Film and television project income prediction method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a film and television project revenue prediction method and device, electronic equipment and a storage medium. The film and television project revenue prediction method comprises the steps of obtaining historical performance data of each main creator in a target card and driver combination corresponding to a target film and television project; performing statistical analysis on the historical performance data of each main creator to obtain a characteristic value of each main creator in each data dimension; further obtaining a feature transformation value of each main creator in each data dimension; based on the feature transformation value of each main creator in each data dimension and the feature weight corresponding to the corresponding data dimension, determining an evaluation score of each main creator; and according to the evaluation score of each main creator and the feature transformation value of each main creator in each data dimension, predicting the income of the target film and television project under the condition of using the target card combination. By applying the technical scheme provided by the embodiment of the invention, the movie and television project income prediction accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer application technology, and in particular to a method, device, electronic device and storage medium for predicting revenue of a film and television project. Background Art

[0002] In the entertainment industry, film and television project evaluation plays an important role for film and television production companies, novel platforms, video platforms, etc. Film and television project evaluation includes predicting the revenue generated by film and television projects after they are launched. By predicting the revenue of film and television projects, the cost rate can be calculated to provide decision support for whether to approve the film and television projects. Film and television projects may have different cast combinations to choose from, and the use of different cast combinations may result in different revenues for film and television projects.

[0003] At present, the revenue prediction of film and television projects using a certain cast combination is mostly done directly through the feature value of a single data dimension. The feature value of a single data dimension is difficult to reflect the advantages and disadvantages. For example, if the single data dimension is the score, the actors of cast combination A have a higher score on a certain platform, but the corresponding number of scorers is small, and the actors of cast combination B have a slightly lower score on the same platform, but the corresponding number of scorers is large. Through the feature value of this single data dimension, it will be predicted that the film and television project will have higher revenue using cast combination A, but from business experience, cast combination B is more suitable. It can be seen from this that simply using the feature value of a single data dimension to predict the revenue of a film and television project using a certain cast combination does not have high prediction accuracy, which makes it impossible to provide decision support for whether to establish a film and television project. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method, device, electronic device and storage medium for predicting the income of a film and television project, so as to improve the accuracy of the forecast of the income of a film and television project and provide decision support for whether to approve the film and television project. The specific technical solution is as follows:

[0005] In a first aspect, a method for predicting income of a film and television project is provided, comprising:

[0006] Obtain the historical performance data of each main creative staff in the target cast combination corresponding to the target film and television project;

[0007] Performing statistical analysis on the historical performance data of each creative staff in the target cast combination to obtain the characteristic value of each creative staff in each data dimension;

[0008] According to the preset transformation rules, the characteristic values ​​of each main creator in each data dimension are transformed respectively to obtain the characteristic transformation value of each main creator in each data dimension;

[0009] Determine the evaluation score of each main creator based on the feature transformation value of each main creator in each data dimension and the feature weight corresponding to the corresponding data dimension;

[0010] According to the evaluation score of each main creative staff and the feature transformation value of each main creative staff in each data dimension, the income of the target film and television project when using the target cast combination is predicted.

[0011] Optionally, transforming the characteristic value of each creative staff in each data dimension according to a preset transformation rule to obtain the characteristic transformation value of each creative staff in each data dimension includes:

[0012] For each data dimension of each main creator, determine the ratio of the characteristic value of the current main creator in the current data dimension to the reference value of the current data dimension, wherein the current main creator is the main creator targeted by the current operation, and the current data dimension is the data dimension targeted by the current operation;

[0013] The natural logarithm value of the ratio is determined as the feature transformation value of the current creative person in the current data dimension.

[0014] Optionally, determining the evaluation score of each main creative staff member based on the feature transformation value of each main creative staff member in each data dimension and the feature weight corresponding to the corresponding data dimension includes:

[0015] For each data dimension of each main creator, the product of the feature transformation value of the current main creator in the current data dimension and the feature weight corresponding to the current data dimension is determined as the feature value score of the current main creator in the current data dimension, wherein the current main creator is the main creator targeted by the current operation, and the current data dimension is the data dimension targeted by the current operation;

[0016] Based on the characteristic value score of the current main creative personnel in each data dimension, an evaluation score of the current main creative personnel is obtained.

[0017] Optionally, after determining the product of the feature transformation value of the current main creator in the current data dimension and the feature weight corresponding to the current data dimension as the feature value score of the current main creator in the current data dimension and before obtaining the evaluation score of the current main creator based on the feature value score of the current main creator in each data dimension, the method further includes:

[0018] The characteristic value score of the current main creative personnel in the current data dimension is amplified.

[0019] Optionally, after determining the evaluation score of each main creative staff member based on the feature transformation value of each main creative staff member in each data dimension and the feature weight corresponding to the corresponding data dimension, and before predicting the income of the target film and television project when the target cast combination is used according to the evaluation score of each main creative staff member and the feature transformation value of each main creative staff member in each data dimension, the method further includes:

[0020] Adjust the evaluation score of each creative staff member to a positive number.

[0021] Optionally, predicting the revenue of the target film and television project when the target cast combination is used according to the evaluation score of each main creative staff and the feature transformation value of each main creative staff in each data dimension includes:

[0022] Input the evaluation score of each main creator and the feature transformation value of each main creator in each data dimension into the pre-trained prediction model;

[0023] Based on the output of the prediction model, the revenue of the target film and television project when using the target cast combination is predicted.

