CN107480136B - A Method Applied to Sentiment Curve Analysis in Movie Scripts - Google Patents

Abstract

The application discloses a method for analyzing an emotional curve in a movie scenario, which comprises the following steps: 1) constructing an emotion dictionary, including a positive dictionary, a negative dictionary and a neutral dictionary; 2) carrying out scene division and sentence extraction on the movie scenario according to scenes to obtain scene sentences and character sentences; 3) segmenting words of the scene sentences and the character sentences to obtain scene vocabularies and character vocabularies; 4) performing part-of-speech division, word frequency statistics and weight determination on scene vocabularies and character vocabularies; 5) importing the word frequency and the weighted value into a calculation formula to obtain a scene emotion index and a character emotion index; 6) and drawing a scene emotion curve and a character emotion curve of the movie scenario. The invention provides a method for analyzing the emotional curve in the movie script, which has no requirement on working experience and is convenient and quick to calculate.

Description

A Method Applied to Sentiment Curve Analysis in Movie Scripts

technical field

The invention relates to the field of movie script analysis, and more particularly, to a method applied to emotional curve analysis in movie scripts.

Background technique

Watching movies is one of people's main leisure and entertainment methods. With the gradual improvement of material life, people's requirements for movies also increase. The excellence of a movie is closely related to the level of the script. Therefore, movie investors When examining whether a movie is worth investing in, it is inevitable to evaluate the movie script accordingly, and the overall emotional changes of the movie script and the emotional changes of each character in the script are important indicators for evaluating the quality of a movie script. Emotional change is the invisible emotional ups and downs hidden between the lines of a movie script. People cannot directly see such emotional changes. At present, it is still difficult to obtain the emotional changes in a movie script. It is mainly carried out by experienced film practitioners reading the movie script. Sentiment analysis, and summarize and evaluate the emotional changes in the script. This analysis method requires high film experience for the staff and takes a long time. How to quickly and effectively analyze the emotions in a movie script Curves, reducing the film industry experience requirements for staff, and reducing the time spent in the evaluation process are urgent problems to be solved in this field.

Therefore, it is necessary to provide a method for emotional curve analysis in movie scripts to solve the above problems.

SUMMARY OF THE INVENTION

In view of this, the present invention provides a method for emotional curve analysis in a movie script, so as to solve the problem that the existing movie script emotional analysis process requires relatively high movie experience for staff and takes a long time.

In order to solve the above-mentioned technical problems, the application has the following technical solutions:

A method applied to sentiment curve analysis in movie scripts, including:

constructing an emotion dictionary of the movie script, including: a positive dictionary, a negative dictionary and a neutral dictionary, and assigning a weight value to each word in the emotion dictionary;

The movie script is divided according to the scene, and sentence extraction is performed on each of the divided scenes to obtain the scene sentence of each of the scenes;

From each of the scene sentences, extract the sentences whose subjects are characters, and classify them according to the characters, so as to obtain the character sentences of each of the characters in each of the scenes;

Perform word segmentation on the scene sentences to obtain the scene vocabulary of each of the scenes,

Perform word segmentation on the character statement to obtain the character vocabulary of each of the characters in each of the scenes;

Divide the scene vocabulary and the character vocabulary into positive words, negative words and neutral words according to the sentiment dictionary, and count the word frequency and the weight value of each of the words;

The word frequency and the weight value of the scene vocabulary are imported into the calculation formula to obtain the scene emotion index of each of the scenes,

The word frequency and the weight value of the character vocabulary are imported into the calculation formula to obtain the character emotion index of each character, wherein,

The calculation formula is:

Wherein: n is the total number of the positive words, m is the total number of the negative words, P _i is the word frequency of the i-th positive word, Wi is the _i -th positive word The weight, N _j is the word frequency of the jth negative word, W _j is the weight of the jth negative word, L is the positive word, the negative word and the neutral The sum of the number of words, x is a positive number not less than 1;

Taking the scene number as an independent variable, and the scene emotion index as a dependent variable, draw the scene emotion curve of the movie script,

Taking the scene number as the independent variable and the character emotion index as the dependent variable, draw a character emotion curve of each of the characters.

Optionally, the sentiment dictionary includes: a Chinese sentiment polarity dictionary, a set of words for sentiment analysis of HowNet, and a self-developed sentiment dictionary.

Optionally, the scene sentence further includes: an atmosphere sentence in which the subject is not a character.

Optionally, the word segmentation adopts a Chinese language processing package or a Chinese lexical analysis system.

Optionally, the weight value of the positive word and the negative word is greater than 0 and not greater than 1, and the weight value of the neutral word is equal to 0.

