Exploration of the Development Path of Network Novel Film and TV Adaptation Drama in the Background of Internet

Web writers upload and publish their works and disseminate them through the Internet, and their audiences are also mainly from the Internet. Under the background of the increasing popularization of the Internet, more and more people start to read and become addicted to web novels and become loyal fans of web novels [1-3]. When the network novels are made into film and television dramas, this group of “loyal fans” is transformed into the audience of adapted dramas, and the popularity of network novels has also brought the fire of adapted dramas.

In the context of the Internet era, network literature is mixed, film and television adaptation production there are many problems, but there are still many film and television adaptation of outstanding talents and works stand out, the film and television drama production team has also stepped people standardized and professional [4-6]. Film and television adaptation of network literature from the creation of the novel itself to the sale of copyright, and then to the film and television script adaptation, as well as the specifications of the filming process, post-production technology, which has formed a complete system, it can be seen that the trend of network literature adapted into movies and television dramas has gradually matured and standardized [7-9]. Network novels and film and television adaptations have been a topic of great concern in the cultural industry since their inception. With the arrival of the era of integrated media, the interaction between network novels and film and television adaptations has become stronger and stronger, showing a diversified development trend [10-12]. The interaction between the originality and adaptation of network novels, as well as the interpenetration between network novels and film and television dramas, has triggered extensive attention and research in both academia and industry. The study of the status quo, characteristics and its development trend of film and television adaptations of network novels from the perspective of melting media has certain value, which is of great significance to the practice of cultural industry, and helps to improve the quality and market influence of film and television adaptations, thus promoting the prosperity of cultural industry [13-15].

In this study, the hierarchical analysis method is firstly applied to the evaluation process of network literature IP film and television adaptation rights, and a judgment matrix is formed based on the scores, and the eigenvalues and eigenvectors of the matrix are derived. The eigenvectors are normalized to determine the degree of contribution of the Internet literature IP to the copyright value of its film adaptation, so as to estimate the value of the film and television adaptation right of the Internet literature IP. Secondly, the XGBoost algorithm in machine learning is utilized to reasonably predict the revenue of network novel film and television adaptations. And the indicators are borrowed to compare the model’s fitting superiority and prediction effect, and the model with the smallest error is adopted for predicting the revenue of network literature adaptations. Finally, the network novel Qing Yu Nian, which was adapted into a movie and television drama, is taken as a research case to evaluate and validate the model and parameter ideas. In this way, the development path of network novel film and television adaptations is proposed from several aspects, such as potential users, user purchase, and conversion rate.

2

Assessment of the value of film and television adaptation rights for web novels

2.1

The Value Path of Network Novel Film and Television Adaptation Rights

Excellent online literature can not only provide content support for its subsequent conversion products, but also its own large number of readers can make its conversion products have a certain fan base at the very beginning. It can be said that the vitality of online literature has contributed to the prosperous development of the entire entertainment industry.

Since the value of network literature film and television adaptation rights cannot be calculated independently, it can only be calculated by divesting the value of film and television adaptation rights in the overall value. Movie and TV adaptation right is a “crossroads” between online literature and movie and TV drama works. From the perspective of network literary works, the right of film and television adaptation is one of the rights belonging to the copyright of network literary works, and from the perspective of film and television drama works, the right of film and television adaptation is one of the factors affecting the income of film and television drama. Therefore, there are two value paths to assess the value of film and television adaptation right, and the value path of network literature film and television adaptation right is shown in Figure 1.

2.2

Value Assessment of Film and TV Adaptations Based on AHP

The principle of Analytic Hierarchy Process (AHP) is to compare different elements within a hierarchy, score them according to their relative importance, form a judgment matrix based on the scores, find the eigenvalues and eigenvectors of the matrix, and normalize the eigenvectors to get the weight of the elements in that hierarchy relative to the previous hierarchy. Based on this principle, the contribution of each element within the criterion layer to the target layer can be obtained. By applying the hierarchical analysis method to the evaluation process of the network literature IP film and television adaptation right, it is possible to determine the degree of contribution of the network literature IP to the copyright value of its film adaptation, and thus get the value of the network literature IP film and television adaptation right.

