An AI model-based fan character assignment management system

Abstract

The application relates to the technical field of fan creation, and specifically discloses a fan character allocation management system based on an AI model, which comprises a character feature acquisition module, a user information input module and an AI model construction end. The character feature acquisition module is used for extracting fan character related data from text and image data. The user information input module is used for inputting personal information, plot information and historical creation data by a creator. The AI model construction end comprises a plot construction module and a character matching module. The plot construction module is used for generating appropriate story lines according to various fan character features and plot information. Through cooperation of multiple modules, the application guarantees that a final scheme can meet creation requirements, ensures that the scheme is consistent with actual creation expectations, optimizes matching results, re-evaluates the matching degree of characters and creators, realizes accurate allocation, and simultaneously achieves the purposes of improving work quality, perfectly matching characters and creators, and outputting high-quality fan works.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of fan works, and particularly relates to a fan character allocation management system based on an AI model. BACKGROUND

[0002] In the field of fan works, creators usually need to select appropriate characters from a large number of fan characters and allocate the characters to different creative tasks to construct rich and varied plots.

[0003] The traditional character allocation mode usually depends on the subjective judgment and experience of creators, and has many disadvantages. On the one hand, manual character allocation cannot comprehensively consider the complex characteristics of a large number of fan characters and the personal advantages and creative styles of creators, resulting in low matching degree of characters and creators, which affects the quality of works. On the other hand, with the continuous expansion of the scale of fan works, the number of characters and the demand for creation are increasing, and the efficiency of manual character allocation is extremely low, which cannot meet the demand for rapid creation. SUMMARY

[0004] The application aims to provide a fan character allocation management system based on an AI model, and solve the following technical problems:

[0005] How to realize accurate matching of fan characters and creators in an intelligent manner.

[0006] The application can be realized by the following technical solutions.

[0007] A fan character allocation management system based on an AI model, the system comprising:

[0008] A character feature collection module for extracting fan character related data from text and image data;

[0009] A user information input module for creators to input personal information, plot information and historical creation data;

[0010] An AI model construction end comprising a plot construction module and a character matching module, the plot construction module being configured to generate appropriate storylines according to the characteristics of each fan character and the plot information, and the character matching module being configured to generate the corresponding matching degree between each fan character and each creator according to the personal information and historical creation data input by the creator, and generate a preliminary character allocation scheme by comprehensively analyzing the generated storylines;

[0011] A feedback input module for creators to input character allocation feedback and adjust the storylines;

[0012] An optimization decision module for regenerating an optimized final character allocation scheme according to the character allocation feedback and the adjusted storylines input by the creators.

[0013] Further, the process of extracting the data related to the fan character includes:

[0014] Using natural language processing technology to analyze the description of the appearance, personality, background story, and special skills of the character in the text, and convert it into structured data features;

[0015] Using image recognition technology to identify the appearance characteristics, clothing features and scene information of the character in the image, and to mine the performance characteristics of the character in a specific scene.

[0016] Further, the process of generating a suitable story line includes:

[0017] Analyzing each character in the plot information, extracting the characteristics of each character, and calculating the similarity between the characteristics of each character and the fan character characteristics;

[0018] Selecting the fan character with the highest similarity to each character and generating a logically coherent story line through intelligent algorithms.

[0019] Further, the process of generating a preliminary role scheme by the role matching module includes:

[0020] Analyzing the personal information and historical creation data input by the creator, building a creator portrait, and mining the creation advantages and style characteristics of the creator;

[0021] According to the characteristics of each fan character used in the story line, calculate the corresponding matching degree between each fan character and each creator;

[0022] Combine the story line generated by the plot construction module, comprehensively analyze the importance and role of each fan character in the story, and generate a preliminary role allocation scheme.

[0023] Further, the process of generating a preliminary role allocation scheme includes:

[0024]

[0025]

[0026]

[0027] The creator i with the highest comprehensive matching degree with a fan character is calculated and analyzed by formulas (1)-(3);

[0028] Wherein, the feature vector of the creator portrait is The feature vector of the fan character is S jis the cosine similarity between the jth creator and the fan character, ω is the importance weight of the fan character in the story, is the role weight of the fan character in the story, there are N creators for the fan character to choose from, H j is the comprehensive matching degree between the jth creator and the fan character, arg max is the index of the maximum value.

