraph into a large language model, so that the large language model outputs the description text corresponding to each dimension of the plot paragraph, obtaining the role description text, role combination description text, and role state text of the plot paragraph; In this embodiment, for any plot paragraph, the content of the plot paragraph can be input into a large language model. In this way, the large language model can be prompted to analyze the input plot paragraph in depth, and based on its algorithm and training data, output detailed description text of each dimension related to the plot paragraph.

[0064] Among them, the dimensions include but are not limited to role description text, role combination description text, and role state text; the role description text will elaborate the characteristics, personality, background story and other information of each character in the plot; the role combination description text will focus on the interactive relationship, cooperation or conflict between different characters; and the role state text will describe the psychological state, physical state of the character at a specific plot moment, and the influence of the environment on the character. Through the description text of multiple dimensions, the characters and plot development in the plot can be more comprehensively understood.

[0065] S203, combining the role description texts corresponding to multiple plot paragraphs in the time sequence of the plot text to obtain a role character setting; In this embodiment, first, the timeline clues and the context of the story development of the plot text are obtained by reading and analyzing the plot text, and then the descriptive text about each character is extracted from multiple plot paragraphs. These descriptions may include the appearance characteristics, personality characteristics, and behavior habits of the characters. By sorting and integrating these scattered character description information in different paragraphs according to the plot time sequence, the complete character setting of the character is gradually constructed.

[0066] S204, combining the role combination description texts corresponding to multiple plot paragraphs in the time sequence of the plot text to obtain a role combination relationship; In this embodiment, first, based on the analysis of the plot text, the characters mentioned in each plot paragraph are sorted and combined according to the time sequence of the story development, and the combination trajectory of the characters and the character combinations in the story is captured. Then, the interactive description, relationship changes and other information of these character combinations in different paragraphs are integrated, so as to construct the combination relationship between the characters. This relationship may include friendship, love, hostility, cooperation and other types.

[0067] S205, combining the role state texts corresponding to multiple plot paragraphs in the time sequence of the plot text to obtain a role growth trajectory; In this embodiment, first, based on the analysis of the plot text, the character state text described in each plot paragraph is sequentially combined according to the time sequence of the story development to form the growth trajectory of the character.

[0068] Specifically, the character state text includes the psychological state, physical state and the change of the environment of the character at different plot moments, and the changes of these states together constitute the growth process of the character in the plot. For example, the character may have experienced the transformation from ignorance to awakening, from weakness to strength, from loneliness to integration into the collective, which is clearly revealed through the description of different plot paragraphs.

[0069] S206, generating a plurality of sub-plot texts of the target character based on the character profile, the character combination relationship and the character growth trajectory respectively.

[0070] In this embodiment, the sub-plot text of the target character is generated based on the character profile, the character combination relationship and the character growth trajectory. Each corresponding sub-plot text records the specific performance and development process of the target character in different dimensions in chronological order, which can more comprehensively understand the inner world and growth trajectory of the target character, and provide more accurate basis for subsequent plot adjustment and character development.

[0071] Further, by integrating the character profile, the character combination relationship and the character growth trajectory, a comprehensive character evaluation result is obtained, which not only reflects the performance and value of the character in the script, but also provides strong data support for subsequent script optimization and character shaping.

[0072] In specific implementation, the dimensions and weights of the character evaluation can be adjusted according to the type of the script, the audience group and the market demand, etc., to ensure the accuracy and practicability of the evaluation result.

[0073] The character evaluation method provided by the embodiments of the present application divides the plot into a plurality of key paragraphs by analyzing the plot text in detail, and uses a large language model to deeply analyze each paragraph to output detailed description texts about the character profile, the character combination relationship and the character state, which not only summarizes the basic characteristics and personality of the character, but also shows the interaction between the characters and the psychological and physiological state of the character at different plot moments. Then, these description texts are sequentially combined according to the time sequence of the plot development to respectively build the complete profile of the character, the combination relationship between the characters and the growth trajectory of the character, which not only ensures the comprehensiveness and accuracy of the character evaluation, but also provides a solid foundation for subsequent plot creation and character development. Through this method, the inner world and growth process of the character can be more deeply understood, which provides strong support for the optimization of the script and the shaping of the character.