Optionally, the scene emotion curve and the character emotion curve are drawn as a line graph, a bar graph, a scatter graph and a fan graph.

Compared with the prior art, the present invention provides a method for analyzing emotion curves in movie scripts, which achieves the following beneficial effects:

1) A method provided by the present invention is applied to emotional curve analysis in movie scripts. There is no step in the whole method that requires personnel to have experience in the film industry, and the calculation process is convenient and fast, which solves the problem of high requirements for personnel experience and high consumption in the past. the question of time;

2) A kind of method that is applied to emotional curve analysis in movie scripts provided by the present invention can analyze the character emotional curve of each character, which is convenient for movie script writers to judge whether the plot arrangement of each character is suitable, and select according to the emotional changes. the best matching actor;

3) A method for analyzing emotion curves in movie scripts provided by the present invention draws invisible emotional changes into visible emotion curves, which provides a basis for investors to judge the overall level of movie scripts and reduces the risk of loss. investment risk.

Description of drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention.

Fig. 1 is the flow chart of "a method applied to emotional curve analysis in movie script" of the present invention

Detailed ways

Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the invention unless specifically stated otherwise.

The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, such techniques, methods, and apparatus should be considered part of the specification.

In all examples shown and discussed herein, any specific values should be construed as illustrative only and not limiting. Accordingly, other instances of the exemplary embodiment may have different values. It should be noted that once an item is defined in one figure, it does not require further discussion in subsequent figures.

Example 1:

Fig. 1 is a flow chart of "a method applied to emotional curve analysis in a movie script" of the present invention.

A method applied to sentiment curve analysis in movie scripts, as shown in Figure 1, includes the following steps:

Step S101: Build an emotion dictionary of the movie script, including: a positive dictionary, a negative dictionary, and a neutral dictionary, and assign a weight value to each word in the emotion dictionary.

Specifically, there is no intersection between the positive dictionary, the negative dictionary and the neutral dictionary, and a word can only belong to one of the dictionaries. The weight of a word is related to the emotional intensity expressed by the word.

Step S102: Divide the movie script according to the scene, extract sentences for each scene after the division, and obtain the scene sentences of each scene; Classification, get the character statement of each character in each scene.

Specifically, when the movie script is divided into scenes, based on the scene number marked in the movie script, all the content in the movie script should be included in the scene statement.

Step S103: Perform word segmentation on the scene sentences to obtain the scene vocabulary of each scene, and perform word segmentation on the character sentences to obtain the character vocabulary of each character in each scene.

Specifically, a threshold is set for the number of character sentences. If the number of character sentences of a certain character is less than the threshold, the character sentences of this character are not counted, and no threshold is set for the number of scene sentences.

Step 104: Perform word segmentation on the scene sentence to obtain the scene vocabulary of each scene, and perform word segmentation on the character sentence to obtain the character vocabulary of each character in each scene.

Specifically, a threshold is set for the number of character vocabulary. If the number of character vocabulary of a certain character is less than the threshold, the character vocabulary of the character is not counted, and no threshold is set for the number of scene vocabulary.

Step 105: Divide the scene vocabulary and character vocabulary into positive words, negative words and neutral words according to the sentiment dictionary, and count the word frequency and weight value of each word;

Specifically, the word frequency refers to the number of times a word appears. For any movie script, the weight value of any word is a fixed value, and for two different movie scripts, the weight value of the same word can be the same or different.

Step 106: Import the word frequency and weight value of the scene vocabulary into the calculation formula to obtain the scene emotion index of each scene, and import the word frequency and weight value of the character vocabulary into the calculation formula to obtain the character emotion index of each character.

The calculation formula is:

Among them: n is the total number of positive words, m is the total number of negative words, P _i is the word frequency of the _i -th positive word, Wi is the weight of the i-th positive word, N _j is the j-th negative word The word frequency of , W _j is the weight of the jth negative word, L is the sum of the number of positive words, negative words and neutral words, and x is a positive number greater than or equal to 1.

Specifically, if the calculated scene emotion index is greater than 0, it means that the emotion of the scene is positive, if the scene emotion index is less than 0, it means that the emotion of the scene is negative, and if the scene emotion index is equal to 0, it means that the emotion of the scene is is neutral.

If the calculated character emotion index is greater than 0, it means that the character's emotion in the scene is positive; if the character emotion index is less than 0, it means that the character's emotion in the scene is negative; if the scene emotion index is equal to 0, it means the The emotions of the characters in the scene are neutral.

Step S107: Using the scene number as the independent variable and the scene emotion index as the dependent variable, draw the scene emotion curve of the movie script, using the scene number as the independent variable and the character emotion index as the dependent variable, draw the character of each character Emotional curve.

