A method and device for evaluating narrative ability of a narrative subject

Abstract

An embodiment of the present invention provides a method and apparatus for evaluating the narrative ability of a narrative subject, comprising: obtaining multiple target media works of the narrative subject and a first score of each target media work on a media display platform; determining a second score for each target media work for each preset narrative dimension associated with the subject matter of the target media work based on a scoring model corresponding to each preset narrative dimension; mapping the first score of each target media work to a normalized scoring interval corresponding to the media display platform to obtain a third score for the target media work; determining a fourth score for the target media work based on the second score of each target media work for each preset narrative dimension and the third score of the target media work; and determining a narrative ability score for the narrative subject for each subject matter based on the fourth score of each target media work. This technical solution can improve the accuracy of narrative ability evaluation for narrative subjects.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to a method and device for evaluating the narrative ability of a narrative subject. Background Art

[0002] The creator of a media work is the narrative subject of the work, and the narrative ability of the narrative subject has a decisive influence on the quality of the media work. The assessment of the narrative subject's narrative ability plays a vital role in the early stages of project establishment and later evaluation of media projects.

[0003] Currently, the assessment of a subject's narrative ability is primarily based on the quantity and quality of their media works. However, this approach suffers from sample bias, leading to inaccurate assessments of their ability. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a method and apparatus for evaluating the narrative ability of a narrative subject, so as to improve the accuracy of the narrative ability evaluation of the narrative subject. The specific technical solution is as follows:

[0005] In a first aspect of the present invention, a method for evaluating the narrative ability of a narrative subject is provided, comprising:

[0006] Obtaining multiple target media works of a narrative object and a first rating of each target media work on a media display platform;

[0007] Determining, for each target media work, a second score for each preset narrative dimension associated with the subject matter of the target media work based on a scoring model corresponding to each preset narrative dimension;

[0008] Mapping the first score of each target media work to a normalized score interval corresponding to the media display platform to obtain a third score of the target media work;

[0009] determining a fourth score of each target media work based on the second score of each preset narrative dimension of each target media work and the third score of the target media work;

[0010] The narrative ability score of the narrative subject in each theme is determined according to the fourth score of each target media work in each theme.

[0011] In a possible implementation, the method further includes:

[0012] Obtaining subject matter related information of each target media work;

[0013] According to the theme association information of each target media work, the theme of the target media work is determined.

[0014] In a possible implementation, the subject matter-related information includes user portrait data and narrative technique information;

[0015] The step of determining the genre of each target media work based on the genre association information of the target media work includes:

[0016] Converting the user portrait data and narrative technique information of each target media work into a feature vector of the target media work;

[0017] A preset clustering algorithm is used to determine the vector class to which the characteristic vector of each target media work belongs. The subject matter corresponding to the vector class is the subject matter to which the target media work belongs.

[0018] In a possible implementation, the subject matter-related information includes a work label and plot summary information;

[0019] The step of determining the genre of each target media work based on the genre association information of the target media work includes:

[0020] Extract keywords for each genre from the work tags and plot synopsis information of each target media work;

[0021] For each target media work, determine the score and value of the keyword corresponding to the target media work in each theme as the theme score of the theme corresponding to the target media work;

[0022] For each target media work, the theme associated with the highest theme score corresponding to the target media work is determined as the theme to which the target media work belongs.

[0023] In one possible implementation, the step of determining, based on the scoring model corresponding to each preset narrative dimension, a second score for each target media work for each preset narrative dimension associated with the subject matter of the target media work includes:

[0024] Extracting, from the score association information associated with each target media work, a score parameter for each preset narrative dimension associated with the subject matter of the target media work;

[0025] The scoring parameters of each target media work in each preset narrative dimension are input into the scoring model corresponding to the preset narrative dimension to obtain a second score of the target media work in the preset narrative dimension.

[0026] In a possible implementation, the rating-related information includes image material and text material;

[0027] The scoring model for each preset narrative dimension associated with the image material is a convolutional neural network model;

[0028] The scoring model for each preset narrative dimension associated with the text material is an NLP (Neuro-Linguistic Programming, natural language processing) model.

[0029] In a possible implementation, the method further includes: for each preset narrative dimension, using the following steps to train a scoring model for the preset narrative dimension:

[0030] Acquiring training data, wherein the training data is score association information having annotated scores of the preset narrative dimension;

[0031] Extracting sample scoring parameters of the preset narrative dimension from the training data;

[0032] Inputting the sample scoring parameters of the preset narrative dimension into the scoring model corresponding to the preset narrative dimension to obtain a predicted score for the preset narrative dimension;

[0033] Determining a model loss for the preset narrative dimension based on the predicted score and the labeled score for the preset narrative dimension;

[0034] If it is determined that the scoring model corresponding to the preset narrative dimension has converged according to the model loss of the preset narrative dimension, then the training of the scoring model corresponding to the preset narrative dimension is terminated;

[0035] If it is determined based on the model loss of the preset narrative dimension that the scoring model corresponding to the preset narrative dimension has not converged, the parameters of the scoring model corresponding to the preset narrative dimension are adjusted, and the step of inputting the sample scoring parameters of the preset narrative dimension into the scoring model corresponding to the preset narrative dimension to obtain the predicted score of the preset narrative dimension is re-executed.

[0036] In one possible implementation, the step of determining the fourth score of each target media work based on the second score of each preset narrative dimension and the third score of the target media work includes:

[0037] The second score of each target media work in each preset narrative dimension and the third score of the target media work are weighted according to the weight coefficient of each preset narrative dimension and the weight coefficient of the media display platform to obtain a fourth score of the target media work; or

[0038] From the second score of the target media work in each preset narrative dimension and the third score of the target media work, the largest score is selected as the fourth score of the target media work.

[0039] In a possible implementation, determining the narrative ability score of the narrative subject in each theme based on the fourth score of each target media work in each theme includes:

[0040] For each theme, the fourth scores of each target media work of the theme are averaged to obtain the narrative ability score of the narrative subject on the theme; or

[0041] For each theme, the largest fourth score is selected from the fourth scores of each target media work on the theme as the narrative ability score of the narrative subject on the theme.

[0042] In a possible implementation, the method further includes:

[0043] Determine the target subject matter of the media work to be created;

[0044] Obtaining narrative ability scores of multiple candidate narrative subjects on the target subject matter;

[0045] The narrative object of the media work to be created is determined from candidate narrative objects whose narrative ability scores are greater than a preset score threshold.

[0046] In a second aspect of the present invention, there is also provided a device for evaluating the narrative ability of a narrative subject, comprising:

[0047] A first acquisition module is used to acquire multiple target media works of a narrative object and a first score of each target media work on a media display platform;

[0048] A first determination module is configured to determine, based on a rating model corresponding to each preset narrative dimension, a second rating of each target media work for each preset narrative dimension associated with the subject matter of the target media work;

[0049] a mapping module, configured to map the first score of each target media work to a normalized score interval corresponding to the media display platform to obtain a third score of the target media work;

[0050] A second determination module is configured to determine a fourth score of each target media work based on the second score of each preset narrative dimension of each target media work and the third score of the target media work;

[0051] The third determining module is configured to determine the narrative ability score of the narrative subject in each theme according to the fourth score of each target media work in each theme.