[0024] Optionally, the target cast combination is any one of a plurality of cast combinations corresponding to the target film and television project, and the method further includes:

[0025] After obtaining the predicted revenue of the target film and television project when using each of the multiple cast combinations, the cast combination selected for the target film and television project is determined based on the predicted revenue of the target film and television project when using each cast combination.

[0026] In a second aspect, a device for predicting income of a film and television project is provided, comprising:

[0027] An acquisition module is used to acquire the historical performance data of each main creative staff in the target cast combination corresponding to the target film and television project;

[0028] An analysis module is used to perform statistical analysis on the historical performance data of each main creative staff in the target cast combination to obtain the characteristic value of each main creative staff in each data dimension;

[0029] A transformation module is used to transform the characteristic value of each main creator in each data dimension according to a preset transformation rule to obtain the characteristic transformation value of each main creator in each data dimension;

[0030] A determination module, used to determine the evaluation score of each main creative staff member based on the feature transformation value of each main creative staff member in each data dimension and the feature weight corresponding to the corresponding data dimension;

[0031] The prediction module is used to predict the income of the target film and television project when using the target cast combination based on the evaluation score of each main creative staff and the feature transformation value of each main creative staff in each data dimension.

[0032] In a third aspect, an electronic device is provided, comprising 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;

[0033] Memory, used to store computer programs;

[0034] The processor is used to implement the steps of the film and television project revenue prediction method described in the first aspect when executing the program stored in the memory.

[0035] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the program is executed by a processor, the steps of the film and television project revenue prediction method described in the first aspect are implemented.

[0036] In a fifth aspect, a computer program product is provided, which includes computer instructions, which are stored in a computer-readable storage medium and are suitable for being read and executed by a processor, so that an electronic device having the processor performs the steps of the film and television project revenue prediction method described in the first aspect.

[0037] Applying the technical solution provided by the embodiment of the present application, first obtain the historical performance data of each creative staff in the target cast combination corresponding to the target film and television project, then statistically analyze the historical performance data of each creative staff in the target cast combination, obtain the characteristic value of each creative staff in each data dimension, and then transform the characteristic value of each creative staff in each data dimension according to the preset transformation rule, obtain the characteristic transformation value of each creative staff in each data dimension, determine the evaluation score of each creative staff based on the characteristic transformation value of each creative staff in each data dimension and the characteristic weight corresponding to the corresponding data dimension, and finally predict the income of the target film and television project under the condition of using the target cast combination according to the evaluation score of each creative staff and the characteristic transformation value of each creative staff in each data dimension. By transforming the characteristic value of each data dimension, and then obtaining the evaluation score of the creative staff based on the characteristic transformation value and the corresponding characteristic weight, and then predicting the income of the target film and television project based on the evaluation score, etc., it is possible to increase the comparability and interpretability between the features, improve the accuracy of the forecast of the income of the film and television project, thereby helping to select a suitable cast combination and provide effective decision support for whether the film and television project is established. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0039] Figure 1 This is a flowchart of an implementation method of a film and television project income prediction method in an embodiment of the present application;

[0040] Figure 2 This is a schematic diagram of the structure of a device for predicting income of a film and television project in an embodiment of the present application;

[0041] Figure 3 This is a schematic diagram of the structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0043] The core of this application is to provide a method for predicting the income of film and television projects. This method can be applied to scenarios such as film and television production companies, novel platforms, and video platforms that require film and television project evaluation and cast combination selection.

[0044] First, the historical performance data of each creative staff in the target cast combination corresponding to the target film and television project is obtained, and then the historical performance data of each creative staff in the target cast combination is statistically analyzed to obtain the characteristic value of each creative staff in each data dimension, and then the characteristic value of each creative staff in each data dimension is transformed according to the preset transformation rules to obtain the characteristic transformation value of each creative staff in each data dimension, and the evaluation score of each creative staff is determined based on the characteristic transformation value of each creative staff in each data dimension and the characteristic weight corresponding to the corresponding data dimension, and finally, the income of the target film and television project is predicted based on the evaluation score of each creative staff and the characteristic transformation value of each creative staff in each data dimension. By transforming the characteristic value of each data dimension, and then obtaining the evaluation score of the creative staff based on the characteristic transformation value and the corresponding characteristic weight, and then predicting the income of the target film and television project based on the evaluation score, etc., it can increase the comparability and interpretability between the features, improve the accuracy of the forecast of the income of the film and television project, and thus help to select a suitable cast combination, and provide effective decision support for whether the film and television project is established.

[0045] See also Figure 1 As shown, it is an implementation flow chart of a film and television project revenue prediction method provided by an embodiment of the present application, and the method may include the following steps:

[0046] S110: Obtain historical performance data of each main creative staff member in the target cast combination corresponding to the target film and television project.

[0047] In the embodiment of the present application, the target film and television project can be any film and television project that a film and television production company, a novel platform, a video platform, etc. want to establish. The target film and television project may have multiple cast combinations to choose from, and the target cast combination is any cast combination among the multiple cast combinations that can be selected for the target film and television project. The cast combination can be a combination of main creative personnel such as a director, a screenwriter, and an actor.