[0029] Further, the process of the optimization decision module generating the optimized final role allocation scheme includes:

[0030] According to the adjusted story line feedback, the story line and the characteristics of each fan character are adjusted;

[0031] According to the creator input role allocation feedback, the matching degree between the fan character and the creator is re-evaluated;

[0032] According to the feedback adjusted story line and the re-evaluated matching degree between the fan character and the creator, a final role allocation scheme is generated.

[0033] Further, the system is deployed in a cloud server, and the creators access the system through a web browser and a creation client to input information.

[0034] Further, the system periodically collects text and image data from major fan creation platforms, extracts role characteristics after data cleaning, and stores them in a role characteristic database;

[0035] Through a data interface, the system is connected with the creation platform, automatically collects historical creation data of the creators, and stores them in a user information database.

[0036] The beneficial effects of the present application are:

[0037] (1) The present application realizes precise allocation by ensuring that the final scheme can meet the creation requirements, ensuring that the scheme meets the actual creation expectations, optimizing the matching results, re-evaluating the matching degree between the role and the creator, and achieving the purpose of improving the quality of works, perfectly matching the role and the creator, and outputting high-quality fan works. BRIEF DESCRIPTION OF DRAWINGS

[0038] The present application will be further described below with reference to the accompanying drawings.

[0039] Figure 1 is a schematic diagram of a fan role allocation management system based on an AI model. DETAILED DESCRIPTION

[0040] With reference to the drawings, the technical solutions in the embodiments of the present application will be clearly and completely described in the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of the present application.

[0041] Please refer to Figure 1 As shown in the drawings, in one embodiment, an AI model-based fan character allocation management system is provided, which comprises:

[0042] A character feature collection module is configured to extract fan character-related data from text and image data. For example, in popular fan fiction creation, through natural speech analysis processing technology, the description of character appearance (such as eye color, hairstyle characteristics), personality (whether cheerful and lively or calm and reserved), background story (growing experience, family origin), special skills (magic ability, combat skill), etc. in novels, forum posts and other texts can be deeply mined, and these unstructured text information can be ingeniously converted into structured data features for subsequent analysis and processing. At the same time, with the help of image recognition technology, fan illustrations, comics and other images are analyzed to identify the appearance characteristics (facial profile, body proportion) of the characters, the clothing characteristics (clothing style, accessory details) and the scene information (the environment they are in, the layout of the battle scene), so as to mine the performance characteristics of the characters in a specific situation.

[0043] A user information input module is configured to input personal information, plot information and historical creation data by the creator. The personal information input by the creator, such as creation preference and field of expertise, helps the system understand the basic situation of the creator. The plot information, including story theme, background setting and plot development, provides basic materials for the system to build story lines. The historical creation data reflects the past creation style and experience of the creator, such as the type of fan works created by the creator in the past, the characteristics of character shaping, etc.

[0044] The AI model construction end includes a story scenario construction module and a character matching module. The story scenario construction module is used to generate appropriate story lines according to various fan character features and story plot information. First, the module deeply analyzes each character involved in the story plot information and extracts its core features. Then, through complex calculation methods, the similarity of these character features and the fan character features collected from text and images is calculated and analyzed. For example, in a fan story set in a campus, the character in the story is designed as a class monitor with excellent grades and a gentle personality. The system will compare this character feature with the existing fan character features and select the fan character with the highest similarity to the character to implant into the story. Then, intelligent algorithms such as scenario logic-based reasoning algorithms are used to link the implanted fan characters to generate a logically coherent story line. The character matching module is used to generate the matching degree between each fan character and each author based on the author's input personal information and historical creation data, and to generate a preliminary role allocation scheme based on the generated story line analysis. First, the author's input personal information and historical creation data are deeply analyzed to build a comprehensive and accurate author portrait. For example, through data mining technology, the author's creation advantages (good at action description, emotional portrayal, etc.) and style characteristics (humorous, realistic, etc.) are mined. Then, according to the characteristics of each fan character in the story line, specific matching algorithms are used to calculate the corresponding matching degree between each fan character and each author. For example, for a character with a complex personality and rich inner play, the system will preferentially match authors who are good at emotional portrayal. Finally, combined with the story line generated by the story scenario construction module, the importance (main character, supporting character, etc.) and role (number of appearances, plot driving role) of each fan character in the story are analyzed to generate a preliminary role allocation scheme.

[0045] The feedback input module is used for authors to input role allocation feedback and adjust the story line. This module allows authors to input role allocation feedback, such as thinking that a certain character does not match their creative style, or having different opinions on the development of certain plot in the story line, thereby adjusting the story line to make the creation more consistent with the author's expectations.