[0052] In a possible implementation, the apparatus further includes:

[0053] The second acquisition module is used to obtain the subject matter related information of each target media work;

[0054] The fourth determining module is configured to determine the genre of each target media work based on the genre association information of the target media work.

[0055] In a possible implementation, the subject matter-related information includes user portrait data and narrative technique information; and the fourth determination module is specifically configured to:

[0056] Converting the user portrait data and narrative technique information of each target media work into a feature vector of the target media work;

[0057] A preset clustering algorithm is used to determine the vector class to which the characteristic vector of each target media work belongs. The subject matter corresponding to the vector class is the subject matter to which the target media work belongs.

[0058] In a possible implementation, the subject matter-related information includes a work label and plot summary information; and the fourth determining module is specifically configured to:

[0059] Extract keywords for each genre from the work tags and plot synopsis information of each target media work;

[0060] For each target media work, determine the score and value of the keyword corresponding to the target media work in each theme as the theme score of the theme corresponding to the target media work;

[0061] For each target media work, the theme associated with the highest theme score corresponding to the target media work is determined as the theme to which the target media work belongs.

[0062] In a possible implementation, the first determining module is specifically configured to:

[0063] Extracting, from the score association information associated with each target media work, a score parameter for each preset narrative dimension associated with the subject matter of the target media work;

[0064] The scoring parameters of each target media work in each preset narrative dimension are input into the scoring model corresponding to the preset narrative dimension to obtain a second score of the target media work in the preset narrative dimension.

[0065] In one possible implementation, the scoring association information includes image materials and text materials; the scoring model for each preset narrative dimension associated with the image materials is a convolutional neural network model; and the scoring model for each preset narrative dimension associated with the text materials is an NLP model.

[0066] In a possible implementation, the apparatus further includes: a training module configured to train, for each preset narrative dimension, a scoring model for the preset narrative dimension, including:

[0067] A third acquisition submodule is configured to acquire training data, wherein the training data is score association information having annotated scores of the preset narrative dimension;

[0068] An extraction submodule, configured to extract sample scoring parameters of the preset narrative dimension from the training data;

[0069] An input submodule, configured to input the sample scoring parameters of the preset narrative dimension into a scoring model corresponding to the preset narrative dimension to obtain a predicted score for the preset narrative dimension;

[0070] The training submodule is used to determine the model loss of the preset narrative dimension based on the predicted score and the labeled score of the preset narrative dimension; if it is determined that the scoring model corresponding to the preset narrative dimension has converged based on the model loss of the preset narrative dimension, then the training of the scoring model corresponding to the preset narrative dimension is terminated; if it is determined that the scoring model corresponding to the preset narrative dimension has not converged based on the model loss of the preset narrative dimension, then the parameters of the scoring model corresponding to the preset narrative dimension are adjusted, and the step of inputting the sample scoring parameters of the preset narrative dimension into the scoring model corresponding to the preset narrative dimension to obtain the predicted score of the preset narrative dimension is re-executed.

[0071] In a possible implementation, the second determining module is specifically configured to:

[0072] The second score of each target media work in each preset narrative dimension and the third score of the target media work are weighted according to the weight coefficient of each preset narrative dimension and the weight coefficient of the media display platform to obtain a fourth score of the target media work; or

[0073] From the second score of the target media work in each preset narrative dimension and the third score of the target media work, the largest score is selected as the fourth score of the target media work.

[0074] In a possible implementation, the third determining module is specifically configured to:

[0075] For each theme, the fourth scores of each target media work of the theme are averaged to obtain the narrative ability score of the narrative subject on the theme; or

[0076] For each theme, the largest fourth score is selected from the fourth scores of each target media work on the theme as the narrative ability score of the narrative subject on the theme.

[0077] In a possible implementation, the apparatus further includes:

[0078] The fifth determination module is used to determine the target subject matter of the media work to be created;

[0079] A fourth acquisition module is used to obtain narrative ability scores of multiple candidate narrative objects on the target theme;

[0080] The sixth determination module is configured to determine the narrative object of the media work to be created from candidate narrative objects whose narrative ability scores are greater than a preset score threshold.

[0081] In the third aspect of the implementation of the present invention, an electronic device is also provided, which includes: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; the memory is used to store computer programs; the processor is used to implement the narrative ability evaluation method of the narrative object described in the first aspect when executing the program stored in the memory.

[0082] In a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the method for evaluating the narrative ability of a narrative object described in the first aspect is implemented.

[0083] In the technical solution provided by the embodiment of the present invention, on the one hand, a normalized scoring interval corresponding to the media display platform, that is, the scores of media works of various themes displayed on the media display platform are all normalized to the same interval, and the first score of the target media work on the media display platform is mapped to the normalized scoring interval corresponding to the media display platform to obtain the third score of the target media work, thereby reducing the sample bias problem caused by the influence of the user group of the media work display platform, that is, improving the rationality of the score of the media work display platform; using the relatively reasonable third score obtained to evaluate the narrative ability of the narrative object, the accuracy of the narrative ability evaluation of the narrative object is improved.

[0084] On the other hand, the narrative ability of the narrative object is subdivided into narrative abilities in various themes, that is, the second scores of multiple target media works in each preset narrative dimension associated with their respective themes and the above-mentioned third scores are combined to evaluate the narrative ability of the narrative object in various themes. This reduces the sample bias problem caused by the gap in the quantity and quality of media works created by the narrative object in different themes, and further improves the accuracy of the narrative ability assessment of the narrative object.

[0085] Of course, it is not necessary to achieve all of the advantages described above simultaneously in order to implement any product or method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for describing the embodiments or the prior art.

[0087] Figure 1 A first flow chart of a method for evaluating the narrative ability of a narrative subject provided by an embodiment of the present invention.

[0088] Figure 2 This is a second flow chart of a method for evaluating the narrative ability of a narrative subject provided by an embodiment of the present invention.

[0089] Figure 3 A third flow chart of a method for evaluating the narrative ability of a narrative subject provided by an embodiment of the present invention.

[0090] Figure 4 This is a fourth flow chart of a method for evaluating the narrative ability of a narrative subject provided by an embodiment of the present invention.

[0091] Figure 5 This is a fifth flow chart of a method for evaluating the narrative ability of a narrative subject provided by an embodiment of the present invention.

[0092] Figure 6 A flowchart of a method for training a scoring model for a preset narrative dimension provided by an embodiment of the present invention.

[0093] Figure 7 This is a sixth flow chart of a method for evaluating the narrative ability of a narrative subject provided by an embodiment of the present invention.

[0094] Figure 8 This is a seventh flow chart of a method for evaluating the narrative ability of a narrative subject provided by an embodiment of the present invention.

[0095] Figure 9 A flowchart of a method for determining a narrative object provided by an embodiment of the present invention.

[0096] Figure 10 A schematic structural diagram of a device for evaluating the narrative ability of a narrative subject provided by an embodiment of the present invention.

[0097] Figure 11 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0098] The technical solutions in the embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention.

[0099] Media works are forms of media intended for the general public. They use narrative techniques to tell a story for viewers, satisfying the psychological needs of target users and ultimately fostering a virtuous cycle of consumer supply and demand. These include, but are not limited to, audio works, visual works, text works, and multimedia works (such as film and television). Audio works include, but are not limited to, audiobooks and radio broadcasts; visual works include, but are not limited to, comics and other media works composed of one or more images; text works include, but are not limited to, online novels and online short articles; and multimedia works include, but are not limited to, television series, films, and variety shows.