[0048] You can first obtain the historical performance data of each creative staff in the target cast combination corresponding to the target film and television project. For each creative staff, the historical performance data of the creative staff can be used to evaluate the business performance, market influence and potential value of the creative staff, which can include various types of data, such as work-related data, audience feedback data, box office and ratings data, social media data, awards and nomination data, professional evaluation data, role type data, ranking data, public opinion data, etc.

[0049] The historical performance data of each creative staff in the target cast combination corresponding to the target film and television project can be obtained through various search platforms, entertainment platforms, social media platforms, etc.

[0050] S120: Statistically analyze the historical performance data of each creative staff member in the target cast combination to obtain the characteristic value of each creative staff member in each data dimension.

[0051] In an embodiment of the present application, at least one data dimension can be pre-designed, and the data dimension is used to represent a specific aspect or attribute of the data. Optionally, according to the business significance of different data sources of historical performance data, multiple data dimensions such as trend, best, worst, and quantile can be designed to fully describe the historical performance of the main creative personnel. Taking the historical performance data of the main creative personnel on a social media platform as an example, the different data dimensions of the historical performance data of the main creative personnel on the social media platform, such as the score, the number of people who scored, the number of short comments, the ranking, etc., can be included to make the data dimensions more comprehensive as much as possible to make up for the shortcomings of insufficient data volume. The data dimensions designed for different types of main creative personnel may be the same or different. The data dimensions designed for the same type of main creative personnel are the same.

[0052] After obtaining the historical performance data of each main creative staff in the target cast combination corresponding to the target film and television project, the historical performance data of each main creative staff in the target cast combination can be further statistically analyzed to obtain the characteristic value of each main creative staff in each data dimension.

[0053] For each creative staff, statistical analysis is performed on the historical performance data of the creative staff, which may include using different statistical analysis methods to perform statistical analysis on different types of data included in the historical performance data of the creative staff. The statistical analysis methods may include number statistics, average statistics, quantile statistics, median statistics, trend change value statistics, percentage statistics, etc.

[0054] For example, through index interfaces such as search platforms, entertainment platforms, and social media platforms, the daily search values ​​of each creative staff in the target cast combination can be obtained. By counting the search values ​​in the past two years, the quantiles and trend change values ​​of the search values ​​of each creative staff in the target cast combination can be obtained, which can make up for the disadvantage of a single dimension of search data. Among them, the trend change value can be the ratio of the average search value of the previous period to the average search value of the next period. For example, the trend change value = the average search value of the past 3 months / the average search value of the past 3-6 months.

[0055] Examples of the characteristic values ​​of the main creative personnel in the target cast combination in multiple data dimensions are shown in Table 1:

[0056]

[0057] Table 1

[0058] S130: According to a preset transformation rule, the characteristic value of each main creative person in each data dimension is transformed respectively to obtain the characteristic transformation value of each main creative person in each data dimension.

[0059] Obtain the historical performance data of each main creative staff in the target cast combination corresponding to the target film and television project, perform statistical analysis on the historical performance data of each main creative staff, and obtain the characteristic value of each main creative staff in each data dimension. Then, according to the preset transformation rules, transform the characteristic value of each main creative staff in each data dimension to obtain the characteristic transformation value of each main creative staff in each data dimension.

[0060] The transformation rules can be set and adjusted according to the actual situation. According to the preset transformation rules, the characteristic values ​​of each data dimension of each main creative personnel are transformed to adjust the data distribution and unify the dimensions.

[0061] Optionally, the transformation rule may include a logarithmic transformation, such as a ln logarithmic transformation, that is, performing a logarithmic transformation on the characteristic value of each main creative staff in each data dimension to obtain a characteristic transformation value of each main creative staff in each data dimension.

[0062] An example of the feature transformation value of each creative staff in the target cast combination in each data dimension is shown in Table 2:

[0063]

[0064] Table 2

[0065] Performing a logarithmic transformation on the eigenvalues ​​of each creative staff in each data dimension helps to compress the scale, change the data distribution, and reduce the impact of abnormal values ​​such as maximum and minimum values ​​in the eigenvalues.

[0066] S140: Determine an evaluation score for each main creative staff member based on the feature transformation value of each main creative staff member in each data dimension and the feature weight corresponding to the corresponding data dimension.

[0067] In the embodiment of the present application, different feature weights can be set for different data dimensions based on data analysis and expert experience.

[0068] Examples of feature weights corresponding to each data dimension are shown in Table 3:

[0069] Data Dimensions Feature weights Number of TV dramas starred 5 TV series weighted rating (by role contribution) 1 0 Weighted ratings of TV series (by number of ratings) 5 TV series maximum rating 5 Maximum number of people who rated a TV series 2 The proportion of short reviews 5 Total number of people who rated TV series 2 Recent TV drama rating trends 5 The most recent work is the number of copies this year 2

[0070] Table 3

[0071] After obtaining the feature transformation value of each creative staff in the target cast combination in each data dimension, the evaluation score of each creative staff can be determined based on the feature transformation value of each creative staff in each data dimension and the feature weight corresponding to the corresponding data dimension.

[0072] S150: Based on the evaluation score of each main creative staff and the feature transformation value of each main creative staff in each data dimension, predict the income of the target film and television project when using the target cast combination.