2.2.1

Hierarchical modeling

In this paper, the hierarchical analysis method is not used to make decisions, but to determine the extent to which an element in the criterion layer contributes to the overall goal. According to the needs of assessment, the hierarchical model of this paper is set as objective layer, criterion layer and indicator layer. The external and internal factors affecting the box office revenue of movies have been analyzed in the previous section. On this basis, this paper divides it into 8 indicators to construct a hierarchical analysis model, and the hierarchical analysis structure of the network literature IP film and television adaptation rights is shown in Figure 2. Among them, the pairwise comparison matrix is a comparison that indicates the relative importance of all the factors, for a certain factor in the previous layer.

1)

Construct judgment matrix

If the target level is $Z$ and the criterion level is $A$ , and there are 5 elements in $A$ , the relative importance scoring table of all elements in $A$ with respect to $Z$ is shown in Table 1. Where $a i j$ indicates the importance score of the $i$ th element in $A$ relative to the $j$ th element.

Based on the scoring table, the following pairwise comparison matrix is obtained: (1) $A = [\begin{matrix} a_{11} & a_{12} & a_{13} & a_{14} & a_{15} \\ a_{21} & a_{22} & a_{23} & a_{24} & a_{25} \\ a_{31} & a_{32} & a_{33} & a_{34} & a_{35} \\ a_{41} & a_{42} & a_{43} & a_{44} & a_{45} \\ a_{51} & a_{52} & a_{53} & a_{54} & a_{55} \end{matrix}]$ where $a i i = 1$ , $a i j = 1 / a j i$

2)

Calculate the single-layer weight vector

A matrix is said to be a positive inverse matrix if each element of the matrix $a i j > 0$ , and satisfies $a i j * a j i = 1$ . In hierarchical analysis, the judgment matrices constructed are all positive inverse matrices. If the positive inverse matrix satisfies $a i j * a j k = a i k$ , it is said to be a consistency matrix.

If there exists a number $λ$ and a nonzero column vector $x$ such that $A x = λ x$ holds, then $λ$ is said to be an eigenvalue of the matrix $A$ and $x$ is an eigenvector. According to the nature of consistency matrices, if the constructed judgment matrix is a consistency matrix, the normalized eigenvector $W = {(W_{1}, W_{2}, \dots \dots .. W_{n})}^{T}$ corresponding to its largest eigenvalue $λ_{\max}$ is the single-layer weight vector. The parameters in the vector then correspond to the degree to which each element of the layer contributes to a particular element of the previous layer, e.g., $A 1$ contributes $w_{1}$ to $Z$ .

In general, most of the judgment matrices actually constructed are not consistency matrices, but they are still treated as consistency matrices, and the normalized eigenvector corresponding to the largest eigenroot is taken as the weight vector. However, whether the weight vector is reasonable or not, a consistency test is needed.

3)

Consistency test

The indicators of consistency test are: (2) $C I = \frac{λ_{\max} - n}{n - 1}$ where $λ_{\max}$ is the largest eigenvalue of the judgment matrix and $n$ is the order of the judgment matrix. $C I = 0$ , indicates that there is complete consistency. $C I$ is close to 0, there is satisfactory consistency. The larger $C I$ is, the more serious the inconsistency is. To measure the magnitude of $C I$ , a random consistency index RI is introduced.

Randomly constructing 500 judgment matrices $A 1, A 2, \dots \dots, A 500$ , the consistency indicator $C I 1, C I 2, \dots \dots, C I 500$ can be obtained.