[0046] The optimization decision module generates an optimized final role allocation scheme based on the author's input role allocation feedback and the adjusted story line. First, the story line and each fan character feature are adjusted based on feedback to ensure that the story line and character features meet the latest needs of the author. Then, the matching degree between the fan characters and the authors is re-evaluated, taking into account the changes in the importance and role of the characters after the story line adjustment, as well as the new requirements raised in the author's feedback. Finally, based on the feedback-adjusted story line and the re-evaluated matching degree between the fan characters and the authors, a final role allocation scheme is generated to achieve the optimization of fan character allocation.

[0047] The process of extracting character-related data includes:

[0048] Using natural language processing technology to analyze the appearance, personality, background story, and special skills of characters in the text, and converting them into structured data features. First, the text is preprocessed, such as word segmentation and part-of-speech tagging, to convert it into a form that computers can understand. Then, through semantic analysis technology, the meaning of each word and sentence in the text is understood, and descriptions related to the appearance, personality, background story, and special skills of the character are identified. For example, for the text description "he has long black hair, a straightforward personality, was born in a mysterious magic family, and has the special skill of controlling fire", natural language processing technology can accurately extract the appearance characteristics (long black hair), personality characteristics (straightforward), background story (born in a mysterious magic family), and special skills (controlling fire) of the character, and convert these characteristics into structured data, such as storing them in the form of key-value pairs: { "appearance": "long black hair", "personality": "straightforward", "background story": "born in a mysterious magic family", "special skills": "controlling fire"};

[0049] Using image recognition technology to identify the appearance characteristics, clothing features, and scene information of characters in images, and to mine the performance characteristics of characters in specific situations. First, image preprocessing is performed, such as grayscale and noise reduction, to improve image quality. Then, the CNN model automatically learns the features in the image through multiple layers of convolution and pooling operations, and identifies the appearance characteristics of the character, such as the shape and proportion of facial features, skin color, etc. At the same time, it can identify the clothing features of the character, such as the style, color, and material of the clothing. In addition, through analysis of the background and scene elements of the image, the scene information of the character can be mined, such as whether it is in a bustling urban street or a quiet forest. Furthermore, through analysis of the posture and expression of the character in the image, the performance characteristics of the character in specific situations can be mined, such as whether the character is in a combat state or a leisure state, and whether the expression is joyful or sad.

[0050] The process of generating appropriate storylines includes:

[0051] The characteristics of each character in the story plot information are analyzed, the similarity between the characteristics of each character and the characteristics of the fan characters is calculated, the key characteristics of each character are first analyzed and extracted, such as the character's personality traits, goal pursuit, interpersonal relationships, etc., and then the characteristics are compared with the fan character characteristics collected from the text and image data. The similarity calculation usually uses cosine similarity algorithm, which measures the similarity between two feature vectors by calculating the cosine value of the included angle between them. For example, assuming that the feature vector of character A in the story is [0.8, 0.6, 0.3, 0.1] (representing the quantified values of bravery, kindness, intelligence, and humor, respectively), and the feature vector of fan character B is [0.7, 0.7, 0.2, 0.2], the similarity between them can be calculated by the cosine similarity formula.

[0052] The fan character with the highest similarity to each character is selected and implanted, and a logically coherent story line is generated by intelligent algorithm, which takes into account the relationship between characters, the theme and goal of the story, etc. For example, in an adventure-themed story, different fan characters may have different skills and personalities, and the algorithm will arrange their interactions and plot development in the story according to these characteristics, such as having the brave character explore dangerous areas first, the intelligent character develop strategies, and the kind character rescue injured companions, etc., so as to build a logically reasonable and dramatic story line.

[0053] The process of generating a preliminary role scheme by the role matching module includes:

[0054] The creator's personal information and historical creation data are analyzed to build a creator portrait, and the creator's creative advantages and style characteristics are mined. Through data mining technology, the basic attributes of the creator are extracted from the personal information, such as age, gender, and years of experience, and the creative style of the creator is mined from the historical creation data, such as through language style analysis of the works to determine whether they are concise and clear or elegant and delicate. Through analysis of the themes of the works, the creator's preferred subject types are understood, such as science fiction, romance, and martial arts, and the creator's creative advantages are mined, such as being good at creating character images and building complex plots, etc. These information is integrated to build a portrait that fully reflects the characteristics of the creator;

[0055] Based on the characteristics of the various fan-created characters used in the storyline, a specific matching algorithm is used to calculate the matching degree between each fan-created character and each creator. The matching algorithm can comprehensively consider multiple factors, such as whether the creator's creative style matches the personality traits of the fan-created character, and whether the creator's creative strengths can fully showcase the charm of the fan-created character. For example, for a fan-created character with fantasy elements and a lively personality, it is more suitable to match a creator with an imaginative creative style who is good at describing vivid scenes and lively characters. Through quantitative analysis of these factors, the matching degree value between the two is obtained.