[0074] Figure 3 A flowchart of another role evaluation method provided by an embodiment of the present application is shown in FIG. 3, which specifically includes the following steps. Figure 3 S301, obtaining character text related to a target role in a script document; In addition to the script, the plot document can also include a character profile. The character profile is a detailed description of the character's personality, role goals, etc. in the script, which helps to better understand the character's inner world and behavior motivation.

[0075] In this embodiment, the character profile included in the script document is obtained, and the character text related to the target role is obtained from the character profile.

[0076] It should be noted that in some scripts, the creator may not provide detailed background information about each character. In this case, the character profile may not be included in the script document. These information can be obtained through other ways, such as discussion with the screenwriter or director, or self-study and creation of the role.

[0077] S302, determining the personality characteristics and shaping goals of the target role according to the character text; S303, inputting the personality characteristics, shaping goals, and multiple sub-plot texts into a multi-dimensional analysis model, so that the multi-dimensional analysis model outputs an evaluation value corresponding to each dimension in combination with the personality characteristics and shaping goals, to obtain an evaluation value set of the target role.

[0078] In this embodiment, during the multi-dimensional analysis process of the multi-dimensional analysis model plot, the personality characteristics and shaping goals can be input into the multi-dimensional analysis model at the same time, so that during the multi-dimensional analysis of the multiple sub-plot texts, the evaluation value corresponding to each dimension is output in combination with the personality characteristics and shaping goals, to obtain the evaluation values of the target role in the role setting, role combination relationship, and role growth trajectory three role shaping dimensions.

[0079] Specifically, the multi-dimensional analysis model can comprehensively consider the personality characteristics, shaping goals of the target role, and specific performance in the plot text, so as to generate more accurate and comprehensive evaluation values.

[0080] For example, in the role setting dimension, the model will analyze whether the personality characteristics of the target role are distinctive, stereoscopic, and consistent with the overall style of the script and the expectations of the audience.

[0081] ​The role evaluation method provided in the embodiments of the present application not only considers the role information directly described in the plot text, but also combines the character features and shaping targets of the role, and performs in-depth analysis and evaluation through a multi-dimensional analysis model, so that the real performance and value of the target role in the script can be more accurately revealed, and powerful data support can be provided for subsequent script creation and role development.

[0082] In an optional solution of the embodiments of the present application, before the multiple sub-plot texts are input into the multi-dimensional analysis model, the method further includes: judging whether the current role evaluation standard is updated; in the case that the current role evaluation standard is updated, adjusting the parameters of the multi-dimensional analysis model according to the update of the role evaluation standard.

[0083] In the field of films and television dramas, the evaluation criteria for hit scripts and hit roles are not constant, but always evolve with the iterative upgrade of audience aesthetic cognition and the passage of time.

[0084] In the embodiments, before the multiple sub-plot texts are input into the multi-dimensional analysis model for in-depth analysis of the plot text, it is first necessary to confirm whether the current role evaluation standard is still applicable, in other words, to determine whether the evaluation criteria for hit roles have been updated or changed; in the case that the evaluation criteria are updated, the parameters of the multi-dimensional analysis model need to be adjusted according to the update of the criteria, so as to ensure that the model can analyze and evaluate according to the latest evaluation criteria, so as to obtain more accurate and in-line-with-the-times evaluation results, and ensure the timeliness and accuracy of the role evaluation method.

[0085] The role evaluation method provided in the embodiments of the present application regularly or irregularly checks whether the role evaluation criteria are updated, and is realized by comparing with a preset evaluation criteria database or by receiving an update notification from an external system; once it is found that the evaluation criteria are updated, the multi-dimensional analysis model is automatically or manually adjusted in parameters by the user, so as to ensure that the model can accurately evaluate the role based on the latest evaluation criteria, can adapt to the changing audience aesthetic and market demand, and can provide more practical and accurate role evaluation results for script creators.