[0073] In an embodiment of the present application, after obtaining the feature transformation value of each main creative staff in the target cast combination in each data dimension and determining the evaluation score of each main creative staff in the target cast combination, the revenue of the target film and television project when using the target cast combination is predicted based on the feature transformation value of each main creative staff in each data dimension and the evaluation score of each main creative staff.

[0074] Applying the method provided by the embodiment of the present application, first obtain the historical performance data of each creative staff in the target cast combination corresponding to the target film and television project, then statistically analyze the historical performance data of each creative staff in the target cast combination, obtain the characteristic value of each creative staff in each data dimension, and then transform the characteristic value of each creative staff in each data dimension according to the preset transformation rule, obtain the characteristic transformation value of each creative staff in each data dimension, determine the evaluation score of each creative staff based on the characteristic transformation value of each creative staff in each data dimension and the characteristic weight corresponding to the corresponding data dimension, and finally predict the income of the target film and television project under the condition of using the target cast combination according to the evaluation score of each creative staff and the characteristic transformation value of each creative staff in each data dimension. By transforming the characteristic value of each data dimension, and then obtaining the evaluation score of the creative staff based on the characteristic transformation value and the corresponding characteristic weight, and then predicting the income of the target film and television project based on the evaluation score, etc., it is possible to increase the comparability and interpretability between the features, improve the accuracy of the forecast of the income of the film and television project, thereby helping to select a suitable cast combination and provide effective decision support for whether the film and television project is established.

[0075] In some embodiments of the present application, step S130 transforms the feature value of each creative person in each data dimension according to a preset transformation rule to obtain the feature transformation value of each creative person in each data dimension, which may include the following steps:

[0076] For each data dimension of each main creator, determine the ratio of the characteristic value of the current main creator in the current data dimension to the benchmark value of the current data dimension, wherein the current main creator is the main creator targeted by the current operation, and the current data dimension is the data dimension targeted by the current operation;

[0077] The natural logarithm value of the ratio is determined as the feature transformation value of the current main creator in the current data dimension.

[0078] For the convenience of description, the above two steps are combined for explanation.

[0079] In the embodiment of the present application, after obtaining the historical performance data of each creative staff in the target cast combination corresponding to the target film and television project, the base value of each data dimension can be obtained by performing descriptive statistical analysis on the historical performance data of each creative staff. Optionally, a set quantile can be used as the base value, such as the 0.75 quantile.

[0080] Examples of benchmark values ​​corresponding to each data dimension are shown in Table 4:

[0081] Data Dimensions Benchmark value Number of TV dramas starred 6 TV series weighted rating (by role contribution) 8 Weighted ratings of TV series (by number of ratings) 8.5 TV series maximum rating 2 Maximum number of people who rated a TV series 50000 The proportion of short reviews 0.3 Total number of TV series ratings 100000 Recent TV drama rating trends 1 The most recent work is the number of copies this year 3

[0082] Table 4

[0083] After statistically analyzing the historical performance data of each creative staff in the target cast combination and obtaining the characteristic value of each creative staff in each data dimension, the ratio of the characteristic value of the current creative staff in the current data dimension to the benchmark value of the current data dimension can be determined for each data dimension of each creative staff. The current creative staff refers to the creative staff targeted by the current operation, and the current data dimension refers to the data dimension targeted by the current operation.

[0084] After obtaining the ratio of the characteristic value of the current main creator in the current data dimension to the reference value of the current data dimension, the natural logarithm value of the ratio can be determined as the characteristic transformation value of the current main creator in the current data dimension.

[0085] Optionally, the feature transformation value of the current main creative person in the current data dimension=ln(the feature value of the current main creative person in the current data dimension / the reference value of the current data dimension).

[0086] The ratio of the main creators' eigenvalues ​​in each data dimension to the baseline value of the corresponding data dimension is logarithmically transformed. This helps to eliminate the dimensions of the eigenvalues ​​of different data dimensions, adjust the data distribution, and help reduce the impact of abnormal values ​​such as maximum and minimum values ​​in the eigenvalues.

[0087] In some embodiments of the present application, step S140 determines the evaluation score of each main creator based on the feature transformation value of each main creator in each data dimension and the feature weight corresponding to the corresponding data dimension, and may include the following steps:

[0088] For each data dimension of each main creator, the product of the feature transformation value of the current main creator in the current data dimension and the feature weight corresponding to the current data dimension is determined as the feature value score of the current main creator in the current data dimension, wherein the current main creator is the main creator targeted by the current operation, and the current data dimension is the data dimension targeted by the current operation;

[0089] Based on the characteristic value scores of the current main creative personnel in each data dimension, the evaluation scores of the current main creative personnel are obtained.

[0090] For the convenience of description, the above two steps are combined for explanation.

[0091] In an embodiment of the present application, the characteristic value of each creative staff in each data dimension is transformed respectively. After obtaining the characteristic transformation value of each creative staff in each data dimension, for each data dimension of each creative staff, the characteristic transformation value of the current creative staff in the current data dimension can be multiplied by the characteristic weight corresponding to the current data dimension to obtain the product of the two, and the product is determined as the characteristic value score of the current creative staff in the current data dimension. The current creative staff refers to the creative staff targeted by the current operation, and the current data dimension refers to the data dimension targeted by the current operation.

[0092] For example, if the feature transformation value of creative staff A in a certain data dimension is X, and the feature weight corresponding to the data dimension is w, then the feature value score of the creative staff in the data dimension is w*X.