[0056] By combining the storylines generated in the story scenario construction module, a comprehensive analysis of the importance and role of each fan fiction character in the story is conducted to generate a preliminary character allocation plan. For fan fiction characters with high importance and a large role, priority is given to matching them with creators who have a high degree of compatibility with the character and strong creative ability. For fan fiction characters with low importance and a small role, they can be appropriately assigned to other creators. In this way, a preliminary character allocation plan is generated to ensure that each fan fiction character can be created by a suitable creator.

[0057] The process of generating the initial role allocation plan includes:

[0058]

[0059]

[0060]

[0061] The creator i with the highest overall matching degree with a certain fan fiction character is obtained by calculation and analysis using formulas (1)-(3);

[0062] The feature vector of the creator's portrait is: The feature vector of a fan-made character can be obtained from the personal information and historical creation data input by the creator. S is generated by extracting relevant features from text and image data and quantizing them. j Let ω be the cosine similarity between the j-th creator and the fanfiction character, and let ω be the importance weight of the fanfiction character in the story, determined based on the character's position in the storyline and their key role in advancing the plot; the higher the importance, the greater the weight. The weight of the fanfiction character's role in the story is determined based on factors such as the character's appearance frequency and plot length. The more role a character has, the greater its weight. μ is the creative ability evaluation factor for the j-th creator. There are N creators available for this fanfiction character to choose from, and H... jThe comprehensive matching degree of the jth creator and the co-protagonist is arg max, which is the index of the maximum value. That is, after calculating the comprehensive matching degree of all creators and a co-protagonist through the formula, the creator with the highest comprehensive matching degree is selected as the best allocation object of the co-protagonist.

[0063] Through the above technical solution, the embodiment provides a method for preliminary allocation of co-protagonists based on an AI model. For creators, the method can play their creative advantages, moderately expand their fields, and improve their creative confidence and ability. In terms of creative quality, the method can improve the matching degree of characters and creation, enhance the coherence and logic of the story, reduce the break-in time and optimize the process in terms of creative efficiency, and make the creation more efficient and orderly.

[0064] The process of generating the optimized final role allocation scheme by the optimization decision module includes:

[0065] According to the adjustment feedback of the creator on the story line, the system reconsiders the rationality and coherence of the story line, adjusts the story line, for example, if the creator proposes that a certain plot development in the story line is too abrupt, the system will re-plan the plot, adjust the actions and interaction methods of the relevant co-protagonists, and according to the feedback of the adjusted story line, adjust the feedback of the story line and the characteristics of each co-protagonist;

[0066] According to the role allocation feedback input by the creator, the matching degree between the co-protagonists and the creators is re-evaluated, considering the changes in the importance and roles of the characters after the adjustment of the story line, and the new requirements proposed by the creator in the feedback, such as the creator's desire to try a new creative style, the system will re-calculate the matching degree between each creator and the co-protagonist, and consider the new factors to modify the previous matching degree results;

[0067] According to the feedback-adjusted story line and the re-evaluated matching degree between the co-protagonists and the creators, the system again uses the optimization algorithm to generate the final role allocation scheme. In this process, the system will fully consider the overall needs of the story line, the characteristics of the characters, and the abilities and willingness of the creators, striving to achieve the best matching between the characters and the creators, and providing the best role allocation scheme for co-creation.

[0068] Through the above technical solution, the embodiment provides a method for final allocation of co-protagonists based on an AI model. The method can ensure that the final scheme can meet the creative needs, adjust the story line and the characteristics of the characters according to the feedback of the creators, ensure that the scheme meets the actual creative expectations, optimize the matching results, re-evaluate the matching degree between the characters and the creators, achieve precise allocation, and simultaneously achieve the purpose of improving the quality of the work, making the characters and the creators perfectly matched, and producing high-quality co-creation works.

[0069] The system is deployed in a cloud server, which can provide powerful computing resources and storage capacity to meet the system's processing needs for massive text and image data, and the storage needs for a large amount of author information and historical creation data. Authors can access the system through a web browser and a creation client to input information, which is not limited by region and device. As long as there is a network connection, authors can use the system to manage fan role allocation anytime and anywhere.