[0086] In an optional solution of the embodiments of the present application, the multi-dimensional analysis model is established as follows: collecting historical plot texts and role evaluation data corresponding to the historical plot texts, the role evaluation data including evaluation index values of the role in each dimension and a comprehensive evaluation score of the role; obtain a plurality of sub-plot texts corresponding to the character setting, the character combination relationship, and the character growth track in the historical plot text, and label the respective evaluation index values in the plurality of sub-plot texts; use the evaluation index values of the character in each dimension, the comprehensive evaluation score of the character, and the plurality of sub-plot texts as training samples, and train a preset learning model using the training samples to obtain a multi-dimensional analysis model.

[0087] In this embodiment, the establishment process of the multi-dimensional analysis model needs to be trained according to the historical plot text and the corresponding character evaluation data as training samples.

[0088] First, a large number of plot texts are collected from a historical script library, which cover different types of scripts and characters, ensuring the diversity and comprehensiveness of the data. Then, for each historical plot text, professional character evaluation data is collected, including the evaluation index values of the character in each dimension (such as character setting, character combination relationship, character growth track, etc.), and the comprehensive evaluation score of the character.

[0089] Then, a plurality of sub-plot texts corresponding to the character setting, the character combination relationship, and the character growth track are extracted from the historical plot text, and the respective evaluation index values are labeled during the extraction of the sub-plot texts, so that the subsequent training process can accurately associate the sub-plot texts with the evaluation index values.

[0090] Finally, the collected character evaluation data (including the evaluation index values of the character in each dimension, the comprehensive evaluation score, and the corresponding sub-plot texts) are input into a preset learning model as training samples; through the training process, the model can learn the relationship between the performance of the character in different dimensions and the evaluation index values, thereby having the ability to perform multi-dimensional analysis on the character in a new plot text; after sufficient training, a multi-dimensional analysis model that can be used for actual character evaluation is obtained; the model can comprehensively consider the character's personality characteristics, shaping goals, and specific performance in the plot text, generating more accurate and comprehensive evaluation results, providing strong data support for script creators.

[0091] The establishment method of the multi-dimensional analysis model provided in the embodiments of the present application ensures the accuracy and reliability of the model in the character evaluation task; by training using a large amount of historical plot text and corresponding character evaluation data, the complex relationship between the performance of the character in different dimensions and the evaluation index values can be learned, thereby enabling comprehensive and in-depth analysis and evaluation of the character in a new plot text, improving the efficiency of character evaluation, ensuring the objectivity and accuracy of the evaluation results, and improving the quality and appeal of the script.

[0092] In an optional solution of the embodiment of the application, the determining of the comprehensive evaluation score of the target role in the script document based on the evaluation value set comprises: obtaining the weight of each role shaping dimension; calculating the weighted sum of the evaluation index values of each dimension in the evaluation value set according to the weight to obtain the comprehensive evaluation score of the target role in the script document.

[0093] In the process of role shaping of a film or television play, multiple aspects are often involved to constitute a stereoscopic image of a role. Therefore, in order to comprehensively evaluate the overall performance of the target role in the script document, the weight of each role shaping dimension (such as role setting, role combination relationship, role growth track, etc.) needs to be determined first. After the weight of each dimension is determined, the evaluation index value of each dimension is multiplied by the weight of the dimension according to the weight, and then the weighted values of all dimensions are added to calculate the weighted sum of the evaluation index values of each dimension in the evaluation value set, thereby obtaining the comprehensive evaluation score of the target role in the script document.

[0094] The determination of the weight can be based on multiple factors such as the type of the script, the preference of the audience group, market demand, and the creative intention of the scriptwriter. For example, in some types of scripts, the psychological state and growth history of the role can be more important, so the weight of the role growth track dimension can be increased accordingly.