[0100] The narrative object is the creator of the media work. The creator can be an independent individual, such as the director and screenwriter of the multimedia work. The creator can also be a creative team, which is not specifically limited in the embodiment of the present invention.

[0101] A media display platform is a platform for displaying media works, which can be video playback software, novel reading software, etc. For example, a user watches a movie on a certain video playback software and can rate the movie on the video playback software. The video playback software can then obtain the movie's rating on the video playback software based on the ratings of all users who watched the movie.

[0102] Currently, the assessment of a subject's narrative ability is primarily based on the quantity and quality of their media works. However, this sample bias leads to inaccurate assessments of the subject's narrative ability. This bias is primarily due to the following two reasons:

[0103] 1. Media works of varying genres are influenced by the user base of the media display platform, resulting in a sample skew in the quantity and quality of media works created by the narrative subject. For example, on a certain media display platform, young users generally give higher ratings to comedic films, while lower ratings are given to historical films. This results in inaccurate assessments of the narrative subject's narrative abilities due to inflated ratings for comedic works and inflated ratings for historical works on the same platform.

[0104] 2. The sample skew problem caused by the disparity in the quantity and quality of the narrative subject's media works on different themes. For example, the narrative subject has produced more media works on comedy themes, and these works have higher ratings, while the narrative subject has produced fewer media works on suspense themes, and these works have lower ratings. The current assessment of the narrative subject's narrative ability is based on all the media works the narrative subject has created, without breaking down the subject types of the media works the narrative subject has created. As a result, only an overall assessment of the narrative subject across all themes is obtained, and the assessment of the narrative subject's narrative ability across different themes is unclear, which in turn leads to an inaccurate assessment of the narrative subject's narrative ability.

[0105] In order to improve the accuracy of the narrative ability evaluation of the narrative object, the embodiment of the present invention provides a narrative ability evaluation method for the narrative object, such as Figure 1 As shown, the method includes the following steps:

[0106] Step S11 : obtaining a plurality of target media works of a narrative object and a first score of each target media work on a media presentation platform.

[0107] Step S12 : determining a second score of each target media work in each preset narrative dimension associated with the subject matter of the target media work based on the scoring model corresponding to each preset narrative dimension.

[0108] Step S13 : Mapping the first score of each target media work to a normalized score interval corresponding to the media display platform to obtain a third score of the target media work.

[0109] Step S14, determining a fourth score of each target media work based on the second score of each preset narrative dimension of each target media work and the third score of the target media work;

[0110] Step S15 : determining the narrative ability score of the narrative subject in each theme according to the fourth score of each target media work in each theme.

[0111] It can be seen that by adopting the technical solution provided by the embodiment of the present invention, on the one hand, a normalized scoring interval corresponding to the media display platform, that is, the scores of media works of various themes displayed on the media display platform are all normalized to the same interval, and the first score of the target media work on the media display platform is mapped to the normalized scoring interval corresponding to the media display platform to obtain the third score of the target media work, thereby reducing the sample bias problem caused by the influence of the user group of the media work display platform, that is, improving the rationality of the score of the media work display platform; using the relatively reasonable third score obtained to evaluate the narrative ability of the narrative object, the accuracy of the narrative ability evaluation of the narrative object is improved.

[0112] On the other hand, the narrative ability of the narrative object is subdivided into narrative abilities in various themes, that is, the second scores of multiple target media works in each preset narrative dimension associated with their respective themes and the above-mentioned third scores are combined to evaluate the narrative ability of the narrative object in various themes. This reduces the sample bias problem caused by the gap in the quantity and quality of media works created by the narrative object in different themes, and further improves the accuracy of the narrative ability assessment of the narrative object.

[0113] For ease of description and understanding, the following description is made using electronic devices as the execution subject, which is not intended to be limiting.

[0114] In step S11, the target media work is a media work created by the narrative subject. When evaluating a narrative subject's narrative ability, the electronic device obtains multiple target media works for the narrative subject and obtains a first rating for each target media work on a media presentation platform. If a target media work is displayed on multiple media presentation platforms, the electronic device obtains a first rating for the target media work for each media presentation platform, for a total of multiple first ratings.

[0115] In the above step S12, the electronic device pre-sets one or more narrative dimensions, namely, preset narrative dimensions. The preset narrative dimensions can be set according to the main selling points of the subject matter of the media work.

[0116] For example, female-oriented themes use the emotional interactions between characters as the main narrative thread to attract a predominantly female user base, and then use the likability of the characters themselves and the good chemistry between them as the main selling points. Based on this, female-oriented themes can be associated with the two pre-set narrative dimensions of likability of the characters themselves and the good chemistry between the characters. Of course, female-oriented themes can also be associated with other pre-set narrative dimensions.

[0117] For example, suspense genres use cases and adventure as the main narrative lines, attracting a primarily young male user base. They also rely on the thrill of the story and the creation of suspense as their main selling points. Based on this, suspense genres can be associated with these two pre-defined narrative dimensions. Of course, suspense genres can also be associated with other pre-defined narrative dimensions.

[0118] The preset narrative dimensions can also include costume and props dimensions and action dimensions, etc.

[0119] The main selling point is the specific narrative technique, also known as the pre-set narrative dimension. The higher the degree of completion (i.e., the second score) of the pre-set narrative dimension associated with the subject matter, the more it satisfies the psychological needs of the media work's users and the more favorable the media work is to users.

[0120] A genre may be associated with multiple pre-defined narrative dimensions, each corresponding to a scoring model. The scoring model's input can be scoring parameters derived from the media work's image, text, and audio data. The output is the media work's score for the corresponding narrative dimension (i.e., the second score). Scoring parameters will be explained in detail later.

[0121] After obtaining target media works created by the narrative subject, the electronic device can determine the genre of each target media work, with each genre being associated with one or more preset narrative dimensions. For each target media work, the electronic device determines a second score for each preset narrative dimension associated with the genre based on the scoring model corresponding to each preset narrative dimension.

[0122] In the embodiment of the present invention, the genre of each target media work can be pre-determined, or the target media work can be classified in a specified manner to determine the genre of the target media work. Two implementation methods for determining the genre of the target media work will be described in detail below and will not be repeated here.

[0123] In step S13, the electronic device pre-sets a normalized rating interval for the media display platform, and the first ratings of each theme displayed on the media display platform are mapped to this normalized rating interval. That is, after obtaining the first ratings of the media works created by the narrative subject on the media display platform, the electronic device maps the first rating of each target media work to the normalized rating interval corresponding to the media display platform, thereby obtaining a third rating for the target media work.

[0124] On the same media display platform, because user groups have varying preferences for media works of different themes, the ratings for media works of different themes are distributed across different rating ranges. In an embodiment of the present invention, the electronic device normalizes the rating ranges for each theme on the media display platform into a single range, namely, the normalized rating range corresponding to the media display platform, and maps the first rating of the target media work on the media display platform to the normalized rating range corresponding to the media display platform. This removes any biased ratings caused by the media display platform's preference for the target media work's theme, and more objectively reflects the target media work's true rating on the media display platform.

