Scenario quality evaluation method, apparatus and device, and storage medium

Abstract

The invention relates to a script quality evaluation method and device, equipment and a storage medium. The script quality evaluation method comprises the steps of obtaining a script text, and determining a plurality of target roles included in the script text; determining at least one interactive plot text recording a plurality of target roles in the script text; through a pre-trained sentiment analysis model, the sentiment relationship of the multiple target roles is analyzed based on the at least one interaction plot text to obtain a sentiment analysis result, and the sentiment analysis result refers to semantic content which is represented by the interaction plot text and reflects sentiment features between the roles; and according to the sentiment analysis result and the at least one interactive plot text, performing risk quantification on the sentiment promotion process among the plurality of target roles through a set risk evaluation index to generate a sentiment quality evaluation result of the script text. According to the method provided by the invention, risk quantification is carried out on the emotion promotion process according to a unified standard, so that the script quality can be accurately controlled.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a script quality assessment method, apparatus, device, and storage medium. Background Technology

[0002] In the realm of scriptwriting, particularly romance dramas and other genres where emotional storylines drive the plot, the emotional interactions between characters are crucial for plot development, enhancing audience engagement, and directly determining script quality. Currently, assessing the risks associated with developing emotional storylines in scripts still heavily relies on subjective judgment based on experience. This includes methods such as the writing team subjectively evaluating the rationality and emotional impact of emotional progression based on their experience, or comparing past successful examples with established patterns. However, this subjective approach has significant limitations: firstly, it is susceptible to individual differences in understanding, leading to a lack of standardized assessment criteria; secondly, it fails to visually represent the pace and dynamic trends of emotional development, making it difficult to adapt to the rapidly evolving aesthetic demands of the audience, thus significantly limiting its effectiveness in early-stage script quality control. Summary of the Invention

[0003] To address the aforementioned technical problems, this disclosure provides a script quality assessment method, apparatus, device, and storage medium.

[0004] In a first aspect, embodiments of this disclosure provide a script quality assessment method, including: Obtain the script text and identify the multiple target characters included in the script text; Identify at least one interactive plot text that records multiple target characters in the script text; By using a pre-trained sentiment analysis model, the emotional relationships of multiple target characters are analyzed based on at least one interactive plot text to obtain sentiment analysis results. The sentiment analysis results refer to the semantic content of the interactive plot text that represents and reflects the emotional characteristics between the characters. Based on the sentiment analysis results and at least one interactive plot text, the risk of the emotional progression process between multiple target characters is quantified by setting risk assessment indicators to generate the sentiment quality assessment results of the script text.

[0005] Secondly, embodiments of this disclosure provide a script quality assessment apparatus, comprising: The acquisition unit is used to acquire the script text and identify multiple target characters included in the script text; A text generation unit is used to identify at least one interactive plot text for multiple target characters recorded in the script text; The sentiment analysis unit is used to analyze the emotional relationships of multiple target characters based on at least one interactive plot text using a pre-trained sentiment analysis model, and obtain sentiment analysis results. The sentiment analysis results refer to the semantic content represented by the interactive plot text that reflects the emotional characteristics between the characters. The results generation unit is used to quantify the risk of the emotional progression process between multiple target characters based on the sentiment analysis results and at least one interactive plot text, through a set risk assessment index, in order to generate the sentiment quality assessment results of the script text.

[0006] Thirdly, embodiments of this disclosure provide an electronic device, including: Memory; Processor; and Computer programs; The computer program is stored in memory and configured to be executed by a processor to implement the first aspect of the method described above.

[0007] Fourthly, embodiments of this disclosure provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect above.

[0008] The script quality assessment method disclosed herein includes: acquiring the script text and identifying multiple target characters included in the script text; identifying at least one interactive plot text recording the multiple target characters in the script text; analyzing the emotional relationships of the multiple target characters based on the at least one interactive plot text using a pre-trained sentiment analysis model to obtain sentiment analysis results, wherein the sentiment analysis results refer to the semantic content represented by the interactive plot text reflecting the emotional characteristics between the characters; and quantifying the risk of the emotional progression process between the multiple target characters based on the sentiment analysis results and the at least one interactive plot text using a set risk assessment index to generate a sentiment quality assessment result for the script text. The method provided in this application quantifies the risk of the emotional progression process according to a unified standard, thereby enabling precise control over script quality. Attached Figure Description

[0009] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0010] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A schematic flowchart illustrating a script quality assessment method provided in this embodiment of the disclosure; Figure 2 An emotion trend chart provided for embodiments of this disclosure; Figure 3 for Figure 1 A detailed flowchart of S104 in a script quality assessment method is shown. Figure 4 A flowchart illustrating a character's emotional line analysis provided in this embodiment of the disclosure; Figure 5 This is a schematic diagram of the structure of a script quality assessment device provided in an embodiment of the present disclosure; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0012] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0013] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0014] To address the aforementioned technical problems, this disclosure provides a method for evaluating script quality. This will be described in detail through one or more of the following embodiments.

[0015] The script quality assessment method provided in this disclosure is applicable to script assessment scenarios involving emotional aspects. This method can be executed by a script quality assessment device, which can be implemented in software and / or hardware and can be integrated into an electronic device. The electronic device can include, but is not limited to, mobile terminals such as smartphones, laptops, digital radio receivers, personal digital assistants (PDAs), tablet computers (Tablet PCs), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), wearable devices, etc., as well as fixed terminals such as digital televisions, desktop computers, smart home devices, etc.

[0016] Figure 1 A flowchart illustrating a script quality assessment method provided in this disclosure embodiment specifically includes, as follows: Figure 1The following steps are shown: S101. Obtain the script text and identify the multiple target characters included in the script text.

[0017] Understandably, scripts undergoing emotional plot analysis can be historical romances, contemporary love stories, or similar genres where the plot is driven by emotion. Quantitatively assessing the emotional tension between characters, such as "fated entanglements" and "love-hate struggles," has become crucial for predicting the performance of film and television dramas. The script text can specifically consist of at least a portion of the episodes in a complete script; for example, the script text might include the first 10 episodes of a 30-episode script. A script character refers to a character entity within the script text that possesses an independent identity, participates in the plot development, and carries out emotional expression or behavioral driving functions. These typically include protagonists, supporting characters, and functional characters. Multiple script characters are the target characters for emotional plot analysis, meaning analyzing the emotional lines between multiple script characters, such as analyzing the emotional line between the male and female protagonists. The following example uses a romance script for detailed explanation.