The formula for the random consistency indicator $R I$ is: (3) $R I = \frac{C I 1 + C I 2 + \dots + C I 500}{500} = \frac{\frac{λ 1 + λ 2 + \dots + λ 500}{500} - n}{n - 1}$

Define the consistency ratio: (4) $C R = \frac{C I}{R I}$

When the consistency ratio is $C R < 0.1$ , it means that the judgment matrix $A$ passes the consistency test and its normalized eigenvectors can be used as the weight vectors. Otherwise, it is necessary to re-score the elements in criterion layer $A$ to construct a new judgment matrix.

Table 1.

The analytic hierarchy process (AHP) scores

Z	A1	A2	A3	A4	A5
A1	a11	a12	a13	a14	a15
A2	a21	a22	a23	a24	a25
A3	a31	a32	a33	a34	a35
A4	a41	a42	a43	a44	a45
A5	a51	a52	a53	a54	a55

2.2.2

Weight vectors of judgment matrices

In this paper, the copyright value of the movie adaptation of online literature IP is taken as the target layer, and the criterion layer includes internal factors and external factors affecting the copyright value. The internal factors include actors, directors, screenwriters, online literature IP and movie genres, and the external factors include movie attention, release schedule and distribution companies, and these eight factors are taken as the index layer, and the results of the division are shown in Table 2. According to the delineated hierarchy, let all elements within each layer be compared two by two, design questionnaires, ask experts to score, and construct judgment matrices based on the scores. The judgment matrix B1 obtained according to the experts’ scores is: (5) $B 1 = (\begin{matrix} 1 & 1 / 5 & 6 & 1 / 2 & 4 \\ 5 & 1 & 7 & 3 & 5 \\ 1 / 6 & 1 / 7 & 1 & 1 / 5 & 2 \\ 2 & 1 / 3 & 5 & 1 & 5 \\ 1 / 4 & 1 / 5 & 1 / 2 & 1 / 5 & 1 \end{matrix})$

Table 2.

Division results

Project	The Standard Layer	Index Level
Internet literature IP adaptation of film copyright value (A)	Internal factors (B1)	Actor (C1)
		Director (C2)
		Screenwriter (C3)
		Internet Literature IP (C4)
		Film genre (C5)
	External factors (B2)	Film attention (C6)
		Release date (C7)
		Issuing Company (C8)

Calculate the eigenvalues and eigenvectors of the judgment matrix using an Excel spreadsheet with the following results: 1)

The largest eigenvalue of judgment matrix $A$ is 2 and the normalized eigenvectors are $W 1 = {(0.6667, 0.3333)}^{T}$ , $C I = 0$ , indicating that matrix $A$ has perfect consistency and weight vector $W 1$ is desirable.

1)

The largest eigenvalue of judgment matrix $B 1$ is 5.52, and the normalized eigenvectors are $W 2 = {(0.1857, 0.4531, 0.0621, 0.2991, 0.0482)}^{T}$ , $C I = 0.0972$ , $C R = 0.0834 < 0.1$ , indicating that matrix $B 1$ has satisfactory agreement and that weight vector $W 2$ is desirable.

3)

The maximum eigenvalue of judgment matrix $B 2$ is 4.05, and the normalized eigenvectors are $W 3 = {(0.6024, 0.2386, 0.1555)}^{T}$ , $C I = 0.0091$ , and $C R = 0.0091 / 0.58 = 0.0158 < 0.1$ , indicating that matrix $B 2$ has satisfactory consistency and weight vector $W 3$ is desirable.

The results of the hierarchical single sorting are organized results are shown in Table 3, experts believe that in the internal and external factors affecting the value of movie copyright, internal factors including actors, directors, screenwriters, online literature IP, movie genres are dominant, and their influence weight reaches about 0.6667. The external factors movie attention, release schedule and distribution company play a secondary role, and their influence weight is only about 0.3333. Among the internal factors, the weight of online literature IP reaches 0.2991.

Table 3.