[0125] In the embodiment of the present invention, the method of normalizing the score interval of each subject to the specified normalized score interval can be set according to actual needs. For example, the score normalization can be achieved by adding or subtracting scores, or by percentage.

[0126] For example, on the same media display platform, users prefer comedy-themed media works more than suspense-themed media works, and their preference for comedy-themed media works is higher than that for historical-themed media works. For example, the ratings of comedy-themed media works are concentrated in the 7-9 range, the ratings of suspense-themed media works are concentrated in the 5-7 range, and the ratings of historical-themed media works are concentrated in the 3-5 range. The electronic device can use a scoring method to normalize the rating ranges of each genre on the media display platform to a range of 6-8. That is, 1 point is subtracted from the rating of comedy-themed media works, 1 point is added to the rating of suspense-themed media works, and 3 points is added to the rating of historical-themed media works. In this case, the target media work is a comedy, and its first rating on the media display platform is 8 points. By mapping the first rating to the normalized rating range corresponding to the media display platform, the third rating of the target media work is obtained, which is 7 points (8 points minus 1 point).

[0127] In the embodiment of the present invention, step S12 may be performed first and then step S13; step S13 may be performed first and then step S12; or step S12 and step S13 may be performed simultaneously. The embodiment of the present invention does not specifically limit the execution order of step S12 and step S13.

[0128] After determining the second score of each target media work in each preset narrative dimension and the third score of the target media work, the above-mentioned step S14 is performed, that is, for each target media work, the fourth score of the target media work is determined based on the second score of the target media work in each preset narrative dimension and the third score of the target media work.

[0129] Each theme may include multiple target media works. After determining the fourth score of each target media work under each theme, the above step S15 is performed, that is, the narrative ability score of the narrative object in different themes is determined based on the fourth score of each target media work under each theme.

[0130] based on Figure 1 The embodiment shown in the figure, the embodiment of the present invention also provides a method for evaluating the narrative ability of a narrative object, such as Figure 2 As shown, the method may include the following steps S21-S27, wherein step S21 is the same as the above step S11, and steps S24-S27 are the same as the above steps S12-S15.

[0131] Step S22: Acquire subject-related information of each target media work.

[0132] The subject-matter information is information required to determine the subject matter of a media work. For example, the subject-matter information may include user profile data, narrative technique information, work tags, and plot summary information.

[0133] User profile data can be obtained by analyzing users' viewing history. This data may include age distribution, viewing duration, occupation, and geographic location. Narrative technique information, such as the aforementioned preset narrative dimensions, is a key element in media works. Different media works have different narrative technique information.

[0134] When determining the subject matter of the target media work, the electronic device obtains subject matter-related information of the target media work.

[0135] Step S23 : determining the genre of each target media work according to the genre association information of the target media work.

[0136] Because different media presentation platforms may classify the subject matter of a target media work differently, accurately assessing the narrative subject's narrative ability across various subject matters based on the classification of the subject matter by different media presentation platforms cannot be achieved. In embodiments of the present invention, a unified approach, i.e., determining the subject matter of a target media work based on its subject-related information, effectively circumvents this issue of different classifications of the subject matter by different media presentation platforms, thereby improving the accuracy of the assessment of the narrative subject's narrative ability across various subject matters.

[0137] In one embodiment of the present invention, the subject matter association information may include user portrait data and narrative technique information. Based on this, the embodiment of the present invention also provides a narrative ability evaluation method for a narrative object, such as Figure 3 As shown, the method may include the following steps: steps S31-S38, wherein steps S31-S32 are the same as steps S21-S22, and steps S35-S38 are the same as steps S24-S27. Steps S33-S34 are an implementation method of step S23.

[0138] Step S33: Convert the user portrait data and narrative technique information of each target media work into a feature vector of the target media work.

[0139] In step S34, a preset clustering algorithm is used to determine the vector class to which the feature vector of each target media work belongs. The theme corresponding to the vector class is the theme to which the target media work belongs.

[0140] The technical solution provided by the embodiments of this invention utilizes a clustering algorithm based on user profile data and narrative techniques from media works to determine the genre of each target media work. This approach establishes a connection between the narrative techniques of a media work and the psychological needs of the user, enabling more accurate determination of the genre of the target media work and, in turn, improving the accuracy of the narrative ability assessment of the narrative subject.

[0141] In the above step S33 , for each target media work, the electronic device converts the user portrait data and narrative technique information of the target media work into a feature vector of the target media work.

[0142] In an embodiment of the present invention, the electronic device may pre-set conversion rules, according to which the electronic device converts the user portrait data and narrative technique information of each target media work into a feature vector of the target media work. The above conversion rules may be set according to actual needs.

[0143] In one example, the conversion rule may be: extracting data of preset dimensions from user portrait data and narrative technique information, converting the extracted data of each preset dimension into feature values ​​based on the pre-stored correspondence between the data of each preset dimension and the feature values, and forming a feature vector from these feature values.

[0144] For example, data 1 of preset dimension 1 and data 2 of preset dimension 2 are extracted from the user portrait data and narrative technique information. The electronic device pre-stores eigenvalue 1 corresponding to data 1 of preset dimension 1 and eigenvalue 2 corresponding to data 2 of preset dimension 2. Therefore, the electronic device can determine the eigenvector as {eigenvalue 1, eigenvalue 2}.

[0145] In another example, the conversion rule may be: inputting user portrait data into the corresponding neural network to obtain a first feature sequence, and inputting narrative technique information into another corresponding neural network to obtain a second feature sequence, and the first feature sequence and the second feature sequence form a feature vector.

[0146] In the above step S34, the preset clustering algorithm may be a K-Means clustering algorithm (K-means clustering algorithm) or a Mean-Shift algorithm (mean shift algorithm), etc., which is not specifically limited in the embodiment of the present invention.

[0147] Vector clustering is achieved by clustering multiple sample media works using a preset clustering algorithm. For example, an electronic device obtains multiple training samples, each of which is subject-related information about the sample media works. The multiple training samples are converted into feature vectors. Using a preset clustering algorithm, a preset number of feature vectors are selected as anchor points, and the feature vectors of the multiple training samples are clustered to obtain a preset number of clusters. Each cluster can be understood as a collection of training samples whose feature vectors are similar to the user profile data and narrative technique information of the anchor point of that cluster. Subsequently, vectors within a cluster whose distance from the anchor point is greater than a preset distance are used as new anchor points, and the feature vectors of the multiple training samples are clustered again to obtain multiple clusters. The multiple stable clusters ultimately formed are each referred to as vector clusters. A vector cluster can also be referred to as a base cluster, where each base cluster corresponds to a subject matter, i.e., each base cluster has different narrative technique information. In embodiments of the present invention, the resulting vector cluster can be considered a psychological demand model for media works that establishes a connection between the narrative technique of a media work and the psychological needs of the user.

[0148] When multiple vector classes are determined, for each target media work, the electronic device uses a preset clustering algorithm to cluster the target media work to obtain the vector class to which the characteristic vector of the target media work belongs. Based on the vector class to which each target media work belongs, the subject matter to which the target media work belongs can be determined, that is, the subject matter corresponding to the vector class can be used as the subject matter to which the target media work belongs.