[0018] Understandably, the character emotional arc is a crucial component in narrative works, depicting character growth and change. It primarily refers to the trajectory of a character's emotional changes throughout the story, reflecting not only their inner world, personality transformation, and evolving motivations, but also directly impacting the audience's or reader's empathy and the overall emotional impact of the work. Specifically, the romantic arc between the male and female leads refers to the development and evolution of their love relationship within the script. This includes not only the budding, development, and culmination of their love, but also their emotional conflicts, contradictions, trials, and ultimate fate. The purpose of advancing the romantic arc is to drive the plot forward, triggering new events, conflicts, and crises, shaping character traits, showcasing their personality traits, values, and emotional needs, making them more three-dimensional and well-rounded, and ultimately evoking audience resonance. Furthermore, the emotional arc is a vital means of engaging the audience, evoking resonance and immersion, allowing them to become more engaged in the story. A key aspect of script quality assessment lies in whether there are risks in the emotional groundwork, turning points, and character motivations, and whether the emotional progression ultimately aligns with character development and the main plot theme.

[0019] Optionally, multiple target characters included in the script text can be identified, which can be achieved through the following steps: Extract multiple script characters from the script text; determine multiple target characters based on the attribute information of the multiple script characters in the script text, wherein the attribute information includes the number of times the character appears in the script text and / or the number of descriptive paragraphs; or, obtain the character description statements to be subjected to sentiment analysis, and determine multiple target characters among the multiple script characters based on the character description statements.

[0020] Understandably, multiple target characters can be directly input by the user, or determined based on the script outline, such as the male and female leads. Multiple characters involved in the plot development can also be extracted from the script text, such as all important characters. Specifically, multiple characters can be extracted using named entity recognition; other possible extraction methods are not elaborated upon. Subsequently, the attributes of multiple characters in the script text are analyzed to obtain attribute information. This attribute information includes the number of times a character appears and / or the number of descriptive paragraphs associated with them. Attribute information is used to measure a character's participation and influence in the plot development. The number of appearances can be the average number of appearances per episode or the total number of appearances across all episodes. The number of descriptive paragraphs reflects the character's importance in the script; for example, the protagonist has a relatively large amount of descriptive text, descriptive paragraphs, descriptive text, and descriptive length. After determining the attribute information of the characters, the final multiple target characters are determined based on this information. For example, the top N characters in terms of appearance count can be selected as the final target characters, and / or the top N characters in terms of the number of descriptive paragraphs can be selected as the final target characters. Other methods for determining the final target characters based on attribute information are not limited. In addition, multiple target characters can be identified based on user-defined character descriptions, such as identifying the male and female leads or the second male and female leads as the final target characters. Other methods for determining the final target characters are not limited. This method of character identification reduces issues such as misjudgment, omissions, or unreasonable weighting when dealing with complex character relationships and ambiguous character naming in historical / modern romance contexts, further improving the accuracy of emotional analysis and market performance prediction.

[0021] S102. Identify at least one interactive plot text in the script text that records multiple target characters.

[0022] Understandably, based on the above S101, the interactive plots between multiple target characters in the script text are summarized and analyzed to determine at least one interactive plot text that matches multiple target characters. Here, interactive plot text refers to a text fragment describing the emotional, behavioral, or linguistic interactions between two or more target characters in a specific plot context, used to portray character relationships, drive plot development, and reflect character personalities and motivations. Interactive plot text includes character identification, character behavior descriptions, and contextual background information. Understandably, at least one interactive plot text can be extracted from each episode text. For example, character A and character B meet at location C, and character A is forced by character B to open the seal at location C, where character A and character B are multiple target characters.

[0023] Optionally, the script text may include at least one interactive plot text that records multiple target characters, which can be achieved through the following steps: Select at least one interactive plot text from the script text in which multiple target characters interact; analyze whether the at least one interactive plot text contains effective interactive plots using a pre-trained large language model, and output the interactive plot text containing effective interactive plots. Here, effective interactive plots refer to changes in the emotional relationship between multiple target characters at interactive plot nodes.

[0024] Understandably, interactive plot texts depicting interactions between multiple target characters are selected from the script text. These interactive plot texts can be the original descriptive text from the script, referring to unprocessed descriptive statements, such as character A casually completing a task and encountering character B at location C. Interactive plot texts can also be simplified descriptive texts resulting from textual analysis of the original descriptive text, for example, character A encountering character B at location C while completing a task. The selected interactive plot texts are then used as input to a pre-trained large language model. Alternatively, the script text itself can be directly used as input to the language model, meaning each episode's script is fed into the large language model. Subsequently, the language model analyzes whether the interactive plot texts contain valid interactive elements. Furthermore, it is used to determine whether the interactive content described in the text contributes to evaluating the quality of emotional exchange between characters and advancing the plot, marking it as valid or invalid interaction. Effective interaction refers to a significant development and change in the emotional relationship between two characters at important plot points. In other words, effective interaction substantially impacts emotional development, plot progression, or changes in character relationships. For example, if character A is saved by character B, character A's feelings towards character B might change from indifference to affection due to the life-saving act. Ineffective interaction, on the other hand, refers to superficial interactions between characters that lack substantial information transmission and make no significant contribution to plot development or emotional evolution. Removing ineffective interaction would not affect the overall narrative or drive emotional change. For example, character A and character B might simply brush past each other.

[0025] Understandably, the effectiveness of interactive plots can also be analyzed from multiple dimensions, such as whether they advance the plot, change character relationships, convey clear emotions, have memorable moments, and can be deleted without affecting the plot.

[0026] Optionally, an interactive plot text is analyzed using a pre-trained large language model to determine whether it contains valid interactive plots, and the interactive plot text containing valid interactive plots is output, including: Summarize the text content of at least one interactive plot text to generate at least one summarized interactive plot text; analyze whether the summarized at least one interactive plot text contains valid interactive plots using a pre-trained large language model, and output the interactive plot text containing valid interactive plots.

[0027] Understandably, the original descriptive text can be summarized to generate an interactive summary text. For example, the interactive summary text could be "Character A and Character B meet at location C." Compared to the original descriptive text, the interactive summary text more concisely reflects the interaction between the characters. In this case, the interactive summary text can also be directly used as the interactive plot text.

[0028] Optionally, the training process for a large language model includes the following steps: Construct a training dataset, which includes interactive plot samples and valid or invalid labels for the interactive plot samples; use the training dataset as input to the large language model to fine-tune the model parameters; use the test dataset as input to the fine-tuned large language model to evaluate the model's performance; optimize the model parameters based on the evaluation results to obtain the trained large language model.