Statistics of single-layer sorting

Project	The standard layer	Index level
Internet literature IP adaptation of film copyright value	Internal factors (0.6667)	Actor (0.1857)
		Director (0.4531)
		Screenwriter (0.0621)
		Internet Literature IP (0.2991)
		Film genre (0.0482)
	External factors (0.3333)	Film attention (0.6024)
		Release date (0.2385)
		Issuing Company (0.1555)

2.2.3

Calculation of the value of film and television adaptation rights

After using the hierarchical analysis method to calculate the share rate of network literature IP in the value of film copyright, it is possible to strip out the value of film and television adaptation rights of network literature IP, and the formula for calculating the value of film and television adaptation rights of network literature IP is as follows: (6) $V_{0} = V \times K = V \times 15.27 %$

Where $V_{0}$ is the value of the network literature IP film and television adaptation rights, and $V$ is the value of the copyright of the adapted movie. $K$ is the value of network literature IP film and television adaptation rights of the share rate, take 15.27%.

2.3

Revenue Assessment of Film and TV Adaptations Based on Machine Learning Algorithms

2.3.1

XGBoost Algorithm

XGBoost is an extreme gradient enhancement algorithm that, while still GBDT in nature, strives to maximize its speed and efficiency.XGBoost effectively implements many of the improvements made to the GBDT algorithm and engineering.The most significant modification to GBDT is to the definition of the objective function.

The objective function of XGBoost is detailed below: (7) $o b j = \sum_{i = 1}^{N} L (y_{i}, {\hat{y}}_{t}) + \sum_{m = 1}^{M} Ω (f_{m})$ (8) $Ω (f_{m}) = γ J + \frac{1}{2} λ {‖ω‖}^{2}$ where $Ω$ is the regular penalty term function and ${\hat{y}}_{t}$ is the predicted value of the model. $M$ iterations are performed, that is, $M$ trees are produced. $f_{m}$ is the $m$ th tree model, which is the $m$ th base learner. In the $Ω (f_{m})$ expression, $J$ denotes the number of leaf nodes in each tree, $ω$ denotes the sum of weights of the leaf nodes of that decision tree, and $γ$ and $λ$ denote the penalty coefficients.

The specific flow of XGBoost algorithm is: 1)

Initialize the base learner to $f_{0} (x)$ .

2)

Compute the first-order derivative and second-order derivative of the loss function for $m = 1, \dots .., M$ : (9) $g_{m t} = \frac{\partial L (y_{t}, f_{m - 1} (x_{t}))}{\partial f_{m - 1} (x_{t})}$ (10) $h_{m t} = \frac{\partial^{2} L (y_{t}, f_{m - 1} (x_{t}))}{\partial f_{m - 1} (x_{t})}$

The objective function of this decision tree is then obtained as: (11) $o b j^{(m)} = \sum_{i = 1}^{N} [g_{i} f_{m} (x_{t}) + \frac{1}{2} h_{t} f_{m}^{2} (x_{i})] + Ω (f_{m})$

3)

Sample traversal is converted to leaf node traversal with loss function: (12) $o b j^{(m)} = \sum_{j = 1}^{J} [(\sum_{i = l_{j}} g_{t}) w_{j} + \frac{1}{2} (\sum_{i = t_{j}} h_{t} + λ) w_{j}^{2}] + γ J$ where $I_{j}$ denotes the set of samples from the $j$ nd leaf node.

4)

Fit the data to generate the base learner $h_{m}$ , the best fit value is to minimize the objective loss function. The process of fitting is the process of decision tree node splitting, that is, selecting attributes for division, and the regression tree can be used to select the attributes of the division with the squared error. So the optimal weight value of each leaf node as well as the objective function value is obtained as: (13) $w_{j}^{*} = - \frac{G_{t}}{H_{t} + λ}$ (14) $o b j = - \frac{1}{2} \sum_{j = 1}^{J} \frac{G_{t}}{H_{t} + λ} + γ J$ where $G_{i} = \sum_{t \in I_{j}} g_{i}$ , $H_{t} = \sum_{i \in I_{j}} h_{t}$ .