Specifically, the drawn scene emotion curve indicates the emotional change trend of the entire movie script, the character emotion curve indicates the emotional change trend of a specific character, and the emotional change trend of any character has nothing to do with the emotional change trend of the movie script.

Example 2

Fig. 1 is a flow chart of "a method applied to emotional curve analysis in a movie script" of the present invention.

A method applied to sentiment curve analysis in movie scripts, as shown in Figure 1, includes the following steps:

Step S101: Build an emotion dictionary of the movie script, including: a positive dictionary, a negative dictionary, and a neutral dictionary, and assign a weight value to each word in the emotion dictionary.

Specifically, there is no intersection between the positive dictionary, the negative dictionary and the neutral dictionary, and a word can only belong to one of the dictionaries. The weight of a word is related to the emotional intensity expressed by the word.

Step S102: Divide the movie script according to the scene, extract sentences for each scene after the division, and obtain the scene sentences of each scene; Classification, get the character statement of each character in each scene.

Specifically, when the movie script is divided into scenes, based on the scene number marked in the movie script, all the content in the movie script should be included in the scene statement.

Step S103: Perform word segmentation on the scene sentences to obtain the scene vocabulary of each scene, and perform word segmentation on the character sentences to obtain the character vocabulary of each character in each scene.

Specifically, a threshold is set for the number of character sentences. If the number of character sentences of a certain character is less than the threshold, the character sentences of this character are not counted, and no threshold is set for the number of scene sentences.

Step 104: Perform word segmentation on the scene sentence to obtain the scene vocabulary of each scene, and perform word segmentation on the character sentence to obtain the character vocabulary of each character in each scene.

Specifically, a threshold is set for the number of character vocabulary. If the number of character vocabulary of a certain character is less than the threshold, the character vocabulary of the character is not counted, and no threshold is set for the number of scene vocabulary.

Step 105: Divide the scene vocabulary and character vocabulary into positive words, negative words and neutral words according to the sentiment dictionary, and count the word frequency and weight value of each word;

Specifically, the word frequency refers to the number of times a word appears. For any movie script, the weight value of any word is a fixed value, and for two different movie scripts, the weight value of the same word can be the same or different.

Step 106: Import the word frequency and weight value of the scene vocabulary into the calculation formula to obtain the scene emotion index of each scene, and import the word frequency and weight value of the character vocabulary into the calculation formula to obtain the character emotion index of each character.

The calculation formula is:

Among them: n is the total number of positive words, m is the total number of negative words, P _i is the word frequency of the _i -th positive word, Wi is the weight of the i-th positive word, N _j is the j-th negative word The word frequency of , W _j is the weight of the jth negative word, L is the sum of the number of positive words, negative words and neutral words, and x is a positive number greater than or equal to 1.

Specifically, if the calculated scene emotion index is greater than 0, it means that the emotion of the scene is positive, if the scene emotion index is less than 0, it means that the emotion of the scene is negative, and if the scene emotion index is equal to 0, it means that the emotion of the scene is is neutral.

If the calculated character emotion index is greater than 0, it means that the character's emotion in the scene is positive; if the character emotion index is less than 0, it means that the character's emotion in the scene is negative; if the scene emotion index is equal to 0, it means the The emotions of the characters in the scene are neutral.

Step S107: Using the scene number as the independent variable and the scene emotion index as the dependent variable, draw the scene emotion curve of the movie script, using the scene number as the independent variable and the character emotion index as the dependent variable, draw the character of each character Emotional curve.

Specifically, the drawn scene emotion curve indicates the emotional change trend of the entire movie script, the character emotion curve indicates the emotional change trend of a specific character, and the emotional change trend of any character has nothing to do with the emotional change trend of the movie script.

Further, the sentiment dictionary includes Chinese sentiment polarity dictionary, HowNet's sentiment analysis word set and self-developed sentiment dictionary.

Specifically, the self-developed sentiment dictionaries include positive dictionaries, negative dictionaries, and neutral dictionaries. For the positive dictionaries and negative dictionaries, further subdivisions will be made according to the intensity of the emotions expressed by the words.

Further, the scene sentence in the scene sentence further includes: the subject is not an atmosphere sentence of a character.

Specifically, the atmosphere sentence can be calculated to draw the atmosphere emotion curve. If the number of atmosphere sentences in the scene sentence is less than the threshold, the atmosphere emotion curve cannot be drawn at this time.

Further, the word segmentation adopts a Chinese language processing package or a Chinese lexical analysis system.

Specifically, the word segmentation process adopts computing software to ensure the accuracy and completeness of the word segmentation.