[0095] It should be noted that the weight of each role shaping dimension is not fixed and can be adjusted according to the specific circumstances of the script to ensure the rationality of the evaluation result.

[0096] The role evaluation method provided by the embodiments of the application can more accurately reflect the comprehensive evaluation score of the target role in the script by determining the weight of each role shaping dimension, comprehensively considers the importance of different dimensions to role shaping, can more comprehensively evaluate the comprehensive value of the target role in the script, and provides strong data support for subsequent script creation and role development.

[0097] In an optional solution of the embodiment of the application, the method further comprises: In the case where the comprehensive evaluation score does not reach the preset standard score, a corresponding optimization strategy is formulated according to the evaluation index value of each role shaping dimension, and the optimization strategy and the comprehensive evaluation score after the optimization strategy is executed are displayed.

[0098] In this embodiment, if the comprehensive evaluation score fails to reach the pre-set passing score, corresponding optimization strategies are formulated according to the evaluation index values of each character shaping dimension. After the optimization strategies are formulated, they are displayed to relevant personnel, and the script is adjusted according to the optimization strategies, and then the script is evaluated again by using the character evaluation device to display the comprehensive evaluation score after the optimization strategies are executed, the differences before and after are compared to ensure the improvement effect of character shaping, and the quality of the script and the attractiveness of the characters can be continuously improved.

[0099] For example, if the evaluation index value of the character setting dimension is low, the distinctness and stereoscopicness of the character personality can be increased, or the background story of the character can be adjusted to enrich the image of the character, so that the character is more consistent with the overall style of the script and the expectations of the audience. Alternatively, if the evaluation index value of the character combination relationship dimension is not ideal, the interaction and conflict between characters can be optimized to enhance the relevance between characters and the tension of the story. The problem of the character growth trajectory dimension can be solved by adjusting the development rhythm of the plot and the experience of the character, so that the growth process of the character is more reasonable and interesting.

[0100] Specifically, the optimization strategies can include increasing or modifying certain key plot points of the character to strengthen the character's personality characteristics or promote the character's growth process; adjusting the interaction between the character and other characters to enhance the dramatic nature of the story or the chemical reaction between the characters; or deepening the psychological state of the character to reveal a more rich inner world of the character. The formulation and execution of these optimization strategies aim to improve the comprehensive evaluation score of the target character in the script, so that it is more consistent with the overall style of the script and the expectations of the audience.

[0101] When the optimization strategies are displayed, various forms such as charts and textual descriptions can be used to intuitively display the improvement of each character shaping dimension and the comprehensive evaluation score after the optimization strategies are executed, providing clear guidance for subsequent script modification and character development. At the same time, through the display of the optimization strategies and their effects, the quality of the script can be continuously improved to achieve the effect of a hit show.

[0102] Figure 4 The flow chart of the character evaluation method provided in the embodiments of the present application is shown in Figure 4 The following steps are mainly included in the analysis of the potential of the character evaluation for shaping a hit show: Step 1: input the script (i.e., the plot text) and the character profile (not optional, i.e., the character text), and extract the main male and female characters from the script by the large model, and proceed to the next step.

[0103] Step 2: call the large language model, input the star level definition of the character setting in the evaluation dimension of the potential of the character shaping for shaping a hit show, summarize the star level rules through the prompt word instruction, and obtain the final star level of this dimension.

[0104] Step three: synchronize step two, call large language model, input the star level definition of CP relationship (i.e. role combination relationship), summarize the star level rule through prompt word instruction, and get the final star level of this dimension.

[0105] Step four: synchronize step two, call large language model, input the star level definition of character arc light (i.e. role growth track), summarize the star level rule through prompt word instruction, and get the final star level of this dimension.

[0106] Step five: calculate the star level of the role.