5)

Update the prediction model for: (15) $f_{m} (x) = f_{m - 1} (x) + h_{m} (x)$

After $M$ iteration, the algorithm stops when the convergence condition we set is reached and the final prediction model is obtained.

2.3.2

Revenue Forecast for Film and TV Adaptations of Web Novels

In this section, machine learning algorithms are utilized to reasonably predict the revenue of web novel film and television adaptations. Through the training process of XGBoost algorithm, it lays the foundation for getting the copyright of web literature film adaptations. In this section, jupyter notebook is set as a python environment, the sample dataset is imported and divided into a training set and a test set, the already encapsulated regression function in the Sklearn library is called, and the residual plots are used to represent the gap between the real values of the training set and the test set and the values obtained by training the model, respectively. And borrow the indicators to compare the model’s goodness of fit and prediction effect, and finally adopt the model with the smallest error for the revenue prediction of online literature adaptations.

The above processed data are imported into the code and run to get the regression results. The fitting results of the four algorithms of linear regression, random forest regression, XGBoost regression and support vector machine regression are compared to find the optimal box office revenue prediction model. After obtaining the predicted value of adaptation revenue, it is necessary to consider the producer’s share rate to calculate the cash flow of copyright revenue, estimate the revenue period, i.e., the number of days of screening, according to the current situation of the movie industry, and in addition, the discount rate is derived according to the capital asset pricing model, so that the three parameters of the revenue method have been obtained, which allows us to arrive at the revenue assessment of the web literature adaptation.

3

Case studies on the assessment of the value of film and television adaptations of web novels

3.1

Training and Testing of Adaptation Benefit Assessment Models

3.1.1

Training of the adaptation benefit assessment model

This paper selects 95 classic cases of online fiction in recent years, and collected 95 samples. And 75 data from 95 samples are used for training, and the inputs are 7 variables: novel rating, author’s fan number, twitter heat, director’s rating, romance, plot, and Spring Festival slot, and the output is the adaptation revenue, i.e., the box office of the movie. Import the software, input the program and construct the model.

1)

The number of nodes in the input layer is 7, the number of nodes in the hidden layer is 15 number networks, and the number of nodes in the output layer is 1.

2)

The transfer function from the hidden layer to the output layer is tansig function with purelin function.

3)

The number of training times is set to 500, the rate is set to 0.05, and the minimum error of the training objective is 0.00005.

4)

The training function is trainlm function.

5)

The weights learning function is the clearngdm function, and the XGBoost algorithm is applied to adjust the weights and thresholds of each connection layer.

6)

The performance function is the mse function, which calculates the mean squared error between the output value and the expected value, imports the software, inputs the program, and constructs the model.

Finally, the gradient descent results of the fitting process are obtained as shown in Figure 3, which can be obtained by adapting the revenue assessment model to the actual iteration to 21 times to meet the requirements of the training accuracy, and training to the 15th generation, the training results are the most ideal, with a mean squared error of 0.032074.

The regression curves of the machine learning algorithm are shown in Fig. 4, where Figs. 4(a) to (d) show the training set, validation set, test set and the model as a whole, respectively. It can be seen that the validation set and test set have the same decreasing trend in error, and the validation and test set sample points are uniformly distributed near the fitting line indicating a good fit. The training goodness of fit of the adaptation revenue assessment model is close to 0.91385, and the overall goodness of fit of the model reaches 0.90051, indicating that the training results of the web novel film and television adaptation value assessment model based on the XGBoost algorithm are good.

The results of the comparison between the predicted and actual values of the model test set are shown in Fig. 5, where it can be seen that the expected and predicted values almost overlap, with a mean absolute error MAE of 0.45037, a mean square error MSE of 0.48715, and a root-mean-square error of 0.71214. It shows that the value assessment model of web novel film and television adaptations based on XGBoost algorithm is constructed successfully, but there is the problem of overfitting.