[0070] The system periodically collects text and image data from major fan creation platforms, which cover various types of fan works such as novels, comics, illustrations, etc. The collected data is cleaned to remove noise data, duplicate data, and other invalid information, and then the role characteristics are extracted using the technology in the role characteristic collection module and stored in the role characteristic database. The role characteristic database provides a rich data foundation for subsequent analysis and role allocation of the system, and the constantly updated and expanded database can better adapt to the development and changes in the fan creation field.

[0071] Through the data interface, the system is connected with the creation platform to automatically collect the historical creation data of the authors and store it in the user information database. The historical creation data includes the fan works published by the authors on various platforms, the related data of the works (such as reading volume, comment volume, etc.), and the interaction data between the authors and other users, etc. These data are stored in the user information database to provide important data support for building author portraits and role matching, which helps the system to better understand the creation situation and characteristics of the authors, and thus to achieve more accurate role allocation management.

[0072] The above describes one embodiment of the present application in detail, but the content is only a preferred embodiment of the present application and cannot be considered as limiting the scope of the present application. Any equivalent changes and improvements made within the scope of the present application should still belong to the patent coverage of the present application.

Claims

1. A fan-created character assignment management system based on an AI model, characterized in that, The system includes: The character feature acquisition module is used to extract fan-made character-related data from text and image data; The user information input module is used by creators to input personal information, story plot information, and historical creation data; The AI ​​model building module includes a story scenario building module and a character matching module. The story scenario building module is used to generate appropriate storylines based on the characteristics and plot information of each fan fiction character. The character matching module is used to generate the corresponding matching degree between each fan fiction character and each creator based on the personal information and historical creation data input by the creator, and to generate a preliminary character allocation plan by comprehensively analyzing the generated storylines. The feedback input module is used by creators to input feedback on character allocation and adjust the storyline; The decision-making module was optimized, and the optimized final character allocation scheme was regenerated based on the creator's feedback on character allocation and the adjusted storyline. The process of extracting data related to fan-created characters includes: Natural speech analysis technology is used to analyze descriptions of characters' appearance, personality, backstory, and special skills in texts and transform them into structured data features. Image recognition technology is used to identify the physical features, clothing characteristics, and scene information of characters in images, and to explore the performance characteristics of characters in specific situations. The process of generating an appropriate storyline includes: The characteristics of each character in the story plot are analyzed, and the similarity between the characteristics of each character and those of fan fiction characters is calculated and analyzed. Select fan-created characters that are most similar to each character and insert them into the story; and generate a logically coherent storyline using intelligent algorithms. The process by which the role matching module generates a preliminary role scheme includes: Analyze the personal information and historical creation data entered by creators to construct creator profiles and explore the creators' creative strengths and stylistic characteristics; Based on the characteristics of the various fan-created characters used in the storyline, calculate the matching degree between each fan-created character and each creator. Based on the storylines generated in the story scenario construction module, a comprehensive analysis of the importance and role of each fan fiction character in the story is conducted to generate a preliminary character allocation plan; The process of generating the initial role allocation plan includes: The creator i with the highest overall matching degree with a certain fan fiction character is obtained by calculation and analysis using formulas (1)-(3); The feature vector of the creator's portrait is: The feature vector of the fan-made character is S j Let μ be the cosine similarity between the j-th creator and the fan-created character. j Let ω be the evaluation factor for the creative ability of the j-th creator, and let ω be the importance weight of the fanfiction character in the story. To determine the screen time weight of this fan-created character in the story, there are N creators available for this fan-created character to choose from, and H... j Let arg max be the overall match between the j-th creator and the fan-created character, and let arg max be the index of the maximum value.

2. The fan-created character allocation and management system based on an AI model according to claim 1, characterized in that, The process by which the optimization decision module generates the optimized final role allocation scheme includes: Based on feedback from the revised storyline, adjustments were made to the storyline and the characteristics of each fan-created character. Based on the feedback from creators regarding character allocation, the match between fan-created characters and creators will be reassessed. Based on feedback, the revised storyline and the reassessed match between fan-created characters and creators were used to generate the final character allocation plan.

3. The fan-created character allocation and management system based on an AI model according to claim 2, characterized in that, The system is deployed on a cloud server, and creators access the system to input information through a web browser and a creation client.

4. The fan fiction role allocation management system based on an AI model according to claim 3, characterized in that, The system regularly collects text and image data from major fan fiction platforms, cleans the data, extracts character features, and stores them in a character feature database. It connects to the creation platform via a data interface, automatically collects creators' historical creation data, and stores it in the user information database.

Images