[0107] Figure 5 The structure diagram of the role evaluation device provided by the embodiment of the application is shown in Figure 5 The device specifically includes: The acquisition module 501 is configured to acquire a plot text related to a target role in a script document. The extraction module 502 is configured to extract a plurality of sub-plot texts corresponding to the target role in multiple dimensions from the plot text. The multi-dimensional analysis module 503 is configured to input the plurality of sub-plot texts into a multi-dimensional analysis model, so that the multi-dimensional analysis model outputs an evaluation value corresponding to each dimension, and obtains an evaluation value set of the target role. The comprehensive evaluation module 504 is configured to determine a comprehensive evaluation score of the target role in the script document based on the evaluation value set.

[0108] In one possible implementation, the extraction module 502 is further configured to divide the plot text into a plurality of plot paragraphs according to a script plot. For any plot paragraph, the plot paragraph is input into a large language model, so that the large language model outputs description text corresponding to each dimension of the plot paragraph, and obtains role description text, role combination description text, and role state text of the plot paragraph; the role description text corresponding to a plurality of plot paragraphs is combined in a time sequence of the plot text to obtain a role profile; the role combination description text corresponding to a plurality of plot paragraphs is combined in a time sequence of the plot text to obtain a role combination relationship; the role state text corresponding to a plurality of plot paragraphs is combined in a time sequence of the plot text to obtain a role growth track; and the plurality of sub-plot texts of the target role are generated based on the role profile, the role combination relationship, and the role growth track.

[0109] In a possible implementation, the multi-dimension analysis module 503 is further configured to acquire character text related to the target role in the script document; determine character features and shaping targets of the target role according to the character text; input the character features, the shaping targets, and the plurality of sub-plot texts into the multi-dimension analysis model, so that the multi-dimension analysis model outputs an evaluation value corresponding to each dimension in combination with the character features and the shaping targets, and obtains an evaluation value set of the target role.

[0110] In a possible implementation, the multi-dimension analysis module 503 is further configured to determine whether the current role evaluation standard is updated; and in the case that the current role evaluation standard is updated, perform parameter adjustment on the multi-dimension analysis model according to the update of the role evaluation standard.

[0111] In a possible implementation, the apparatus further includes an establishment module 505 (not shown in the figure) configured to collect historical plot texts and role evaluation data corresponding to the historical plot texts, the role evaluation data including evaluation index values of the role in each dimension and a comprehensive evaluation score of the role; acquire a plurality of sub-plot texts corresponding to the role character, the role combination relationship, and the role growth track in the historical plot texts, and mark respective evaluation index values in the plurality of sub-plot texts; take the evaluation index values of the role in each dimension, the comprehensive evaluation score of the role, and the plurality of sub-plot texts as training samples, and train a preset learning model by using the training samples to obtain the multi-dimension analysis model.

[0112] In a possible implementation, the comprehensive evaluation module 504 is further configured to acquire weights of each role shaping dimension; calculate a weighted sum of the evaluation index values of each dimension in the evaluation value set according to the weights, and obtain a comprehensive evaluation score of the target role in the script document.

[0113] In a possible implementation, the comprehensive evaluation module 504 is further configured to, in the case that the comprehensive evaluation score does not reach a preset target score, formulate a corresponding optimization strategy according to the evaluation index values of each role shaping dimension, and display the optimization strategy and the comprehensive evaluation score after the optimization strategy is executed.

[0114] The role evaluation apparatus provided in this embodiment can be a role evaluation apparatus as shown in Figure 5 , can perform all steps of role evaluation as shown in Figures 1-4 , and thus realizes the technical effects of role evaluation as shown in Figures 1-4 . For details, refer to the related description, which will not be repeated here for brevity. Figures 1-4

[0115] ​The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0116] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application, such as... Figure 6 As shown, this application embodiment provides an electronic device, including a processor 601, a communication interface 602, a memory 603, and a communication bus 604. The processor 601, communication interface 602, and memory 603 communicate with each other via the communication bus 604. The memory 603 stores computer programs. When the processor 601 executes the program stored in the memory 603, it implements the role evaluation steps provided in any of the aforementioned method embodiments. Obtain the plot text related to the target character from the script document; extract multiple sub-plot texts corresponding to the target character in multiple dimensions from the plot text; input the multiple sub-plot texts into a multi-dimensional analysis model so that the multi-dimensional analysis model outputs the evaluation value corresponding to each dimension, thereby obtaining the evaluation value set of the target character; and determine the comprehensive evaluation score of the target character in the script document based on the evaluation value set.