[0149] In one embodiment of the present invention, the subject matter related information includes work labels and plot summary information. Based on this, the embodiment of the present invention also provides a method for evaluating the narrative ability of a narrative object, such as Figure 4 As shown, the method may include the following steps: Steps S41-S49, wherein Steps S41-S42 are the same as Steps S21-S22, and Steps S46-S49 are the same as Steps S24-S27. Steps S43-S45 are an implementation method of Step S23.

[0150] Step S43 , extracting keywords for each theme from the work label and plot summary information of each target media work.

[0151] Step S44 : for each target media work, determine the score and value of the keyword corresponding to the target media work in each theme as the theme score of the theme corresponding to the target media work.

[0152] Step S45 : For each target media work, determine the theme associated with the highest theme score corresponding to the target media work, and use it as the theme of the target media work.

[0153] In the technical solution provided by the embodiments of the present invention, the theme score of a target media work across various themes is determined to determine the theme to which the target media work belongs. That is, a cross-value system is used to score the theme keywords extracted from the work tag and plot synopsis information of each target media work. The scores are accumulated across the themes, and the theme with the highest score is determined as the theme to which the target media work belongs. As can be seen, in the embodiments of the present invention, media works are cross-validated across different themes, and the degree to which the works meet the psychological needs of users is quantified. Since the theme score of the target media work across various themes is taken into account, the theme to which the target media work belongs can be determined more accurately, thereby improving the accuracy of the narrative ability assessment of the narrative subject.

[0154] In step S43, the electronic device can pre-set keywords for each theme and the scores of each keyword in different themes, which can also be understood as the weights of keywords in different themes. Keywords are words closely related to the theme. Keywords in different themes can be the same or different, and the scores of keywords in different themes can be the same or different. However, keywords and keyword scores in different themes are not exactly the same.

[0155] For example, keywords for comedy themes include police, funny, and romance, while keywords for police and gangster themes include police, gangsters, and romance. In a comedy theme, the keyword "police" scores 50 points, the keyword "funny" scores 90 points, and the keyword "romance" scores 40 points. In a police and gangster theme, the keyword "police" scores 95 points, the keyword "gangster" scores 90 points, and the keyword "romance" scores 40 points.

[0156] For each theme of a target media work, keywords for the theme are extracted from the work tag and plot summary information of the target media work. The electronic device may have multiple keywords pre-set for a theme. When extracting keywords for the theme from the work tag and plot summary information of the target media work, all, some, or none of the pre-set keywords for the theme may be extracted.

[0157] In the above step S44, for a target media work, for each theme, the electronic device accumulates the various keywords belonging to the theme to obtain the score and value of the keywords corresponding to the target media work in the theme. The score and value is the theme score of the theme corresponding to the target media work.

[0158] In step S45 , for each target media work, the electronic device selects the highest theme score from the theme scores corresponding to the target media work, and uses the theme associated with the highest theme score as the theme of the target media work.

[0159] For example, the keywords "detective", "adventure", "reasoning", and "hilarious" were extracted from a target media work. In the suspense genre, "detective" scored 9 points, "adventure" scored 3 points, "reasoning" scored 7 points, and "hilarious" scored 0 points; in the adventure genre, "detective" scored 0 points, "adventure" scored 9 points, "reasoning" scored 0 points, and "hilarious" scored 0 points; in the comedy genre, "detective" scored 0 points, "adventure" scored 0 points, and "hilarious" scored 8 points, excluding "reasoning". Therefore, it can be concluded that the keyword score sum value corresponding to the target media work in the suspense genre is 19 points (9+7+3 points), the keyword score sum value corresponding to the target media work in the adventure genre is 9 points, and the keyword score sum value corresponding to the target media work in the comedy genre is 8 points. The highest keyword score sum value is 19 points, that is, the highest genre score is 19 points, and 19 points corresponds to the suspense genre. Then, it can be determined that the genre of the target media work is suspense.

[0160] based on Figure 1 The embodiment shown in the figure, the embodiment of the present invention also provides a method for evaluating the narrative ability of a narrative object, such as Figure 5 As shown, the method may include the following steps S51-S56. Step S51 is the same as the above-mentioned step S11, and steps S54-S56 are the same as the above-mentioned steps S13-S15. Steps S52-S53 are an implementation method of step S12.

[0161] Step S52 : extracting the scoring parameters of each preset narrative dimension associated with the subject matter of each target media work from the scoring association information associated with the target media work.

[0162] Step S53 : Input the scoring parameters of each target media work in each preset narrative dimension into the scoring model corresponding to the preset narrative dimension to obtain a second score of the target media work in the preset narrative dimension.

[0163] In the technical solution provided by the embodiment of the present invention, the narrative object is scored on each preset narrative dimension in different themes, and then the narrative ability requirements of the narrative object in different themes are limited, thereby further improving the accuracy of the narrative ability assessment of the narrative object.

[0164] In step S52, each target media work is associated with rating-related information. The rating-related information may include, but is not limited to, image materials, text materials, and audio materials. Image materials may include video clips or pictures of the media work, text materials may include a brief introduction to the media work, user reviews of the media work, and audio materials may include audio clips of the media work.

[0165] The pre-set narrative dimension can be understood as the main selling point of each genre. The scoring parameters for the pre-set narrative dimension can be understood as a parameter that reflects this main selling point. For example, a suspense film's main selling points are the thrilling story and the suspense it creates. This includes the two pre-set narrative dimensions of thrilling story and suspense it creates. Parameters that reflect the thrilling story, and thus the scoring parameters for the thrilling pre-set narrative dimension, include action smoothness and plot compactness. Parameters that reflect the suspense it creates, and thus the scoring parameters for the suspense it creates, include the number of plot twists. A pre-set narrative dimension can have one or more scoring parameters.

[0166] For each target media work, the electronic device extracts, from the target media work's rating association information, rating parameters for each preset narrative dimension associated with the target media work's subject matter. In this case, one or more rating parameters are extracted for each preset narrative dimension for each target media work.

[0167] In the above step S53, a subject matter may be associated with multiple preset narrative dimensions, and each preset narrative dimension corresponds to a scoring model. Multiple subjects may be associated with the same preset narrative dimension, and the scoring model corresponding to the preset narrative dimension can be shared by multiple subjects. The scoring model can be determined based on the different types of scoring association information that it needs to process. The scoring model can be used to quantitatively evaluate the score (i.e., degree of completion) of the media work in the corresponding narrative dimension based on the scoring association information of the media work. For example, when the scoring association information is an image material, the scoring model can be a convolutional neural network model, that is, the scoring model for each preset narrative dimension associated with the image material is a convolutional neural network model; when the scoring association information is text material, the scoring model can be an NLP model, that is, the scoring model for each preset narrative dimension associated with the text material is an NLP model.

[0168] After the electronic device extracts the scoring parameters of each preset narrative dimension associated with the subject matter of a target media work, for each preset narrative dimension, the scoring parameters of the preset narrative dimension are input into a scoring model corresponding to the preset narrative dimension. After the scoring model processes the input scoring parameters, it outputs a second score of the target media work in the preset narrative dimension.

[0169] The above network model can of course also be other network models, such as a deep neural network model, etc. The specific selection can be determined according to actual needs and is not specifically limited here.

[0170] based on Figure 5 The embodiment shown in the figure, the embodiment of the present invention also provides a training method for a scoring model with a preset narrative dimension, such as Figure 6 As shown, for each preset narrative dimension, electronic devices can refer to Figure 6 The process shown is to train and obtain the scoring model of the preset narrative dimension. The training method of the above-mentioned scoring model may include the following steps.