[0029] Understandably, a large amount of romantic drama scripts, novel excerpts, or emotional interaction plots written by professional screenwriters are collected as raw data, and a large amount of interactive plot text is extracted from them as training samples. Several professional screenwriters or film and television evaluation experts are hired to label each interactive plot sample according to pre-set unified standards (such as whether it drives changes in character relationships, whether it advances the core conflict, whether it has emotional tension, etc.). The labeling results are usually binary classification, i.e., effective interaction or ineffective interaction, or they can be labeled with their effectiveness level (e.g., 1-5 points). Subsequently, the opinions of multiple experts are combined to generate ground truth labels for each interactive plot sample. A general, pre-trained large language model (such as BERT, RoBERTa, etc.) is selected as the base model, and the training dataset with effective / ineffective labels is input into the base model for supervised fine-tuning. During training, the model associates the features of the input interactive text with the effective / ineffective labels, thereby adjusting its internal parameters. After training, the model's performance is evaluated using a batch of test datasets that were not used in the training. The degree to which its judgments match the true values ​​(such as accuracy and F1 score) is checked. Based on the evaluation results, the model is further optimized or supplemented with training data until its performance reaches a satisfactory level.

[0030] Among them, multiple target roles include the first role and the second role.

[0031] Understandably, for example, the first role mentioned above is role A, and the second role is role B.

[0032] Optionally, at least one interactive plot text describing the plurality of target characters is determined in the script text, which can be achieved through the following steps: Based on the script text, generate a first interactive text describing the interaction between the first character and the second character from the perspective of the first character; generate a second interactive text describing the interaction between the second character and the first character from the perspective of the second character from the perspective of the second character; wherein, the interactive plot text includes the first interactive text and the second interactive text.

[0033] Understandably, the first-person perspective refers to the character's viewpoint as the narrator or observer, reflecting their subjective feelings. From the first character's perspective, extracting or summarizing the interaction between them and the second character from the script text yields the first interactive text. For example, meeting character B at location C and being forced by character B to open the seal of location C. From the second character's perspective, extracting or summarizing the interaction between them and the first character from the script text yields the second interactive text. For example, meeting character A at location C and forcing character A to open the seal of location C. In this case, the interactive plot text refers to a complete collection of interactive segments including the first and second interactive texts. It encompasses the two-way emotional expression, behavioral reactions, and cognitive differences between the characters. The emotional differences between the two perspectives (e.g., A to B vs. B to A) better reflect the complex emotional relationships between the characters.

[0034] Understandably, the first and second interactive texts may correspond or not, meaning that the interactive plots of characters from different perspectives may be different.

[0035] Understandably, the primary character can be used as the target character for the emotional arc to be diagnosed. The other characters in the script interact with the primary character; that is, from the primary character's perspective, the interaction scenes between the primary character and the other characters in the script are selected—a one-to-many format. For example, from the female protagonist's perspective, individual or joint interaction scenes between her and the male protagonist, male supporting character, and female supporting character are selected, resulting in at least one interaction scene text. The determination of the interaction scene texts between multiple characters in the script can be set according to user needs and will not be elaborated here.

[0036] S103. Using a pre-trained sentiment analysis model, analyze the emotional relationships of multiple target characters based on at least one interactive plot text to obtain sentiment analysis results.

[0037] Among them, the sentiment analysis results refer to the semantic content of the interactive plot text that reflects the emotional characteristics between the characters.

[0038] Understandably, based on S102 above, sentiment analysis refers to the use of natural language processing technology to identify emotional tendencies and relationships in interactive plot texts. Sentiment analysis results include multiple sentiment information (sentiment features). Each interactive plot text may yield at least one sentiment information, which is used to characterize the emotional states between characters (such as affection, hostility, dependence, etc.), sentiment types (such as positive, negative), interaction patterns, and sentiment intensity. Specific sentiment analysis methods are not limited.

[0039] S104. Based on the sentiment analysis results and at least one interactive plot text, quantify the risk of the emotional progression process between multiple target characters through the set risk assessment indicators, so as to generate the sentiment quality assessment results of the script text.

[0040] Understandably, based on the aforementioned S103, and using the results of sentiment analysis and interactive plot text, a pre-defined risk assessment system is used to objectively measure the development trajectory and progression of the emotional relationships between characters, generating a comprehensive evaluation report on the quality of the script's emotional storyline. Key risk assessment indicators typically include: Pace of emotional progression risk, used to determine if the pace of emotional development is too uniformly fast; excessively fast development may be unconvincing to the audience, while excessively slow development may lead to a dragging plot. Risk assessment indicators may also include emotional logical coherence risk, used to determine if the emotional changes resulting from a particular interaction are supported by reasonable motivations and if there are abrupt jumps in relationships. Risk assessment indicators may also include emotional progression stability risk, used to determine if the emotional relationship deepens steadily through effective interactions or stagnates for a long period, or even if plot points contradict the characters' established traits. Risk assessment indicators may also include emotional tension risk, used to determine if the entire emotional storyline possesses sufficient conflict and appeal, avoiding a bland and unengaging narrative. Subsequently, the quantitative results of all indicators are integrated to form a comprehensive risk score or level, and specific weak links and high-risk sections in the emotional progression process are clearly identified, thereby generating a structured emotional quality assessment report and completing the script instruction assessment in terms of emotion.

[0041] The script quality assessment method provided in this disclosure constructs a complete system for predicting and evaluating character emotional arcs through in-depth analysis of the characters' emotional arcs in the script. It can not only accurately reconstruct the emotional evolution between characters in the script, but also predict their future acceptance by audiences and market performance, thus providing strong data support and technical assurance for script evaluation, content optimization, and business decisions.

[0042] Based on the above embodiments, Figure 3 for Figure 1The diagram illustrates a detailed process flow of S104 in a script quality assessment method. Optionally, based on the sentiment analysis results and at least one interactive plot text, the risk of the emotional progression process between multiple target characters is quantified using set risk assessment indicators to generate a sentiment quality assessment result for the script text. This specifically includes, for example: Figure 3 The following steps are shown: S301. Obtain a pre-built emotional interaction level table.

[0043] The emotional interaction rating scale is used to represent the correspondence between emotional characteristics and rating values.

[0044] Understandably, the emotional interaction rating scale can be understood as a scoring rule for interactive plot points in terms of character emotions, used to calculate scores for subsequent related plot points. The emotional interaction rating scale is used to represent the correspondence between various emotional characteristics and rating values; for example, one rating value corresponds to emotional characteristics across multiple dimensions.

[0045] The emotional characteristics in the emotional interaction level table include at least one of the following: emotional descriptive words, interaction characteristics, power characteristics, and expression characteristics. Among them, emotional connection refers to words that describe the emotional connection state between characters, interaction characteristics are used to characterize the emotional relationship and interaction pattern when characters interact, power characteristics are used to characterize the power relationship between characters, and expression characteristics are used to characterize the expression form of emotions between characters.