[0117] In one possible implementation, the plot text is divided into multiple plot segments according to the script's plot; For any given plot segment, the plot segment is input into a large language model, which outputs descriptive text for each dimension of the plot segment, resulting in character description text, character combination description text, and character status text for that plot segment. The character description texts corresponding to multiple plot segments are combined according to the chronological order of the plot texts to obtain the character's persona. The character combination description texts corresponding to multiple plot segments are combined according to the chronological order of the plot texts to obtain the character combination relationship. The character status texts corresponding to multiple plot segments are combined according to the chronological order of the plot texts to obtain the character's growth trajectory. Based on the character persona, the character combination relationship, and the character growth trajectory, multiple sub-plot texts for the target character are generated.

[0118] In a possible implementation, the character text related to the target role in the script document is acquired; the character feature and shaping target of the target role are determined according to the character text; and the character feature, shaping target and multiple sub-plot texts are input into the multi-dimensional analysis model, so that the multi-dimensional analysis model outputs an evaluation value corresponding to each dimension in combination with the character feature and shaping target, and an evaluation value set of the target role is obtained.

[0119] In a possible implementation, it is determined whether the current role evaluation standard is updated; and in the case that the current role evaluation standard is updated, the multi-dimensional analysis model is adjusted in parameters according to the update of the role evaluation standard.

[0120] In a possible implementation, historical plot texts and role evaluation data corresponding to the historical plot texts are collected, the role evaluation data including evaluation index values of the role in each dimension and a comprehensive evaluation score of the role; multiple sub-plot texts corresponding to the role character, role combination relationship and role growth track in the historical plot texts are acquired, and respective evaluation index values are labeled in the multiple sub-plot texts; the evaluation index values of the role in each dimension, the comprehensive evaluation score of the role and the multiple sub-plot texts are taken as training samples, and a preset learning model is trained by using the training samples, so as to obtain the multi-dimensional analysis model.

[0121] In a possible implementation, the weight of each role shaping dimension is acquired; and a weighted sum of the evaluation index values of each dimension in the evaluation value set is calculated according to the weight, so as to obtain a comprehensive evaluation score of the target role in the script document.

[0122] In a possible implementation, in the case that the comprehensive evaluation score does not reach a preset standard score, a corresponding optimization strategy is formulated according to the evaluation index value of each role shaping dimension, and the optimization strategy and the comprehensive evaluation score after the optimization strategy is executed are displayed.

[0123] Through the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be implemented by means of software plus a general hardware platform, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions essentially or in other words, the part that contributes to the related art can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0124] It is to be understood that the terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. The terms "comprises", "comprising", "includes", "including" and "has" are inclusive and therefore specify the presence of stated features, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and / or groups thereof. The method steps, processes, and operations described herein are not to be construed as necessarily requiring their performance in the particular order

[0125] The above description is merely that of the specific embodiments of the application and as such is not to be taken in a limiting sense. Various modifications and alterations of the embodiments described herein will become apparent to those skilled in the art from the foregoing description, which does not limit the generality presented. It is the intention that all such modifications and alterations be considered equaliy by the spirit and scope of this application. It is therefore intended to cover in the appended claims all such changes and alterations that come within the scope of this application.

Claims

1. A method of character assessment, characterized by, The method comprises the following steps: obtaining a plot text related to a target character in a script document; extracting a plurality of sub-plot texts corresponding to the target character in multiple dimensions from the plot text; inputting the plurality of sub-plot texts into a multi-dimensional analysis model to make the multi-dimensional analysis model output an evaluation value corresponding to each dimension, thereby obtaining an evaluation value set of the target character; based on the evaluation value set, determining a comprehensive evaluation score of the target character in the script document.