[0171] Step S61: acquiring training data, wherein the training data is score association information of annotated scores with preset narrative dimensions;

[0172] The training data can include ratings associated with multiple media works. The specific number can be set based on actual needs. For example, when the scoring model requires high scoring accuracy, the ratings associated with a larger number of media works can be used; when the device performance is poor, the ratings associated with a smaller number of media works can be used.

[0173] Step S62: extracting sample scoring parameters of the preset narrative dimension from the training data.

[0174] Step S63: input the sample scoring parameters of the preset narrative dimension into the scoring model corresponding to the preset narrative dimension to obtain a predicted score of the preset narrative dimension.

[0175] Step S64: determining the model loss of the preset narrative dimension based on the predicted score and the labeled score of the preset narrative dimension.

[0176] In the embodiment of the present invention, the electronic device may pre-set a loss threshold. After obtaining the model loss, if the model loss is less than the loss threshold, it can be determined that the scoring model has converged; otherwise, it is determined that the scoring model has not converged.

[0177] In an embodiment of the present invention, the electronic device may also pre-set a loss threshold and an iteration threshold. After obtaining the model loss, if the model loss is less than the loss threshold, the scoring model is determined to have converged. If the number of training iterations is greater than the iteration threshold, the scoring model is determined to have converged. Otherwise, the scoring model is determined to have not converged.

[0178] The above loss threshold and iteration threshold can be set according to actual needs.

[0179] Step S65: If it is determined that the scoring model corresponding to the preset narrative dimension has converged based on the model loss of the preset narrative dimension, then the scoring model training corresponding to the preset narrative dimension is terminated;

[0180] Step S66 : If it is determined based on the model loss of the preset narrative dimension that the scoring model corresponding to the preset narrative dimension has not converged, the parameters of the scoring model corresponding to the preset narrative dimension are adjusted, and step S63 is executed again.

[0181] By adopting the technical solution provided by the embodiment of the present invention, the scoring model of the preset narrative dimension is trained using the training data, so that the scoring model can fully learn the change rules of the training data, and then use the trained scoring model to more accurately predict the second score of the target media work in each preset narrative dimension, and make more accurate limitations on the narrative ability requirements of the narrative object in different themes, thereby further improving the accuracy of the narrative ability evaluation of the narrative object.

[0182] The description of steps S61-S66 is relatively simple. For details, please refer to the above Figure 1-5 Part of the description.

[0183] based on Figure 1 The embodiment shown in the figure, the embodiment of the present invention also provides a method for evaluating the narrative ability of a narrative object, such as Figure 7 As shown, the method may include the following steps S71-S75, wherein steps S71-S73 are the same as the above steps S11-S13, and step S75 is the same as the above step S15. Step S74 is an implementation method of step S14.

[0184] S74, based on the weight coefficient of each preset narrative dimension and the weight coefficient of the media display platform, the second score of each target media work in each preset narrative dimension and the third score of the target media work are weighted to obtain a fourth score of the target media work.

[0185] In an embodiment of the present invention, the electronic device pre-records the weight coefficient of each preset narrative dimension and the weight coefficient of the media display platform. For each target media work, the electronic device can use the weight coefficient of each preset narrative dimension and the weight coefficient of the media display platform to weight the second score of the target media work in each preset narrative dimension and the third score of the target media work to obtain the fourth score of the target media work. The weight coefficient of each preset narrative dimension and the weight coefficient of the media display platform can also be manually assigned or obtained through neural network training, and the embodiment of the present invention does not specifically limit this.

[0186] For example, the electronic device pre-records the following: weight coefficient 1 for preset narrative dimension 1, weight coefficient 2 for preset narrative dimension 2, and weight coefficient 3 for the media presentation platform. Target media work a has a second score of 1 for preset narrative dimension 1, a second score of 2 for preset narrative dimension 2, and a third score of 3. Therefore, the electronic device can determine the fourth score of target media work a as: score 1 * weight coefficient 1 + score 2 * weight coefficient 2 + score 3 * weight coefficient 3.

[0187] Optionally, the sum of the weight coefficients of multiple preset narrative dimensions and the weight coefficient of the media presentation platform is 1, which facilitates the calculation of the fourth score of the media work by the electronic device.

[0188] In the technical solution provided by the embodiment of the present invention, the fourth score of the target media work is obtained by weighting the second score of the target media work in each preset narrative dimension and the third score of the target media work, further eliminating the impact of sample bias on the score of the target media work, making the score of the target media work more objective and fair, and further improving the accuracy of the narrative ability assessment of the narrative object.

[0189] In embodiments of the present invention, the electronic device may also use other methods to determine the fourth score of the target media work. For example, the electronic device may select the largest score from the second score of the target media work in each preset narrative dimension and the third score of the target media work as the fourth score of the target media work. For example, the second score of target media work a in preset narrative dimension 1 is score 1, the second score in preset narrative dimension 2 is score 2, and the third score of the target media work is score 3. If score 1 > score 2 > score 3, the electronic device may determine score 1 as the fourth score of target media work a.

[0190] based on Figure 1 The embodiment shown in the figure, the embodiment of the present invention also provides a method for evaluating the narrative ability of a narrative object, such as Figure 8 As shown, the method may include the following steps S81-S85, wherein steps S81-S84 are the same as the above steps S11-S14, and step S85 is an implementation method of step S15.

[0191] S85, for each theme, average the fourth scores of each target media work on the theme to obtain the narrative ability score of the narrative subject on the theme.

[0192] In this embodiment of the present invention, a theme may have one or more target media works. After determining the fourth scores of the target media works created by the narrative subject for each theme, the fourth scores of each target media work for each theme are averaged to obtain the narrative subject's narrative ability score for that theme.

[0193] For example, the target media works for theme b include target media works 1-3, where the fourth score of target media work 1 is score 1', the fourth score of target media work 2 is score 2', and the fourth score of target media work 3 is score 3'. The electronic device can then determine the narrative subject's narrative ability score for theme b as (score 1' + score 2' + score 3') / 3.

[0194] In the technical solution provided by the embodiment of the present invention, the second scores and third scores of multiple target media works in each preset narrative dimension associated with their respective themes are combined to obtain the narrative ability score of the narrative object in each themes. This reduces the sample bias problem caused by the gap in quantity and quality of media works created by the narrative object in different themes, and further improves the accuracy of the narrative ability assessment of the narrative object.

[0195] In embodiments of the present invention, the electronic device may also use other methods to determine the narrative subject's narrative ability score for each genre. For example, for each genre, the electronic device may select the largest fourth score from the fourth scores of each target media work for that genre as the narrative subject's narrative ability score for that genre.

[0196] For example, the target media works of subject matter c include target media works 1-3, among which the fourth score of target media work 1 is score 1, the fourth score of target media work 2 is score 2, and the fourth score of target media work 3 is score 3. Score 1>score 2>score 3. Then, the electronic device can determine that the narrative ability score of the narrative object on subject matter c is: score 1.

[0197] based on Figure 1-8 The embodiment shown in the figure, the embodiment of the present invention also provides a method for determining a narrative object, such as Figure 9 As shown, the method may include the following steps S91-S93.