[0046] Understandably, the emotional characteristics in the emotional interaction level table include at least one of the following: emotional descriptive words, interaction characteristics, power characteristics, and expression characteristics. Emotional descriptive words refer to words describing the degree of emotion between characters, such as "bound by fate," "inseparable," and "mutually hostile." Interaction characteristics refer to the type of emotional relationship and interaction methods manifested when characters interact in specific plots; for example, the interaction characteristic corresponding to "bound by fate" is "mutual achievement, shared growth, and inseparability." Power characteristics refer to the relationship between characters in terms of status, power, and control, which can influence the way characters interact and the freedom of their emotional expression; for example, the power characteristic corresponding to "bound by fate" is "balanced power, mutual respect, and joint decision-making." Expression characteristics refer to the way characters express their emotions and their attitudes; for example, the expression characteristic corresponding to "bound by fate" is "frank and direct, full of love and gratitude." The emotional interaction level table is shown in Table 1.

[0047] Table 1:

[0048] Understandably, at least one dimension of emotional information can be extracted from at least one interactive plot text to match the emotional information in at least one dimension with emotional features in multiple dimensions of the rating table, thereby determining the interaction value. Other possible dimensions of emotional features are not elaborated upon.

[0049] S302. Perform a matching analysis between the sentiment features in the sentiment analysis results and the sentiment features in the sentiment interaction level table to determine at least one target sentiment feature that matches.

[0050] Understandably, based on the above S301, the emotional interaction level table transforms the unstructured emotional characteristics in the emotional analysis results into quantifiable numerical indicators, providing a unified quantitative basis for subsequent emotional trend modeling, emotional progression risk assessment, and interaction numerical calculation.

[0051] Optionally, the sentiment features in the sentiment analysis results can be matched with the sentiment features in the sentiment interaction level table to determine at least one matching target sentiment feature. This can be achieved through the following steps: At least one emotional feature to be matched is determined according to the priority of at least one emotional feature in the emotional interaction level table; the initial matching degree between at least one emotional feature in the sentiment analysis result and at least one emotional feature to be matched is calculated; the emotional feature to be matched with the initial matching degree greater than a set threshold is determined as the target emotional feature; or, the initial matching degree between at least one emotional feature in the sentiment analysis result and at least one emotional feature in the emotional interaction level table is calculated; the emotional feature with the final matching degree greater than a set threshold is determined as the target emotional feature, wherein the final matching degree is calculated based on the initial matching degree and the preset weight of at least one emotional feature in the emotional interaction level table.

[0052] Understandably, when multiple dimensions of sentiment features in a rating scale correspond to a single rating value, the priority of each dimension's sentiment features can be prioritized. A certain number of sentiment features to be matched are then determined according to this priority. These features are then matched against the sentiment information, and an initial matching degree is calculated. This means matching a subset of sentiment features based on priority. The initial matching degree can be understood as the similarity between the sentiment feature and the sentiment information. If the initial matching degree is greater than a set threshold, it is identified as the target sentiment feature, indicating that a sentiment feature has been matched against the sentiment information in this match. If the initial matching degree is less than or equal to the set threshold, the next sentiment feature to be matched is determined, meaning the sentiment features to be matched are determined in batches according to priority until a match is found. Alternatively, the sentiment information can be matched against the sentiment features of each dimension, calculating the initial matching degree. The initial matching degree refers to the degree of matching between a sentiment information and a sentiment feature of a dimension. Features with high initial matching degrees exceeding the set threshold can then be identified as target sentiment features. Alternatively, the final matching degree can be calculated based on the preset weights of the sentiment features of each dimension, and features with high final matching degrees exceeding the set threshold can then be identified as target sentiment features. Other methods for determining the target's emotional characteristics are not limited.

[0053] S303. Based on the temporal relationship of at least one interactive plot text in the script text, construct the target emotion line trend based on the level value corresponding to at least one target emotion feature.

[0054] Among them, the target emotion line trend represents the changing trend of the emotional intensity between multiple target characters as the plot progresses.

[0055] Understandably, based on the above S302, the level value (also called interaction value) corresponding to the target emotional feature is determined in the emotional interaction level table. This involves quantifying the emotional features / information in the emotional analysis results. Quantification refers to converting non-numerical emotional features (such as expressing affection) into calculable numerical indicators (such as emotional polarity quantification value, emotional intensity value, interaction value, or level value). The interaction value is used to represent the numerical result of the emotional degree between characters, measuring emotional tension, relationship intimacy, or plot progression. For example, a direct expression of affection between characters is mapped to +5 (i.e., a positive interaction value of 5), indicating that the characters admire each other and have a high degree of affection. Subsequently, the temporal relationship of the interactive plot text in the script text is analyzed, that is, the order in which the interactive plots occur, such as the chronological relationship of the characters' first encounter, confession, and falling in love. Then, an emotional line trend is constructed based on the interaction value and combined with the interactive plot text to generate the emotional line analysis result. The emotional line analysis result can be understood as a structured emotional relationship evolution report, including information such as emotional trends, key nodes, and relationship changes. Subsequent analysis based on the emotional storyline results will be used to optimize the script text, predict future relationships between characters, and make market forecasts. For example, key emotional events will be added to enhance audience resonance, and the correlation between the strength of the emotional storyline and viewership ratings will be analyzed to optimize script development strategies.

[0056] Understandably, before generating the emotional line analysis results, an emotional line analysis table can be constructed based on the emotional lines between characters, specific plot points, and interaction values, as shown in Table 2.

[0057] Specifically, the basic components of an emotional storyline include the initial relationship, emotional trigger point, emotional change / development, emotional climax, and final emotion. The initial relationship refers to the character's emotional state and background at the beginning of the story. In romance dramas, it's particularly important to focus on the character's attitude towards love or romantic relationships. This is usually reflected in the character's biography and may also be revealed as the plot unfolds (e.g., flashbacks). The initial relationship specifically includes the character's background and initial emotional state. Background can include whether the character is single or has experienced emotional trauma, while the initial emotional state can be the character's current emotional / psychological state, such as sadness, loneliness, or a loss of faith in love. The emotional trigger point refers to an event or situation at the beginning of the story that triggers the character's initial emotional change. In romance dramas, this is specifically manifested as the first encounter between the male and female leads. This encounter or event alters the characters' original emotional state, such as igniting their attraction or interest in each other. Emotional development refers to the gradual emotional changes that occur as the character faces various situations. In romance dramas, this manifests in several ways: establishing connection, emotional conflict, and deepening the emotional bond. Establishing connection refers to the characters gradually getting to know each other through interaction, building an initial emotional connection, such as dating or participating in shared activities. Emotional conflict refers to minor misunderstandings or conflicts that test the depth of the characters' feelings and their ability to resolve conflicts. Deepening the emotional bond refers to the gradual deepening of mutual understanding and dependence, with shared experiences solidifying the characters' relationship. The climax refers to a crucial moment in the story where the characters' emotions reach their peak, usually a turning point or a critical decision. In romance dramas, this typically manifests as a major conflict or emotional crisis between the male and female leads, leading to a temporary separation or severe emotional fluctuations. At this moment, the characters' emotions reach their peak, forcing them to confront real emotional issues and make crucial choices, such as reconciliation or confession. The ending refers to the characters' emotional state at the end of the story, showing growth or change compared to their initial state. In romance dramas, this manifests in several ways: resolving conflicts, mutual understanding, and emotional growth. Conflict resolution refers to the characters overcoming their major emotional troubles, reunifying and establishing a deeper emotional connection. Mutual understanding refers to a more thorough understanding of each other and making commitments for the future, such as deciding to spend the rest of their lives together or getting married. Emotional growth refers to the characters experiencing significant emotional growth and change, and gaining a deeper understanding of the meaning of love and being loved.