[0093] After determining the characteristic value score of the current main creative personnel in each data dimension, the evaluation score of the current main creative personnel can be obtained based on the characteristic value score of the current main creative personnel in each data dimension.

[0094] Optionally, the characteristic value scores of the current main creator in each data dimension may be added together to obtain the evaluation score of the current main creator.

[0095] By assigning corresponding feature weights to the feature values ​​of each main creator in each data dimension, the feature values ​​can be made more comparable and the accuracy of the evaluation score of each main creator can be improved.

[0096] In some embodiments of the present application, after determining the product of the feature transformation value of the current main creator in the current data dimension and the feature weight corresponding to the current data dimension as the feature value score of the current main creator in the current data dimension and before obtaining the evaluation score of the current main creator based on the feature value score of the current main creator in each data dimension, the method may further include the following steps:

[0097] Amplify the characteristic value score of the current main creator in the current data dimension.

[0098] In an embodiment of the present application, after the product of the feature transformation value of the current main creative personnel in the current data dimension and the feature weight corresponding to the current data dimension is determined as the feature value score of the current main creative personnel in the current data dimension, the feature value score of the current main creative personnel in the current data dimension can be amplified.

[0099] Optionally, the feature value score of the current main creator in the current data dimension can be multiplied by the magnification factor, and the amplified feature value score is used to replace the previous feature value score. For example, the feature value score of the current main creator in the current data dimension = the feature transformation value of the current main creator in the current data dimension * the feature weight corresponding to the current data dimension * factor, where factor represents the magnification factor.

[0100] After the characteristic value score of the current main creative personnel in each data dimension is amplified, the evaluation score of the current main creative personnel can be further obtained based on the amplified characteristic value score of the current main creative personnel in each data dimension.

[0101] Amplifying the eigenvalue scores of each creative staff in each data dimension can increase the magnitude of the data, making it more beautiful and intuitive.

[0102] An example of the eigenvalue score of each creative staff in each data dimension is shown in Table 5:

[0103]

[0104] Table 5

[0105] In some embodiments of the present application, after determining the evaluation score of each creative staff member based on the feature transformation value of each creative staff member in each data dimension and the feature weight corresponding to the corresponding data dimension in step S140, and before predicting the income of the target film and television project when the target cast combination is used in step S150 based on the evaluation score of each creative staff member and the feature transformation value of each creative staff member in each data dimension, the method may further include the following steps:

[0106] Adjust the evaluation score of each creative staff member to a positive number.

[0107] In an embodiment of the present application, after determining the evaluation score of each main creative staff member based on the feature transformation value of each main creative staff member in each data dimension and the feature weight corresponding to the corresponding data dimension, the evaluation score of each main creative staff member can be adjusted to a positive number, such as adding an offset to the evaluation score of each main creative staff member to avoid errors caused by negative number calculations. The size relationship of the evaluation scores of different main creative staff members after adjustment is the same as the size relationship of the evaluation scores before adjustment.

[0108] For example, the evaluation score of main creative staff A is 27520, and the evaluation score of main creative staff B is 28075.

[0109] After adjusting the evaluation score of each main creative staff member, the revenue of the target film and television project when using the target cast combination can be further predicted based on the adjusted evaluation score of each main creative staff member and the characteristic transformation value of each main creative staff member in each data dimension.

[0110] The embodiment of the present application draws on the odds ratio (odds), weight of evidence (woe), and credit scoring methods of financial risk control, and uses the eigenvalue / benchmark value to approximately replace the odds ratio, and performs a ln logarithmic transformation to approximately replace the woe indicator. Then, feature weights are assigned through expert experience to achieve dimensional unification of eigenvalues, making eigenvalues ​​of different data dimensions comparable.

[0111] Among them, the odds ratio can represent how much greater the probability p of an event occurring is than the probability 1-p of not occurring, odds = p / (1-p). According to the logistic regression (logit) model in statistics, there is the following formula:

[0112]

[0113] Among them, β0 represents the intercept term of the linear equation f(x), which is a constant obtained by model training, X1~X k Represent the original eigenvalues ​​of k data dimensions, β1~β k They represent the feature weights corresponding to the k data dimensions respectively.

[0114] When training the model, in order to increase the stability of the model, you can choose to use X1~X k The woe value corresponding to the original eigenvalue replaces the original eigenvalue.

[0115] Definition of woe: The original eigenvalues ​​of each data dimension are segmented (such as segmented according to equal frequency and equal distance). For any original eigenvalue, woe of the segment where the original eigenvalue is located = ln(odds of the current segment) = ln(p of the current segment / (1-p of the current segment)), and the current segment is the segment where the original eigenvalue is located.

[0116] The formula corresponding to the above logit model can be further written as:

[0117]

[0118] Among them, k represents the total number of data dimensions, n represents the number of feature segments, j corresponds to k, representing the specific data dimension, and i corresponds to n, representing the i-th segment of the j-th data dimension.

[0119] For the main creative personnel, the p in the above formula refers to the probability that this main creative personnel is a good main creative personnel. Under normal circumstances, it is necessary to manually mark the good and bad of a large number of film and television projects and a large number of main creative personnel at different times, which is time-consuming and the effect may not be much higher. Therefore, in order to quickly develop the iteration speed, the embodiment of the present application uses a similar corresponding indicator instead, that is, ln (eigenvalue / baseline value) is used instead of woe in the segment where the original eigenvalue is located.