[0198] S91, determine the target subject matter of the media work to be created.

[0199] When a media work needs to be created (i.e., a media work to be created), the electronic device can obtain the subject matter of the media work to be created, i.e., the target subject matter. The target subject matter can be input by the user into the electronic device, or can be obtained through Figure 2 、 Figure 3 or Figure 4The method shown determines the target subject matter of the media work to be created, but is not limited to this.

[0200] S92, obtaining narrative ability scores of multiple candidate narrative objects on the target subject matter.

[0201] For the determination of the narrative ability score of each candidate narrative subject on the target theme, please refer to Figure 1-8 Part of the description.

[0202] S93, determining a narrative object of the media work to be created from candidate narrative objects whose narrative ability scores are greater than a preset score threshold.

[0203] After obtaining the narrative ability scores of multiple candidate narrative objects on the target subject matter, the electronic device screens out candidate narrative objects whose narrative ability scores on the target subject matter are greater than a preset score threshold from the multiple candidate narrative objects, and determines the narrative object of the media work to be created from the screened candidate narrative objects.

[0204] For example, the electronic device may output a plurality of selected candidate narrative objects in descending order of narrative ability scores, so that the user can select a narrative object for the media work to be created based on the output candidate narrative objects.

[0205] For another example, the electronic device may select a candidate narrative object with the highest narrative ability score as the narrative object for the media work to be created.

[0206] In an embodiment of the present invention, the electronic device may also determine the narrative object for the media work to be created from the screened candidate narrative objects based on information such as the media work production cycle length and the number of media works of the target theme that have been created. For example, the electronic device may determine the candidate narrative object with the shortest media work production cycle from the screened candidate narrative objects as the narrative object for the media work to be created.

[0207] The technical solution provided by the embodiments of the present invention determines the narrative subject of a media work to be created based on the target subject matter of the work and the narrative ability scores of candidate narrative subjects on the target subject matter. By improving the accuracy of the narrative subject's narrative ability assessment on the target subject matter, suitable narrative subjects can be recommended for different types of media works to be created, resulting in higher-quality media works and, in turn, increasing the commercial value of the media works to be created.

[0208] Based on the same inventive concept, according to the above embodiment of the present invention, a narrative ability evaluation method for a narrative object is provided. Accordingly, an embodiment of the present invention provides a narrative ability evaluation device for a narrative object, the structural diagram of which is shown in FIG. Figure 10 Shown, including:

[0209] A first acquisition module 101 is configured to acquire a plurality of target media works of a narrative object and a first score of each target media work on a media presentation platform;

[0210] A first determining module 102 is configured to determine, based on a scoring model corresponding to each preset narrative dimension, a second score for each target media work associated with each preset narrative dimension of the subject matter of the target media work;

[0211] A mapping module 103 is configured to map the first score of each target media work to a normalized score interval corresponding to the media display platform to obtain a third score of the target media work;

[0212] A second determining module 104 is configured to determine a fourth score of each target media work based on the second score of each preset narrative dimension of each target media work and the third score of the target media work;

[0213] The third determining module 105 is configured to determine the narrative ability score of the narrative subject in each theme according to the fourth score of each target media work in each theme.

[0214] Optionally, the narrative ability assessment device of the narrative subject may further include:

[0215] The second acquisition module is used to obtain the subject matter related information of each target media work;

[0216] The fourth determining module is configured to determine the genre of each target media work based on the genre association information of the target media work.

[0217] Optionally, the subject matter-related information includes user portrait data and narrative technique information;

[0218] The fourth determination module may be specifically configured to: convert the user portrait data and narrative technique information of each target media work into a feature vector of the target media work;

[0219] A preset clustering algorithm is used to determine the vector class to which the characteristic vector of each target media work belongs. The subject matter corresponding to the vector class is the subject matter to which the target media work belongs.

[0220] Optionally, the subject matter-related information includes work tags and plot summary information;

[0221] The fourth determination module may be specifically configured to: extract keywords for each theme from the work label and plot summary information of each target media work;

[0222] For each target media work, determine the score and value of the keyword corresponding to the target media work in each theme as the theme score of the theme corresponding to the target media work;

[0223] For each target media work, the theme associated with the highest theme score corresponding to the target media work is determined as the theme to which the target media work belongs.

[0224] Optionally, the first determining module 102 may be specifically configured to:

[0225] Extracting, from the score association information associated with each target media work, a score parameter for each preset narrative dimension associated with the subject matter of the target media work;

[0226] The scoring parameters of each target media work in each preset narrative dimension are input into the scoring model corresponding to the preset narrative dimension to obtain a second score of the target media work in the preset narrative dimension.

[0227] Optionally, the scoring association information includes image materials and text materials; the scoring model for each preset narrative dimension associated with the image materials is a convolutional neural network model; and the scoring model for each preset narrative dimension associated with the text materials is an NLP model.

[0228] Optionally, the narrative ability assessment device of the narrative subject may further include:

[0229] A third acquisition module is used to acquire training data, where the training data is score association information with annotated scores of the preset narrative dimension;

[0230] An extraction module, configured to extract sample scoring parameters of the preset narrative dimension from the training data;

[0231] An input module, configured to input the sample scoring parameters of the preset narrative dimension into a scoring model corresponding to the preset narrative dimension to obtain a predicted score for the preset narrative dimension;

[0232] A training module is used to determine the model loss of the preset narrative dimension based on the predicted score and the labeled score of the preset narrative dimension; if it is determined that the scoring model corresponding to the preset narrative dimension has converged based on the model loss of the preset narrative dimension, then the training of the scoring model corresponding to the preset narrative dimension is terminated; if it is determined that the scoring model corresponding to the preset narrative dimension has not converged based on the model loss of the preset narrative dimension, then the parameters of the scoring model corresponding to the preset narrative dimension are adjusted, and the sample scoring parameters of the preset narrative dimension are input into the scoring model corresponding to the preset narrative dimension again to obtain the predicted score of the preset narrative dimension.

[0233] Optionally, the second determining module 104 may be specifically configured to:

[0234] The second score of each target media work in each preset narrative dimension and the third score of the target media work are weighted according to the weight coefficient of each preset narrative dimension and the weight coefficient of the media display platform to obtain a fourth score of the target media work; or

[0235] From the second score of the target media work in each preset narrative dimension and the third score of the target media work, the largest score is selected as the fourth score of the target media work.

[0236] Optionally, the third determining module 105 may be specifically configured to:

[0237] For each theme, average the fourth score of each target media work on that theme to obtain the narrative subject's narrative ability score on that theme; or

[0238] For each theme, the largest fourth score is selected from the fourth scores of each target media work on the theme as the narrative ability score of the narrative subject on the theme.

[0239] Optionally, the narrative ability assessment device of the narrative subject may further include:

[0240] The fifth determination module is used to determine the target subject matter of the media work to be created;

[0241] The fourth acquisition module is used to obtain narrative ability scores of multiple candidate narrative objects on the target subject matter;

[0242] The sixth determination module is used to determine the narrative object of the media work to be created from the candidate narrative objects whose narrative ability scores are greater than a preset score threshold.

[0243] The embodiment of the present invention further provides an electronic device, such as Figure 11 As shown, it includes a processor 111, a communication interface 112, a memory 113 and a communication bus 114, wherein the processor 111, the communication interface 112, and the memory 113 communicate with each other through the communication bus 114.