[0058] Table 2:

[0059] As is understandable, Table 2 shows the direction of the characters' emotional lines, specific plot points, and interaction values ​​at different emotional stages in the emotional storyline. The interaction values ​​can also be understood as the degree of favorability. In Table 2, the interactive plots from different characters' perspectives may not correspond, and the favorability represented by these plots from different characters' perspectives will also differ. That is, different actions have different degrees of impact on different characters. For example, the interactive plot in Table 2, "Character A, to protect Character B, drugged him and sent him away," reflects Character A's favorability towards Character B as 1, meaning Character A has favorability towards Character B and would let him leave to protect him. However, the interactive plot in Table 2, "Character B, upon learning that Character A decided to send him away to protect him, was deeply moved and, while Character A was asleep, placed a bracelet formed from his own heart's blood on Character A's wrist to protect him," reflects Character B's favorability towards Character A as 3. That is, the favorability towards the other person differs depending on the perspective of the same event; Character B, as the beneficiary, has a higher degree of favorability towards Character A's action.

[0060] Understandably, Table 2 maps a mutual assistance value to each specific plot point, and there is no cumulative interaction value between plots. In other words, the interaction plot "Character A, to protect Character B, drug him and then sends him away" reflects Character A's favorability towards Character B as 1. There is no situation where the favorability was previously 0.5 and then increased by 0.5 in this interaction plot to reach 1. That is, when mapping interaction plots to interaction values, the contextual plot information is not considered; only the interaction plot itself needs to be analyzed.

[0061] Understandably, the cumulative interaction value belonging to the same emotional stage can also be calculated based on user needs. For example, in the initial relationship stage, the emotional line from Character A to Character B may have other interactive events besides "meeting Character B at location C, kissing and then switching bodies, and being forced by Character B to break the seal at location C." If the interaction value for each of these other interactive events is 1, the sum of the two interaction values ​​will result in an actual interaction value of 0 (i.e., 1-1=0). Here, an interaction value of -1 represents a negative emotional line, while an interaction value of 1 or +1 represents a positive emotional line. The calculation method for other possible interaction values ​​is not limited.

[0062] Optionally, the method also includes: Based on the temporal relationship, an emotional trend chart is constructed for interactive values ​​and interactive plot points. Interactive plot points are used to identify interactive plot texts and / or the episodes corresponding to interactive plot texts in the script text.

[0063] Understandably, the order in which interactive plot texts occur within the series is determined. For example, interactive plot text 'a' describes the first interactive action in the first episode, such as "Character A and Character B meet at location C," which represents the initial relationship stage. This order can be denoted as 1-1, signifying the first interactive action in the first episode. Based on this rule, the order of other interactive plot texts is determined. After determining the order of each interactive plot text, the interaction values ​​and corresponding plot points are sorted according to this order to generate an emotional trend chart. The emotional trend chart visually reflects the trend of the characters' emotional lines, that is, the emotional driving force. One interactive plot text can be recorded as one emotional plot point. The horizontal axis of the emotional trend chart represents the key plot points of effective interaction, i.e., the interactive plot points, while the vertical axis represents the interaction values, used to characterize emotional tendency and intensity. Specifically, as shown... Figure 2 As shown, Figure 2 An emotion trend chart provided in this embodiment of the disclosure is constructed based on Table 2 above. Figure 2 This includes the first emotional curve of Character A's feelings for Character B (from Character A's perspective) and the second emotional curve of Character B's feelings for Character A (from Character B's perspective). Based on these two emotional curves, we can determine the changes and fluctuations in Character A's feelings for Character B, the changes and fluctuations in Character B's feelings for Character A, the pace of the emotional progression between Characters A and B, and the major emotional interaction nodes between Characters A and B. The emotional nodes on the horizontal axis (1-4) are ordered in chronological order. For example, when Characters A and B first meet, the emotional intensity / interaction value is 0; as various interactive events occur subsequently, the interaction value between Characters A and B changes accordingly. Understandably, the emotional trend chart can also display specific interactive event text, the episode to which the interactive event belongs, and other information.

[0064] S304. Based on the trend of the target emotional line and at least one interactive plot text, the risk of the emotional progression process between multiple target characters is quantified by setting risk assessment indicators to generate an emotional quality assessment result of the script text.

[0065] Understandably, based on the aforementioned S303, by comprehensively considering the interactive plot text and emotional trends through the established risk assessment indicators, the emotional progression process between multiple target characters is quantified to generate an emotional quality assessment result for the script text. This result includes a script emotional overview, an episode emotional overview, emotional progression risk points, and / or emotional progression suggestions addressing these risk points. The script emotional overview describes the overall emotional trajectory and core emotional themes among multiple characters in the script. The episode emotional overview summarizes the emotions of a specific episode, reflecting the main characteristics of emotional changes between characters in that episode. Each episode may have its own emotional overview, or only some episodes may have one; the specifics can be determined based on user needs. Emotional progression risk points refer to potential problems in the current script's emotional development (e.g., excessively fast pacing, abrupt transitions, lack of tension). Emotional progression suggestions provide specific improvement recommendations for identified risk points (e.g., adding foreshadowing, strengthening conflict, delaying the climax).