2. The method of claim 1, wherein, The multiple dimensions include character setting, character combination relationship, and character growth trajectory. The method of extracting a plurality of sub-plot texts corresponding to the target character in multiple dimensions from the plot text comprises the following steps: dividing the plot text into a plurality of plot paragraphs according to the plot scenario; for any plot paragraph, inputting the plot paragraph into a large language model to make the large language model output a description text corresponding to each dimension of the plot paragraph, thereby obtaining a character description text, a character combination description text, and a character state text of the plot paragraph; combining the character description texts corresponding to a plurality of plot paragraphs in the time sequence of the plot text to obtain a character setting; combining the character combination description texts corresponding to a plurality of plot paragraphs in the time sequence of the plot text to obtain a character combination relationship; combining the character state texts corresponding to a plurality of plot paragraphs in the time sequence of the plot text to obtain a character growth trajectory; generating a plurality of sub-plot texts of the target character based on the character setting, the character combination relationship, and the character growth trajectory.

3. The method of claim 1, wherein, The method of inputting a plurality of sub-plot texts into a multi-dimensional analysis model to make the multi-dimensional analysis model output an evaluation value corresponding to each dimension, thereby obtaining an evaluation value set of the target character, comprises the following steps: obtaining a character text related to the target character in the script document; determining the character trait and shaping target of the target character according to the character text; inputting the character trait, shaping target, and a plurality of sub-plot texts into a multi-dimensional analysis model to make the multi-dimensional analysis model output an evaluation value corresponding to each dimension in combination with the character trait and shaping target, thereby obtaining an evaluation value set of the target character.

4. The method of claim 2, wherein, Before inputting a plurality of sub-plot texts into a multi-dimensional analysis model, the method further comprises the following steps: determining whether the current character evaluation standard is updated; in the case that the current character evaluation standard is updated, adjusting the parameters of the multi-dimensional analysis model according to the update of the character evaluation standard.

5. The method of claim 1, wherein, The method of establishing the multi-dimensional analysis model comprises the following steps: collecting historical plot texts and character evaluation data corresponding to the historical plot texts, wherein the character evaluation data comprises evaluation index values of the character in each dimension and a comprehensive evaluation score of the character; obtaining a plurality of sub-plot texts corresponding to the character setting, character combination relationship, and character growth trajectory in the historical plot texts, and labeling the corresponding evaluation index values in the plurality of sub-plot texts; The evaluation indicator value of the character in each dimension, the comprehensive evaluation score of the character, and the plurality of sub-plot texts are taken as training samples, and a preset learning model is trained using the training samples to obtain a multi-dimensional analysis model.

6. The method of claim 1, wherein, The method further includes: The method further includes: In a case where the comprehensive evaluation score does not reach a preset standard score, an optimization strategy corresponding to each character shaping dimension is formulated according to the evaluation indicator value of each character shaping dimension, and the optimization strategy and the comprehensive evaluation score after the optimization strategy is executed are displayed.

7. The method of claim 1, wherein, The method further includes: The method further includes:

8. A character evaluation device characterized by comprising: The method further includes: An obtaining module is configured to obtain plot texts related to a target character in a script document; An extracting module is configured to extract a plurality of sub-plot texts corresponding to the target character in multiple dimensions from the plot texts; A multi-dimensional analysis module is configured to input the plurality of sub-plot texts into a multi-dimensional analysis model, so that the multi-dimensional analysis model outputs an evaluation value corresponding to each dimension to obtain an evaluation value set of the target character; A comprehensive evaluation module is configured to determine a comprehensive evaluation score of the target character in the script document based on the evaluation value set.

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

10. A storage medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to implement the steps of the character evaluation method in any one of claims 1-7.

Images