[0244] Memory 113, for storing computer programs;

[0245] The processor 111 is configured to execute the program stored in the memory 113 by performing at least the following steps:

[0246] Obtaining multiple target media works of a narrative object and a first rating of each target media work on a media display platform;

[0247] determining a second score for each target media work on each predetermined narrative dimension associated with the subject matter of the target media work;

[0248] Mapping the first score of each target media work to the normalized score interval corresponding to the media display platform to obtain the third score of the target media work;

[0249] determining a fourth score of each target media work based on the second score of each preset narrative dimension of each target media work and the third score of the target media work;

[0250] According to the fourth score of each target media work of each theme, the narrative ability score of the narrative subject on the theme is determined.

[0251] The communication bus mentioned in the terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used in the figure, but this does not mean that there is only one bus or only one type of bus.

[0252] The communication interface is used for communication between the above terminal and other devices.

[0253] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0254] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0255] In another embodiment provided by the present invention, a computer-readable storage medium is also provided, in which a computer program is stored. When the computer program is executed by a processor, the narrative ability evaluation method of any narrative object in the above embodiments is implemented.

[0256] In another embodiment provided by the present invention, a computer program product comprising instructions is provided, which, when executed on a computer, enables the computer to execute the narrative ability assessment method of a narrative subject in any one of the above embodiments.

[0257] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0258] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0259] Each embodiment in this specification is described in a related manner. Similar portions between the embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences from other embodiments. In particular, the device embodiments are generally similar to the method embodiments, so their description is relatively simple. For related portions, refer to the description of the method embodiments.

[0260] The above are only preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention are included in the scope of protection of the present invention.

Claims

1. A method for evaluating the narrative ability of a narrative subject, characterized in that: include: Obtaining multiple target media works of a narrative object and a first rating of each target media work on a media display platform; Determining, for each target media work, a second score for each preset narrative dimension associated with the subject matter of the target media work based on a scoring model corresponding to each preset narrative dimension; Mapping the first score of each target media work to a normalized score interval corresponding to the media display platform to obtain a third score of the target media work; determining a fourth score of each target media work based on the second score of each preset narrative dimension of each target media work and the third score of the target media work; The narrative ability score of the narrative subject in each theme is determined according to the fourth score of each target media work in each theme.

2. The method according to claim 1, characterized in that The method further comprises: Obtaining subject matter related information of each target media work; According to the theme association information of each target media work, the theme of the target media work is determined.

3. The method according to claim 2, characterized in that The subject matter related information includes user portrait data and narrative technique information; The step of determining the genre of each target media work based on the genre association information of the target media work includes: Converting the user portrait data and narrative technique information of each target media work into a feature vector of the target media work; A preset clustering algorithm is used to determine the vector class to which the characteristic vector of each target media work belongs. The subject matter corresponding to the vector class is the subject matter to which the target media work belongs.

4. The method according to claim 2, characterized in that The subject matter related information includes work labels and plot summary information; The step of determining the subject matter of each target media work based on the subject matter association information of the target media work includes: Extract keywords for each genre from the work tags and plot synopsis information of each target media work; For each target media work, determine the score and value of the keyword corresponding to the target media work in each theme as the theme score of the theme corresponding to the target media work; For each target media work, the theme associated with the highest theme score corresponding to the target media work is determined as the theme to which the target media work belongs.

5. The method according to claim 1, wherein The step of determining, based on the scoring model corresponding to each preset narrative dimension, a second score for each target media work associated with the subject matter of the target media work for each preset narrative dimension includes: Extracting, from the score association information associated with each target media work, a score parameter for each preset narrative dimension associated with the subject matter of the target media work; The scoring parameters of each target media work in each preset narrative dimension are input into the scoring model corresponding to the preset narrative dimension to obtain a second score of the target media work in the preset narrative dimension.

6. The method according to claim 5, characterized in that The rating-related information includes image materials and text materials; The scoring model for each preset narrative dimension associated with the image material is a convolutional neural network model; The scoring model for each preset narrative dimension associated with the text material is a natural language processing (NLP) model.

7. The method according to claim 5 or 6, characterized in that The method further includes: for each preset narrative dimension, using the following steps to train a scoring model for the preset narrative dimension: Acquiring training data, wherein the training data is score association information having annotated scores of the preset narrative dimension; Extracting sample scoring parameters of the preset narrative dimension from the training data; Inputting the sample scoring parameters of the preset narrative dimension into the scoring model corresponding to the preset narrative dimension to obtain a predicted score for the preset narrative dimension; Determining a model loss for the preset narrative dimension based on the predicted score and the labeled score for the preset narrative dimension; If it is determined that the scoring model corresponding to the preset narrative dimension has converged according to the model loss of the preset narrative dimension, then the training of the scoring model corresponding to the preset narrative dimension is terminated; If it is determined based on the model loss of the preset narrative dimension that the scoring model corresponding to the preset narrative dimension has not converged, the parameters of the scoring model corresponding to the preset narrative dimension are adjusted, and the step of inputting the sample scoring parameters of the preset narrative dimension into the scoring model corresponding to the preset narrative dimension to obtain the predicted score of the preset narrative dimension is re-executed.

8. The method according to claim 1, characterized in that The step of determining the fourth score of each target media work based on the second score of each preset narrative dimension of each target media work and the third score of the target media work includes: The second score of each target media work in each preset narrative dimension and the third score of the target media work are weighted according to the weight coefficient of each preset narrative dimension and the weight coefficient of the media display platform to obtain a fourth score of the target media work; or From the second score of the target media work in each preset narrative dimension and the third score of the target media work, the largest score is selected as the fourth score of the target media work.

9. The method according to claim 1, characterized in that Determining the narrative ability score of the narrative subject in each subject matter based on the fourth score of each target media work in each subject matter includes: For each theme, the fourth scores of each target media work of the theme are averaged to obtain the narrative ability score of the narrative subject on the theme; or For each theme, the largest fourth score is selected from the fourth scores of each target media work of the theme as the narrative ability score of the narrative subject on the theme.

10. The method according to claim 1, characterized in that The method further comprises: Determine the target subject matter of the media work to be created; Obtaining narrative ability scores of multiple candidate narrative subjects on the target subject matter; The narrative object of the media work to be created is determined from candidate narrative objects whose narrative ability scores are greater than a preset score threshold.

11. A device for evaluating the narrative ability of a narrative subject, characterized in that: include: A first acquisition module is used to acquire multiple target media works of a narrative object and a first score of each target media work on a media display platform; A first determination module is configured to determine, based on a rating model corresponding to each preset narrative dimension, a second rating of each target media work for each preset narrative dimension associated with the subject matter of the target media work; a mapping module, configured to map the first score of each target media work to a normalized score interval corresponding to the media display platform to obtain a third score of the target media work; A second determination module is configured to determine a fourth score of each target media work based on the second score of each preset narrative dimension of each target media work and the third score of the target media work; The third determining module is configured to determine the narrative ability score of the narrative subject in each theme according to the fourth score of each target media work in each theme.

12. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory for storing computer programs; A processor, configured to implement the method steps described in any one of claims 1 to 10 when executing a program stored in a memory.

13. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps of any one of claims 1 to 10 are implemented.

Images