3.1.2

Tests of the model for assessing the benefits of adaptation

In this paper, 10 movies A~J predicted box office adapted by network literature IP are selected, and the predicted values of the test samples are compared and analyzed with the actual values to further test whether the model in this paper is constructed successfully. The error between the model assessment revenue and the actual revenue is shown in Table 4, and it can be seen that the absolute value of the error fluctuates between 0.25% and 8.00%, and the error is basically controlled within 10%, which indicates that the error is small, and the revenue assessment model of the IP adaptation of network novels constructed in this paper is established.

Table 4.

The error between the model evaluation income and the actual income

Movie name	Actual box office	Prediction box office	Error
A	36.05	33.89	6.01%
B	7.79	7.21	7.55%
C	13.55	13.52	0.28%
D	34.11	34.02	0.25%
E	12.84	12.57	2.09%
F	47.41	43.98	7.24%
G	49.59	48.86	1.46%
H	22.68	22.11	2.50%
I	19.91	19.78	0.65%
J	31.23	28.73	8.00%

3.2

Empirical Case Analysis of Film and Television Adaptation of Web Novels

3.2.1

Assessment of basic matters

1)

Object of appraisal

In the 95 classic cases selected in this article, the film and television adaptation of Qing Yu Nian is the most powerful and representative. Therefore, the object of this appraisal is the copyright value of the filmization of the online literary work Qing Yu Nian.

2)

Purpose of appraisal

The purpose of this appraisal is to provide a reference for the value of the copyright filmization of “Qing Yu Nian”, i.e. the price at which the work is assumed to be traded for adaptation rights.

3)

Valuation principles

The appraisal follows the principles of independence, fairness and impartiality, and scientificity.

4)

Type of valuation

The type of value for this appraisal is the market value of the film and television rights of the internet literature Qing Yu Nian as at the valuation reference date. Market value refers to the price at which a voluntary buyer and a voluntary seller would have normally traded on the valuation reference date under their respective rational and unoppressed circumstances.

5)

Valuation benchmark date

As of 2023, the film and television adaptation of Qing Yu Nian is available in the form of a TV series, therefore this article assumes that a film and television production company is planning a film adaptation of Qing Yu Nian, and therefore sets the valuation benchmark date as December 31, 2023.

3.2.2

Determination of the split rate

This paper identifies three major driving factors for movie adaptation of network literature, network literature IP works, copyright operation, and external environmental factors. And a questionnaire is constructed to calculate the share rate of network literature works. This paper invites groups such as practitioners in the online literature industry and experts in the evaluation industry to conduct a questionnaire survey, and 100 questionnaires were collected. After processing, the scale method was used to assign values to the importance of each layer of indicators. After the 100 questionnaires collected were sorted and analyzed, the results of the weight vector and the maximum characteristic root obtained by the hierarchical analysis method are shown in Table 5. Among them, the maximum characteristic root of the value of the movie and television production of online literature is 3.0985, and the maximum weight vector of this dimension is 0.6935 of the copyright operation, and the maximum characteristic root of the works of online literature is 3.1482, and the maximum weight vector of this dimension is 0.6284 of the contents of the works.

Table 5.

Weight vector and maximum feature root results

Solution layer	Criterion layer	Weighting (Wi)	Maximum characteristic root
The value of the net film	The communication of the online work	0.1315	3.0985
	Copyright operation	0.6935
	Macroscopic environment	0.1750
Online literature	Work content	0.6284	3.1482
	Work influence	0.2321
	Social value	0.1395
Internal environment	Outside the movie	0.3483	2.9843
	Network platform	0.2519
	Inside the movie	0.3998
External environment	Policy environment	0.3500	2.9937
External environment	Industrial development level	0.6500	2.9937