[0066] Optionally, based on the trend of the target emotional line and at least one interactive plot text, the risk of the emotional progression process between multiple target characters is quantified using a set risk assessment index to generate an emotional quality assessment result for the script text. This can be achieved through the following steps: The process involves: 1) Counting the first number of emotional interaction events between multiple target characters in at least one interactive plot text; 2) Comparing this first number with a set interaction threshold to analyze the emotional interaction density and generate emotional interaction results; 3) Counting the second number of emotional turning points between multiple target characters due to external forces based on at least one interactive plot text; 4) Comparing this second number with a set turning point threshold to analyze the naturalness of the plot setting and generate plot setting results; 5) Calculating the emotional progression rate based on the target emotional line trend; 6) Comparing the emotional progression rate with a set progression rate threshold to analyze the emotional progression rhythm and generate emotional progression results; 7) Analyzing the consistency of the emotional line trend between the target emotional line trend and the reference emotional line trend and generating emotional progression results; and 8) Integrating the emotional interaction results, plot setting results, and emotional progression results to generate an emotional quality assessment result for the script text.

[0067] Understandably, the first count of emotional interaction events between target characters in interactive plot text is statistically analyzed and compared with a preset interaction threshold. This generates emotional interaction results by calculating the frequency of interaction between characters within a unit of text, objectively quantifying the density of character relationships and avoiding a plot that appears distant due to too little interaction or cumbersome due to too much. The second count of emotional turning points between target characters driven by external factors, such as unexpected events or third-party intervention, is analyzed and compared with a turning point threshold to generate plot setting results. This identifies whether the script relies excessively on coincidence or forced setups to drive the emotional arc, focusing more on emotional changes driven by internal factors between characters, thereby assessing the naturalness and internal logic of plot development. Subsequently, the rate of change of emotional state per unit time, i.e., the emotional progression rate, is calculated based on the emotional trajectory trend. This rate is then compared with a set progression rate threshold to generate the emotional progression result. Alternatively, the target emotional trajectory trend can be compared with a benchmark emotional trajectory trend (such as the emotional trajectory trend of a classic script) to generate the emotional progression result. This allows for precise capture of the plot's pacing, and by comparing it with successful examples, the rationality and market acceptance of the trend can be evaluated. Finally, the results from emotional interaction, plot setting, and emotional progression are integrated to generate a comprehensive emotional quality assessment report. This provides screenwriters with an objective, multi-dimensional, and quantitative scientific analysis, significantly improving the objectivity, accuracy, and efficiency of the assessment. Analysis results for other possible aspects of emotional progression risk points are not limited and can be determined according to user needs.

[0068] Understandably, emotional line comments can also be generated directly based on at least one interactive plot text and / or the emotional line analysis table shown in Table 2. Emotional line comments refer to a comprehensive evaluation text of the evolution of the emotional relationship between characters in the script. Then, the emotional line comments can guide the optimization or creative adjustment of the plot, and obtain the emotional quality assessment result of the script.

[0069] For example, the evaluation of the emotional arc from Character A to Character B in the script is as follows: Character A's feelings for Character B progress gradually from initial fear to dependence and then to a vague affection (i.e., the content of the script's emotional overview). In the first episode, Character A accidentally triggers Character B's resurrection and swaps bodies with him, with fear of his identity and power dominating. From the second to the fourth episode, through their daily interactions, Character B's domineering nature and occasional protectiveness gradually develop feelings for Character A. During this period, the connection of the Heart-Connecting Mantra becomes an important emotional catalyst. After Character A learns Character B's identity, he feels deceived, and his affection decreases. In the tenth episode, when Character B takes Character A back to location D, his concern for Character A becomes more obvious (i.e., the content of the episode's emotional overview). The evaluation result of the emotional quality of Character A to Character B in the script is as follows: The risk point of the emotional progress in the first ten episodes lies in the somewhat abrupt setting of the Heart-Connecting Mantra. Character A's emotional transformation partly depends on this setting, weakening the character's own agency (i.e., the content of the risk point in the emotional arc progression). While it aligns with the setting logic of fantasy dramas, more details are needed to show character A's emotional growth under the influence of character B, and to avoid the emotional transformation appearing too passive (i.e., suggestions for advancing the emotional storyline).

[0070] For example, the evaluation of the emotional line between Character B and Character A in the script is as follows: In the first episode, Character B's feelings for Character A begin with using the Heart-to-Heart Mantra as a medium to see her as a tool to revive the karmic fire. In the second episode, the attempt to remove the spell fails. As Character A makes sacrifices and the Heart-to-Heart Mantra takes effect, Character B begins to pay attention to Character A's emotions, such as whether she is happy. In the third episode, to protect Character A, Character B forcibly shakes the Soul-Transforming Cauldron, breaking the principle of prioritizing self-interest. The emotional drive in the first three episodes still relies on fantasy mechanisms such as body swapping and spell backlash. In the tenth episode, Character B's action of taking Character A back to location D reveals a hint of concern for Character A. The evaluation result of the emotional quality of Character B's emotional line between Character B and Character A in the script is as follows: The risk point of the emotional development in the first ten episodes lies in the fact that Character B's emotional transformation is relatively subtle, lacking sufficient emotional groundwork and external expression. Character B's character setting is a cold and domineering demon lord, but the overly restrained emotional expression makes it difficult for the audience to perceive his inner changes, thus weakening the persuasiveness of the emotional line. The subsequent plot needs more details to show the emotional changes of character B, such as through eye contact, micro-expressions, and behavioral contrasts, to make the gradually warming inner world beneath her cold exterior more clearly visible (i.e., the content of the emotional line development suggestion).

[0071] For example, one can analyze the risks of developing the emotional storyline not only for all episodes together, but also for each episode individually. For instance, regarding the first three episodes: In the first episode, the body-swapping between characters A and B through a kiss is too abrupt and risks viewers abandoning the show. In the second episode, the connection between characters A and B through a love spell is revealed, which seems too coincidental. Furthermore, in the thirtieth episode, there is little emotional interaction between characters A and B, mainly focusing on character A learning of character B's hidden secret. In the thirty-first episode, excessive screen time for supporting characters compresses the progress of the emotional storyline between characters A and B, risking a slow pace and viewers abandoning the show.

[0072] This disclosure provides a script quality assessment method that quantifies the emotional connections, interaction characteristics, power relationships, and expression methods between characters using an emotional interaction level table to determine corresponding interaction values. This achieves a structured representation of complex emotional relationships in a script, improves the accuracy and objectivity of emotional line analysis, and provides a quantifiable technical basis for character relationship modeling and script evaluation.

[0073] Based on the above embodiments, Figure 4 This is a flowchart illustrating a character emotion line analysis method provided in this embodiment of the disclosure. The processing flow of the character emotion line analysis system includes: 1) Obtain the original script; 2) Extract characters from the original script; 3) Filter out effective interactive plots of the characters from the original script; 4) Quantify the emotional relationship reflected by the effective interaction and determine the interaction value; 5) Draw an emotional curve based on the interaction value and effective interactive plots.