[0120] Based on this, the formula corresponding to the above logit model can be further written as:

[0121]

[0122] At this point, ln(odds) can be converted into an evaluation score according to the credit scoring formula:

[0123]

[0124] Among them, the factor value can be determined according to the result of ln(odds), which is used to amplify the eigenvalue score, widen the data gap, make the data magnitude range reasonable, and make it easier to understand in subsequent analysis, thus generating information gap;

[0125] The offset value is determined by the result of factor*ln(odds). It is mainly used to make the evaluation score > 0 to avoid calculation errors caused by negative numbers in subsequent weighted calculations.

[0126] In some embodiments of the present application, step S150 predicts the revenue of the target film and television project when the target cast combination is used according to the evaluation score of each creative staff and the feature transformation value of each creative staff in each data dimension, which may include the following steps:

[0127] Input the evaluation score of each main creator and the feature transformation value of each main creator in each data dimension into the pre-trained prediction model;

[0128] Based on the output of the prediction model, predict the revenue of the target film and television project using the target cast combination.

[0129] For the convenience of description, the above steps are combined for explanation.

[0130] In an embodiment of the present application, a prediction model can be pre-trained, the input of the prediction model is the evaluation score of the main creative personnel and the feature transformation value of the main creative personnel in each data dimension, and the output of the prediction model is the project income.

[0131] The evaluation score of each main creative staff in the target cast combination and the feature transformation value of each main creative staff in each data dimension are input into the prediction model. According to the output of the prediction model, the predicted income of the target film and television project when using the target cast combination can be obtained.

[0132] Revenue forecasting through forecasting models helps improve forecasting efficiency.

[0133] In addition, after obtaining the actual revenue of the target film and television project using the target cast combination, the relevant data can be added to the training data to optimize the prediction model to improve the accuracy of the prediction model.

[0134] In some embodiments of the present application, after predicting the revenue of the target film and television project when the target cast combination is used in step S150, the method may further include the following steps:

[0135] Compare the revenue of the target film and television project using the target cast combination with the budget of the target film and television project;

[0136] Based on the comparison results, determine whether to select the target cast combination for the target film and television project.

[0137] For the convenience of description, the above two steps are combined for explanation.

[0138] After predicting the revenue of the target film and television project when the target cast combination is used, the predicted revenue of the target film and television project when the target cast combination is used can be compared with the budget of the target film and television project. Based on the comparison result, it can be determined whether the target film and television project selects the target cast combination.

[0139] Alternatively, if the predicted income corresponding to the target cast combination is much higher than the budget, it can be considered that the target film and television project will bring better returns by using the target cast combination, and the target cast combination can be selected for the target film and television project. On the contrary, if the predicted income corresponding to the target cast combination is lower than or equal to the budget, it can be considered that the target film and television project will not bring better returns by using the target cast combination, and it can be ignored. This helps to select a suitable cast combination.

[0140] In some embodiments of the present application, the target cast combination is any one of a plurality of cast combinations corresponding to the target film and television project, and the method may further include the following steps:

[0141] After obtaining the predicted revenue of the target film or television project when using each of the multiple cast combinations, the cast combination selected for the target film or television project is determined based on the predicted revenue of the target film or television project when using each cast combination.

[0142] In the embodiment of the present application, the target cast combination is any one of the multiple cast combinations corresponding to the target film and television project. According to the technical solution provided in the embodiment of the present application, the predicted income of the target film and television project when each cast combination is used can be obtained. Based on the obtained predicted income, the cast combination selected for the target film and television project can be determined.

[0143] Optionally, the cast combination with the highest predicted revenue may be selected as the cast combination ultimately used in the target film and television project.

[0144] Selecting the cast combination based on the predicted revenue corresponding to each cast combination will help choose the right cast combination and bring better returns.

[0145] The technical solution provided in the embodiments of the present application enhances the comparability between main creative personnel, enhances information gain, stability, and ranking, and can provide strong support for decision-making such as project establishment and production.

[0146] Corresponding to the above method embodiment, the embodiment of the present application also provides a film and television project revenue prediction device. The film and television project revenue prediction device described below and the film and television project revenue prediction method described above can be referenced to each other.

[0147] See also Figure 2 As shown, the film and television project income prediction device 200 may include the following modules:

[0148] An acquisition module 210 is used to acquire historical performance data of each main creative staff member in a target cast combination corresponding to a target film and television project;

[0149] Analysis module 220, for performing statistical analysis on the historical performance data of each creative staff in the target cast combination, and obtaining the characteristic value of each creative staff in each data dimension;

[0150] The transformation module 230 is used to transform the characteristic value of each main creator in each data dimension according to a preset transformation rule to obtain the characteristic transformation value of each main creator in each data dimension;

[0151] A determination module 240, for determining an evaluation score for each main creative staff member based on a feature transformation value of each main creative staff member in each data dimension and a feature weight corresponding to the corresponding data dimension;

[0152] The prediction module 250 is used to predict the revenue of the target film and television project when the target cast combination is used based on the evaluation score of each main creative staff and the feature transformation value of each main creative staff in each data dimension.