3.2.3

Adaptation value analysis

Bringing the split rate into the hierarchical structure model, the evaluation result of the value of the film and television adaptation of Qing Yu Nian is calculated to be $V$ =352.73 million yuan. The rate of IP belonging to network literature $K$ = 12.84%, the discount rate is $r$ = 15.27%, and the discount period is $t$ for 2 years. Therefore, the copyright value of film and television adaptation of Qing Yu Nian is $V' = V * K / (1 + r)$ = 52.2064 million yuan. It is understood that “Qing Yu Nian” has not yet been developed for the film, but its film and television copyright has been sold in 2017 for 20 million, taking into account industry factors and consulting practitioners in the network literature industry, it can be found that the transaction price in 2017 is obviously low, and its true value is underestimated. From this, it can be found that the idea of using online literature fans as an entry point to predict the revenue of the filmization and development scenario of online literature copyrights, combined with the income approach for evaluation is feasible.

This paper carries out a sensitivity analysis for some of the major parameters in this assessment. Due to the characteristics of the formula, the adaptation impact coefficient $β$ , the potential user D and the user purchase P are linearly related to the copyright value, and therefore the changes are of the same magnitude, so we will not repeat them here. The sensitivity analysis of the important coefficients for the copyright value assessment of the film and television adaptation of Qing Yu Nian is shown in Table 6. Starting from the paid users U, the genuine rate K, the discount rate r, and the conversion rate T, it can be seen that the discount rate r and the conversion rate T have a greater degree of influence on the copyright value. When r rises by 10%, copyright value V falls by 2.37%. When T rises by 10%, copyright value V rises by 0.98%. In contrast, paying users U, the licensing rate K, have a lesser degree of influence on the results; when U rises by 10%, the copyright value V falls by 0.45%. The resultant copyright value rises by 0.40% when the genuine rate K rises by 10%.

Table 6.

Sensitivity analysis of important coefficients

Paid user (U)				Validity (k)
Variation	U	V	Sensibility	Variation	K	V	Sensibility
10%	60.918	5197.147	-0.45%	10%	0.1412	5241.523	0.40%
5%	58.149	5208.633	-0.23%	5%	0.1348	5231.603	0.21%
0	55.38	5220.64	0%	0	0.1284	5220.64	0%
-5%	52.611	5232.647	0.23%	-5%	0.1220	5208.633	-0.23%
10%	49.842	5244.133	0.45%	10%	0.1156	5195.581	-0.48%
Discount rate (r)				Conversion (T)
10%	0.1680	5096.911	-2.37%	10%	0.88	5271.802	0.98%
5%	0.1603	5160.603	-1.15%	5%	0.84	5247.265	0.51%
0	0.1527	5220.64	0%	0	0.80	5220.64	0%
-5%	0.1451	5287.464	1.28%	-5%	0.76	2558.114	-0.51
10%	0.1374	5358.987	2.65%	10%	0.72	5169.478	-0.98%

3.3

Development Path of Film and Television Adaptation of Network Novels

Through the above analysis, it can be seen that potential users, user purchase and conversion rate have a greater impact on the copyright value of network literature film and television. In the case that the fans of the original work have been fixed, the publicity work of the work should be strengthened through social media to expand the potential users and improve the conversion rate of the fans. It is also possible to use famous actors and directors in the production stage of the movie to convert their fans into the fans of the derivative work, so that the work can maximize value-added in the case of a certain quality. Aiming at the problems faced by the industrialized development of network novels, this paper puts forward the following paths for the development of film and television adaptations of network novels: 1)

Strengthen the interactive relationship and insist that content is king

Many authors excessively pursue immediate fame and fortune, creating a large number of low-quality novels with vulgar plots and coarse language, bringing criticism to the whole network novel industry. Although these works have a certain reader market, from the perspective of the long-term results of the development of network novel film and television adaptation, high-quality novels of high quality have more space for development. Therefore, whether it is the market prospect of network novel film and television adaptation, or the demand of the whole society for high-quality network novel works, all put forward strict requirements for the content of network novels, and high-quality content is the key to the development of its film and television adaptation.