[0074] Figure 5 This is a schematic diagram of a script quality assessment device provided in an embodiment of the present disclosure. The script quality assessment device provided in this embodiment can execute the processing flow provided in the script quality assessment method embodiment, such as… Figure 5 As shown, the script quality assessment device 500 includes: Acquisition unit 501 is used to acquire the script text and determine the multiple target characters included in the script text; The text generation unit 502 is used to determine at least one interactive plot text of multiple target characters recorded in the script text; The sentiment analysis unit 503 is used to analyze the emotional relationships of multiple target characters based on at least one interactive plot text using a pre-trained sentiment analysis model, and obtain sentiment analysis results. The sentiment analysis results refer to the semantic content represented by the interactive plot text that reflects the emotional characteristics between the characters. The result generation unit 504 is used to quantify the risk of the emotional progression process between multiple target characters based on the sentiment analysis results and at least one interactive plot text, through a set risk assessment index, so as to generate the sentiment quality assessment result of the script text.

[0075] Optionally, the acquisition unit 501 is used for: Extract multiple characters from the script text; Based on the attribute information of multiple characters in the script text, multiple target characters are identified, where the attribute information includes the number of times the character appears in the script text and / or the number of descriptive paragraphs; or, Obtain the character description statements to be analyzed for sentiment, and identify multiple target characters from multiple script characters based on the character description statements.

[0076] Optionally, the text generation unit 502 is used for: The script text identifies at least one interactive plot text describing the plurality of target characters, including: Select at least one interactive plot text from the script text in which multiple target characters participate in the interaction; The system analyzes at least one interactive plot text to determine whether it contains effective interactive plots using a pre-trained large language model, and outputs interactive plot texts containing effective interactive plots. Effective interactive plots refer to changes in the emotional relationships between multiple target characters at interactive plot nodes.

[0077] Optionally, the text generation unit 502 is used for: Summarize the text content of at least one interactive plot text to generate a summarized interactive plot text. The system analyzes at least one interactive plot text after generalization using a pre-trained large language model to determine whether it contains valid interactive plots, and outputs the interactive plot text containing valid interactive plots.

[0078] Optionally, the result generation unit 504 is used for: Obtain a pre-constructed emotional interaction level table, which is used to represent the correspondence between emotional characteristics and level values; Match the sentiment features in the sentiment analysis results with the sentiment features in the sentiment interaction level table to identify at least one target sentiment feature that matches. Based on the temporal relationship of at least one interactive plot text in the script text, a target emotion line trend is constructed based on the level value corresponding to at least one target emotion feature. The target emotion line trend represents the changing trend of the emotional intensity between multiple target characters as the plot progresses. Based on the trend of the target emotional line and at least one interactive plot text, the risk of the emotional progression process between multiple target characters is quantified by setting risk assessment indicators in order to generate an emotional quality assessment result of the script text.

[0079] The emotional characteristics in the emotional interaction level table include at least one of the following: emotional descriptive words, interaction characteristics, power characteristics, and expression characteristics. Among them, emotional connection refers to words that describe the emotional connection state between characters, interaction characteristics are used to characterize the emotional relationship and interaction pattern when characters interact, power characteristics are used to characterize the power relationship between characters, and expression characteristics are used to characterize the expression form of emotions between characters.

[0080] Optionally, the result generation unit 504 is used for: At least one emotional feature to be matched is determined according to the priority of at least one emotional feature in the emotional interaction level table; the initial matching degree between at least one emotional feature and at least one emotional feature to be matched in the sentiment analysis results is calculated; the emotional features to be matched with an initial matching degree greater than a set threshold are determined as target emotional features; or, Calculate the initial matching degree between at least one sentiment feature in the sentiment analysis results and at least one sentiment feature in the sentiment interaction level table; determine the sentiment feature with a final matching degree greater than a set threshold as the target sentiment feature, wherein the final matching degree is calculated based on the initial matching degree and the preset weight of at least one sentiment feature in the sentiment interaction level table.

[0081] Optionally, the result generation unit 504 is used for: The first number of emotional interaction events between multiple target characters in at least one interactive plot text is counted, and the first number is compared with a set interaction threshold to analyze the emotional interaction density and generate emotional interaction results. Based on at least one interactive plot text, analyze the second number of emotional turning points between multiple target characters due to external driving forces, compare the second number with a set turning point threshold, analyze the naturalness of the plot setting, and generate plot setting results; The emotional progression rate is calculated based on the target emotional line trend, and the emotional progression rate is compared with a set progression rate threshold to analyze the emotional progression rhythm and generate emotional progression results; and / or, the consistency of the emotional line trend between the target emotional line trend and the reference emotional line trend is analyzed to generate emotional progression results. By integrating the results of emotional interaction, plot setting, and emotional progression, an emotional quality assessment of the script text is generated.

[0082] The script includes multiple roles, including a first role and a second role.

[0083] Optionally, the text generation unit 502 is used for: Based on the script text, generate a first interactive text that describes the interaction between the first character and the second character from the perspective of the first character. Based on the script text, generate a second interactive text that describes the interaction between the second character and the first character from the perspective of the second character. The interactive plot text includes the first interactive text and the second interactive text.

[0084] Figure 5 The script quality assessment device shown in the embodiment can be used to execute the technical solution of the above method embodiment. Its implementation principle and technical effect are similar, and will not be repeated here.

[0085] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. See below for details. Figure 6 The diagram illustrates a structural schematic suitable for implementing the electronic device 600 in the embodiments of this disclosure. The electronic device 600 in the embodiments of this disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), wearable electronic devices, etc., as well as fixed terminals such as digital TVs, desktop computers, smart home devices, etc. Figure 6 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0086] like Figure 6 As shown, electronic device 600 may include processing device 601 (e.g., central processing unit, graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 602 or a program loaded from storage device 608 into random access memory (RAM) 603 to implement the script quality assessment method as described in the embodiments of this disclosure. Various programs and data required for the operation of electronic device 600 are also stored in RAM 603. Processing device 601, ROM 602, and RAM 603 are interconnected via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0087] Typically, the following devices can be connected to I / O interface 605: input devices 606 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 607 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 608 including, for example, magnetic tapes, hard disks, etc.; and communication devices 609. Communication device 609 allows electronic device 600 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 6 An electronic device 600 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0088] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts, thereby implementing the script quality assessment method described above. In such embodiments, the computer program can be downloaded and installed from a network via communication device 609, or installed from storage device 608, or installed from ROM 602. When the computer program is executed by processing device 601, it performs the functions defined above in the methods of embodiments of this disclosure.

[0089] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0090] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0091] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0092] Optionally, when one or more of the above-described procedures are executed by the electronic device, the electronic device may also execute other steps of the above embodiments.