[0153] By using the device provided in the embodiment of the present application, the historical performance data of each creative staff in the target cast combination corresponding to the target film and television project is first obtained, and then the historical performance data of each creative staff in the target cast combination is statistically analyzed to obtain the characteristic value of each creative staff in each data dimension, and then the characteristic value of each creative staff in each data dimension is transformed according to the preset transformation rule to obtain the characteristic transformation value of each creative staff in each data dimension, and the evaluation score of each creative staff is determined based on the characteristic transformation value of each creative staff in each data dimension and the characteristic weight corresponding to the corresponding data dimension, and finally, according to the evaluation score of each creative staff and the characteristic transformation value of each creative staff in each data dimension, the income of the target film and television project is predicted when the target cast combination is used. By transforming the characteristic value of each data dimension, and then obtaining the evaluation score of the creative staff based on the characteristic transformation value and the corresponding characteristic weight, and then predicting the income of the target film and television project based on the evaluation score, etc., the comparability and interpretability between the features can be increased, and the accuracy of the prediction of the income of the film and television project can be improved, so as to help select a suitable cast combination and provide effective decision support for whether the film and television project is established.

[0154] In some embodiments of the present application, the transformation module 230 is specifically configured to:

[0155] For each data dimension of each main creator, determine the ratio of the characteristic value of the current main creator in the current data dimension to the benchmark value of the current data dimension, wherein the current main creator is the main creator targeted by the current operation, and the current data dimension is the data dimension targeted by the current operation;

[0156] The natural logarithm value of the ratio is determined as the feature transformation value of the current main creator in the current data dimension.

[0157] In some embodiments of the present application, the determination module 240 is specifically configured to:

[0158] For each data dimension of each main creator, the product of the feature transformation value of the current main creator in the current data dimension and the feature weight corresponding to the current data dimension is determined as the feature value score of the current main creator in the current data dimension, wherein the current main creator is the main creator targeted by the current operation, and the current data dimension is the data dimension targeted by the current operation;

[0159] Based on the characteristic value scores of the current main creative personnel in each data dimension, the evaluation scores of the current main creative personnel are obtained.

[0160] In some embodiments of the present application, the film and television project income prediction device 200 further includes a first processing module, which is used to:

[0161] After the product of the feature transformation value of the current main creative personnel in the current data dimension and the feature weight corresponding to the current data dimension is determined as the feature value score of the current main creative personnel in the current data dimension and before the evaluation score of the current main creative personnel is obtained based on the feature value scores of the current main creative personnel in each data dimension, the feature value scores of the current main creative personnel in the current data dimension are amplified.

[0162] In some embodiments of the present application, the film and television project income prediction device 200 further includes a second processing module for:

[0163] After determining the evaluation score of each main creative staff member based on the feature transformation value of each main creative staff member in each data dimension and the feature weight corresponding to the corresponding data dimension, and before predicting the income of the target film and television project when using the target cast combination according to the evaluation score of each main creative staff member and the feature transformation value of each main creative staff member in each data dimension, the evaluation score of each main creative staff member is adjusted to a positive number.

[0164] In some embodiments of the present application, the prediction module 250 is specifically used to:

[0165] Input the evaluation score of each main creator and the feature transformation value of each main creator in each data dimension into the pre-trained prediction model;

[0166] Based on the output of the prediction model, predict the revenue of the target film and television project using the target cast combination.

[0167] In some embodiments of the present application, the target cast combination is any one of the multiple cast combinations corresponding to the target film and television project, and the film and television project revenue prediction device 200 further includes a selection module for:

[0168] After obtaining the predicted revenue of the target film or television project when using each of the multiple cast combinations, the cast combination selected for the target film or television project is determined based on the predicted revenue of the target film or television project when using each cast combination.

[0169] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0170] The present application also provides an electronic device, such as Figure 3 As shown, it includes a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302, and the memory 303 communicate with each other through the communication bus 304.

[0171] Memory 303, used for storing computer programs;

[0172] The processor 301 is used to execute the program stored in the memory 303 to implement the following steps:

[0173] Obtain the historical performance data of each main creative staff in the target cast combination corresponding to the target film and television project;

[0174] Conduct statistical analysis on the historical performance data of each creative staff in the target cast combination to obtain the characteristic value of each creative staff in each data dimension;

[0175] According to the preset transformation rules, the characteristic values ​​of each main creator in each data dimension are transformed respectively to obtain the characteristic transformation value of each main creator in each data dimension;

[0176] Determine the evaluation score of each main creator based on the feature transformation value of each main creator in each data dimension and the feature weight corresponding to the corresponding data dimension;

[0177] Based on the evaluation score of each main creative staff and the feature transformation value of each main creative staff in each data dimension, the revenue of the target film and television project when using the target cast combination is predicted.

[0178] The communication bus 304 mentioned in the above terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus 304 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0179] The communication interface 302 is used for communication between the terminal and other devices.

[0180] The memory 303 may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory 303 may also be at least one storage device located away from the aforementioned processor.

[0181] The above-mentioned processor 301 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 gates or transistor logic devices, discrete hardware components.

[0182] In another embodiment provided in the present application, a computer-readable storage medium is also provided, in which instructions are stored. When the computer-readable storage medium is run on a computer, the computer executes the steps of the film and television project revenue prediction method in any of the above embodiments.