2)

Strengthening the Union of Industrial Media

Although there is nothing wrong with granting the adaptation rights of each segment of the industrialization of network novels to different subjects, the same subject with multiple adaptation rights can better unite the different forms of industry and promote the development of each other. Adaptation of animation and TV series on line, to ensure the linkage of different forms of IP, so that fans continue to flow between different areas of IP, forming a multi-directional cycle of readers - players - viewers.

3)

Playing the role of “gatekeeper” to create quality movie and TV dramas

Although in recent years, network novels adapted to film and television drama, was a spurt of development. However, there are very few quality works with good quality and good reputation. Many film and television drama developers blindly pursue the current interests, only pay attention to the film and television drama of the click rate and the amount of airplay, while ignoring the quality of the work of this key issue. But the quality is a film and television drama, to win the favor of the audience, the key to establish a reputation. And from the overall perspective of the industrialization of network novels, a well-produced quality drama can represent the success of the industrialization, and then effectively promote the subsequent production of the novel, and ultimately build it into an IP brand. Therefore, content is the king, and the production of dramas is emphasized. The adaptation of the drama into a high-quality drama should become the focus of the future development of the network novel adaptation industry.

4

Conclusion

In today’s network novel comprehensive film and television, its development process inevitably brings some problems, recognize these problems, conducive to the healthy development of network novel film and television. This paper adopts the hierarchical analysis method to evaluate the value of network novel film and television adaptation rights, and constructs the revenue assessment model of film and television adaptation based on XGBoost algorithm. The study shows that the training goodness of fit of the revenue assessment model is 0.91385, and the overall goodness of fit is 0.90051, which indicates that the training results of the model are good, and it can be applied to the assessment of the revenue of network novel film and television adaptation. Combined with the relevant value influencing factors to construct the copyright value index system of network literature film and television, through the case of “Qing Yu Nian” for verification. This paper introduces the relevant theories of communication science for revenue prediction, and the idea of measuring the share rate with the copyright value index system of network literature film and televisionization can provide reference for the benign development of network novel film and televisionization.

Language:: English

Publication timeframe:: 1 times per year
Journal Subjects:: Life Sciences, Life Sciences, other, Mathematics, Applied Mathematics, General Mathematics, Physics, Physics, other

Journal RSS Feed

Z	A1	A2	A3	A4	A5
A1	a11	a12	a13	a14	a15
A2	a21	a22	a23	a24	a25
A3	a31	a32	a33	a34	a35
A4	a41	a42	a43	a44	a45
A5	a51	a52	a53	a54	a55

Z	A1	A2	A3	A4	A5
A1	a11	a12	a13	a14	a15
A2	a21	a22	a23	a24	a25
A3	a31	a32	a33	a34	a35
A4	a41	a42	a43	a44	a45
A5	a51	a52	a53	a54	a55

Exploration of the Development Path of Network Novel Film and TV Adaptation Drama in the Background of Internet

Ye Wang

Xingang Zhang

Published Online: Sep 24, 2025

Received: Jan 23, 2025

Accepted: May 09, 2025

DOI: https://doi.org/10.2478/amns-2025-0998

KeywordsHierarchical analysis, XGBoost algorithm, Film and television adaptation rights, Revenue evaluation model

© 2025 Songyu Ye Wang and Xingang Zhang published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Keywords
Hierarchical analysis, XGBoost algorithm, Film and television adaptation rights, Revenue evaluation model

Z	A1	A2	A3	A4	A5
A1	a11	a12	a13	a14	a15
A2	a21	a22	a23	a24	a25
A3	a31	a32	a33	a34	a35
A4	a41	a42	a43	a44	a45
A5	a51	a52	a53	a54	a55