[0093] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed 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 can 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 can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0094] 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 this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated 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, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0095] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0096] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0097] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0098] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or gateway that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or gateway. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or gateway that includes the element.

[0099] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method of script quality assessment, characterized by, The method comprises the following steps: obtaining a script text and determining a plurality of target roles included in the script text; determining at least one interactive plot text recording the plurality of target roles in the script text; analyzing the emotional relationship of the plurality of target roles based on the at least one interactive plot text by a pre-trained emotion analysis model to obtain an emotion analysis result, wherein the emotion analysis result is the semantic content reflecting the emotional characteristics between the roles represented by the interactive plot text; quantifying the emotional promotion process between the plurality of target roles by a set of risk evaluation indexes according to the emotion analysis result and the at least one interactive plot text to generate an emotional quality evaluation result of the script text.

2. The method of claim 1, wherein, The determination of the plurality of target roles included in the script text comprises: extracting a plurality of script roles from the script text; determining a plurality of target roles according to the attribute information of the plurality of script roles in the script text, wherein the attribute information includes the number of appearances and / or the number of description paragraphs of the roles in the script text; or obtaining a role description sentence to be analyzed and determining the plurality of target roles from the plurality of script roles according to the role description sentence.

3. The method of claim 1, wherein, The determination of at least one interactive plot text recording the plurality of target roles in the script text comprises: screening at least one interactive plot text in which the plurality of target roles participate from the script text; analyzing whether the at least one interactive plot text contains an effective interactive plot by a pre-trained large language model and outputting an interactive plot text containing an effective interactive plot, wherein the effective interactive plot refers to a change in the emotional relationship between the plurality of target roles at an interactive plot node.

4. The method of claim 3, wherein, The analysis of whether the at least one interactive plot text contains an effective interactive plot by a pre-trained large language model and the output of an interactive plot text containing an effective interactive plot comprise: summarizing the text content of the at least one interactive plot text to generate a summarized at least one interactive plot text; analyzing whether the summarized at least one interactive plot text contains an effective interactive plot by a pre-trained large language model and outputting an interactive plot text containing an effective interactive plot.

5. The method of claim 1, wherein, The quantification of the emotional promotion process between the plurality of target roles by a set of risk evaluation indexes according to the emotion analysis result and the at least one interactive plot text to generate an emotional quality evaluation result of the script text comprises: obtaining a pre-constructed emotional interaction level table, wherein the emotional interaction level table is used to represent the corresponding relationship between emotional characteristics and level values; matching and analyzing the emotional characteristics in the emotion analysis result and the emotional characteristics in the emotional interaction level table to determine at least one target emotional characteristic that matches; construct a target emotion line trend based on the grade value corresponding to the at least one target emotion feature according to the time sequence relationship of the at least one interactive plot text in the script text, wherein the target emotion line trend represents a change trend of the emotional intensity between the target characters with the plot advancing; quantify the emotional advancing process between the target characters by using a set of risk evaluation indexes according to the target emotion line trend and the at least one interactive plot text, so as to generate the emotional quality evaluation result of the script text.

6. The method of claim 5, wherein, The emotion features in the emotion interaction grade table include at least one of an emotion descriptor, an interaction feature, a power feature and an expression feature, wherein the emotion descriptor is a word describing the emotional connection state between characters, the interaction feature is used to represent the emotional relationship and interaction mode when the characters interact with each other, the power feature is used to represent the power relationship between the characters, and the expression feature is used to represent the expression form of the emotional connection between the characters. The matching analysis of the emotion features in the emotion analysis result and the emotion features in the emotion interaction grade table to determine at least one target emotion feature includes: determining at least one to-be-matched emotion feature according to the priority of at least one emotion feature in the emotion interaction grade table; calculating the initial matching degree between at least one emotion feature in the emotion analysis result and the at least one to-be-matched emotion feature; determining the to-be-matched emotion feature with the initial matching degree greater than a set threshold as the target emotion feature; or calculating the initial matching degree between at least one emotion feature in the emotion analysis result and at least one emotion feature in the emotion interaction grade table; determining the emotion feature with the final matching degree greater than the set threshold as the target emotion feature, wherein the final matching degree is calculated according to the initial matching degree and the preset weight of at least one emotion feature in the emotion interaction grade table.

7. The method of claim 5, wherein, The risk quantification of the emotional advancing process between the target characters by using a set of risk evaluation indexes according to the target emotion line trend and the at least one interactive plot text to generate the emotional quality evaluation result of the script text includes: statistically counting a first number of emotional interaction events between the target characters in the at least one interactive plot text, comparing the first number with a set interaction threshold, analyzing the emotional interaction density, and generating an emotional interaction result; analyzing a second number of emotional turning events between the target characters due to external driving according to the at least one interactive plot text, comparing the second number with a set turning threshold, analyzing the natural degree of plot setting, and generating a plot setting result; calculating an emotional advancing rate according to the target emotion line trend, comparing the emotional advancing rate with a set advancing rate threshold, analyzing the emotional advancing rhythm, and generating an emotional advancing result; and / or analyzing the consistency of the target emotion line trend with a reference emotion line trend, and generating an emotional advancing result; Integrate the emotional interaction result, the plot setting result and the emotional promotion result to generate the emotional quality evaluation result of the script text.

8. The method of claim 1, wherein, The multiple target roles include a first role and a second role, and the at least one interaction plot text recording the multiple target roles in the script text includes: Based on the script text, generate a first interaction text describing the interaction content between the first role and the second role from the perspective of the first role; Based on the script text, generate a second interaction text describing the interaction content between the first role and the second role from the perspective of the second role; The interaction plot text includes the first interaction text and the second interaction text.

9. A script quality evaluation device characterized by comprising: It includes: An acquisition unit configured to acquire a script text and determine multiple target roles included in the script text; A text generation unit configured to determine at least one interaction plot text recording the multiple target roles in the script text; An emotion analysis unit configured to analyze emotional relationships of the multiple target roles based on the at least one interaction plot text through a pre-trained emotion analysis model to obtain an emotion analysis result, wherein the emotion analysis result refers to semantic content reflecting emotional characteristics between characters represented by the interaction plot text; A result generation unit configured to perform risk quantification on an emotional promotion process between the multiple target roles through a set risk evaluation index according to the emotion analysis result and the at least one interaction plot text to generate an emotional quality evaluation result of the script text.

10. An electronic device, comprising: It includes: A memory; A processor; And A computer program; The computer program is stored in the memory and is configured to be executed by the processor to implement the script quality evaluation method according to any one of claims 1 to 8.

11. A computer readable 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 script quality evaluation method according to any one of claims 1 to 8.

Images