[0183] In another embodiment provided in the present application, a computer program product comprising instructions is also provided, which, when executed on a computer, enables the computer to execute the steps of any of the film and television project revenue prediction methods in the above-mentioned embodiments.

[0184] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loading and executing computer program instructions on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. 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 a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server, a data center, etc. that contains one or more available media integrated. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk Solid State Disk (SSD)), etc.

[0185] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0186] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0187] The above are only preferred embodiments of the present application and are not intended to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application are included in the protection scope of the present application.

Claims

1. A method for predicting income of film and television projects, It is characterized in that include: Obtain the historical performance data of each main creative staff in the target cast combination corresponding to the target film and television project; Performing statistical analysis on the historical performance data of each creative staff in the target cast combination to obtain the characteristic value of each creative staff in each data dimension; According to the preset transformation rules, the characteristic values ​​of each main creator in each data dimension are transformed respectively to obtain the characteristic transformation value of each main creator in each data dimension; Determine the evaluation score of each main creator based on the feature transformation value of each main creator in each data dimension and the feature weight corresponding to the corresponding data dimension; According to the evaluation score of each main creative staff and the feature transformation value of each main creative staff in each data dimension, the income of the target film and television project when using the target cast combination is predicted.

2. The method according to claim 1, It is characterized in that According to the preset transformation rules, the characteristic values ​​of each main creator in each data dimension are transformed respectively to obtain the characteristic transformation value of each main creator in each data dimension, including: For each data dimension of each main creator, determine the ratio of the characteristic value of the current main creator in the current data dimension to the reference value of the current data dimension, wherein the current main creator is the main creator targeted by the current operation, and the current data dimension is the data dimension targeted by the current operation; The natural logarithm value of the ratio is determined as the feature transformation value of the current creative person in the current data dimension.

3. The method according to claim 1, It is characterized in that The step of determining the evaluation score of each main creator based on the feature transformation value of each main creator in each data dimension and the feature weight corresponding to the corresponding data dimension includes: For each data dimension of each main creator, the product of the feature transformation value of the current main creator in the current data dimension and the feature weight corresponding to the current data dimension is determined as the feature value score of the current main creator in the current data dimension, wherein the current main creator is the main creator targeted by the current operation, and the current data dimension is the data dimension targeted by the current operation; Based on the characteristic value score of the current main creative personnel in each data dimension, an evaluation score of the current main creative personnel is obtained.

4. The method according to claim 3, It is characterized in that After determining the product of the feature transformation value of the current main creator in the current data dimension and the feature weight corresponding to the current data dimension as the feature value score of the current main creator in the current data dimension and before obtaining the evaluation score of the current main creator based on the feature value score of the current main creator in each data dimension, the method further includes: The characteristic value score of the current main creative personnel in the current data dimension is amplified.

5. The method according to claim 1, It is characterized in that After determining the evaluation score of each main creative staff member based on the feature transformation value of each main creative staff member in each data dimension and the feature weight corresponding to the corresponding data dimension, and before predicting the income of the target film and television project when the target cast combination is used according to the evaluation score of each main creative staff member and the feature transformation value of each main creative staff member in each data dimension, the method further includes: Adjust the evaluation score of each creative staff member to a positive number.

6. The method according to claim 1, It is characterized in that The step of predicting the revenue of the target film and television project when the target cast combination is used according to the evaluation score of each main creative staff and the feature transformation value of each main creative staff in each data dimension includes: Input the evaluation score of each main creator and the feature transformation value of each main creator in each data dimension into the pre-trained prediction model; Based on the output of the prediction model, the revenue of the target film and television project when using the target cast combination is predicted.

7. The method according to any one of claims 1 to 6, It is characterized in that The target cast combination is any one of a plurality of cast combinations corresponding to the target film and television project, and the method further includes: After obtaining the predicted revenue of the target film and television project when using each of the multiple cast combinations, the cast combination selected for the target film and television project is determined based on the predicted revenue of the target film and television project when using each cast combination.

8. A device for predicting income of film and television projects, It is characterized in that include: An acquisition module is used to acquire the historical performance data of each main creative staff in the target cast combination corresponding to the target film and television project; An analysis module is used to perform statistical analysis on the historical performance data of each main creative staff in the target cast combination to obtain the characteristic value of each main creative staff in each data dimension; A transformation module is used to transform the characteristic value of each main creator in each data dimension according to a preset transformation rule to obtain the characteristic transformation value of each main creator in each data dimension; A determination module, used to determine the evaluation score of each main creative staff member based on the feature transformation value of each main creative staff member in each data dimension and the feature weight corresponding to the corresponding data dimension; The prediction module is used to predict the income of the target film and television project when using the target cast combination based on the evaluation score of each main creative staff and the feature transformation value of each main creative staff in each data dimension.

9. An electronic device, It is 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 through the communication bus; Memory, used to store computer programs; The processor is used to implement the steps of the film and television project revenue prediction method as described in any one of claims 1 to 7 when executing the program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the program is executed by a processor, the steps of the method for predicting the income of a film or television project as described in any one of claims 1 to 7 are implemented.

11. A computer program product, comprising computer instructions, which are stored in a computer-readable storage medium and are suitable for being read and executed by a processor, so that an electronic device having the processor performs the steps of the film and television project revenue prediction method as described in any one of claims 1 to 7.

Images