Further, the weight value of positive words and negative words is greater than 0 and not greater than 1, and the weight value of neutral words is equal to 0.

Specifically, the weight value is a value preset in the sentiment dictionary. When calculating any movie script, the weight value is a fixed value, but when calculating different movie scripts, the weight value can be the same or different. Positive words and The weight value of negative words will be adjusted accordingly according to different movie scripts. The basis of this adjustment includes: the meaning and emotional intensity expressed by the words change with the times.

Further, the scene emotion curve and the character emotion curve are drawn into a line graph, a bar graph, a scatter graph and a fan graph.

Specifically, all the scene emotion curves and character emotion curves may be drawn, or only a part of them may be drawn. The drawn scene emotion curves and character emotion curves include: two-dimensional curves and three-dimensional curves. The expressions of two-dimensional and three-dimensional curves include: static and dynamic.

It can be seen from the above embodiments that the present invention has achieved the following beneficial effects:

1) A method provided by the present invention is applied to emotional curve analysis in movie scripts. There is no step in the whole method that requires personnel to have experience in the film industry, and the calculation process is convenient and fast, which solves the problem of high requirements for personnel experience and high consumption in the past. the question of time;

2) A kind of method that is applied to emotional curve analysis in movie scripts provided by the present invention can analyze the character emotional curve of each character, which is convenient for movie script writers to judge whether the plot arrangement of each character is suitable, and select according to the emotional changes. the best matching actor;

3) A method for analyzing emotion curves in movie scripts provided by the present invention draws invisible emotional changes into visible emotion curves, which provides a basis for investors to judge the overall level of movie scripts and reduces the risk of loss. investment risk.

Although some specific embodiments of the present invention have been described in detail by way of examples, those skilled in the art should understand that the above examples are provided for illustration only and not for the purpose of limiting the scope of the present invention. Those skilled in the art will appreciate that modifications may be made to the above embodiments without departing from the scope and spirit of the present invention. The scope of the invention is defined by the appended claims.

Claims (6)

1. a method that is applied to emotional curve analysis in movie script, is characterized in that, comprises: constructing an emotion dictionary of the movie script, including: a positive dictionary, a negative dictionary and a neutral dictionary, and assigning a weight value to each word in the emotion dictionary; The movie script is divided according to the scene, and sentence extraction is performed on each of the divided scenes to obtain the scene sentence of each of the scenes; From each of the scene sentences, extract the sentences whose subjects are characters, and classify them according to the characters, so as to obtain the character sentences of each of the characters in each of the scenes; Perform word segmentation on the scene sentences to obtain the scene vocabulary of each of the scenes, Perform word segmentation on the character statement to obtain the character vocabulary of each of the characters in each of the scenes; Divide the scene vocabulary and the character vocabulary into positive words, negative words and neutral words according to the sentiment dictionary, and count the word frequency and the weight value of each of the words; The word frequency and the weight value of the scene vocabulary are imported into the calculation formula to obtain the scene emotion index of each of the scenes, The word frequency and the weight value of the character vocabulary are imported into the calculation formula to obtain the character emotion index of each character, wherein, The calculation formula is:

Wherein: n is the total number of the positive words, m is the total number of the negative words, P _i is the word frequency of the i-th positive word, Wi is the _i -th positive word The weight, N _j is the word frequency of the jth negative word, W _j is the weight of the jth negative word, L is the positive word, the negative word and the neutral The sum of the number of words, x is a positive number not less than 1; Taking the scene number as an independent variable, and the scene emotion index as a dependent variable, draw the scene emotion curve of the movie script, Taking the scene number as the independent variable and the character emotion index as the dependent variable, draw a character emotion curve of each of the characters. 2. a kind of method that is applied to emotional curve analysis in movie script as claimed in claim 1, it is characterized in that, described emotional dictionary, comprises: Chinese emotional polarity dictionary, the sentiment analysis of HowNet uses word set and self-developed Emotional Dictionary. 3 . The method of claim 1 , wherein the scene sentence further comprises: an atmosphere sentence whose subject is not a character. 4 . 4 . The method as claimed in claim 1 , wherein the word segmentation adopts a Chinese language processing package or a Chinese lexical analysis system. 5 . 5. The method applied to sentiment curve analysis in movie scripts according to claim 1, wherein the weights of the positive words and the negative words are greater than 0 and not greater than 1, and the weights of the positive words and the negative words are The weight value for neutral words is equal to zero. 6. a kind of method that is applied to emotional curve analysis in movie script as claimed in claim 1, it is characterized in that, described scene emotional curve and described character emotional curve are drawn into line graph, bar graph, scatter graph and Fan chart.

Images