Font Size: a A A

Analysis Of Animation Movie Network Comments Based On Text Mining

Posted on:2021-04-29Degree:MasterType:Thesis
Country:ChinaCandidate:Y HeFull Text:PDF
GTID:2415330605458450Subject:Master of Statistics in Applied Statistics
Abstract/Summary:PDF Full Text Request
In recent years,China’s excellent animation films have been pouring in front of people,especially in the summer vacation of 2019,the explosion of "the devil child of Nezha" has shifted the public’s attention to the domestic animation films.Although domestic animation films have been making progress in recent years,they still lag behind the United States and Japan,which are powerful animation countries in the world.In this paper,from the perspective of network viewers’ comments,the method of combining text data mining and statistical analysis is adopted to study the network comments of animation films in different countries,and useful information and the gap between the animation of China,the United States and Japan are mined out from the comments,so as to provide some help for the rise of domestic animation films.First of all,this paper uses Python software to grab the review data of the top 10 animated movies from Douban website from 2015 to 2019 as the analysis object;second,data cleaning and preprocessing of the reviews;third,using naive Bayesian method to construct a classifier to classify the online review data of animated movies into positive ones face to face comments and negative comments;fourth,use LDA theme model to extract the positive comments and negative comments in different countries’ animation film network comments,and further analyze the deeper reasons behind these theme words.In this paper,naive Bayesian text classification method is used to classify positive and negative samples of relevant film reviews.It is found that this classification method can effectively,accurately,comprehensively and objectively classify positive and negative texts.Then we use the LDA theme model to classify the positive and negative texts from naive Bayes,and extract the theme words.The positive theme words in the United States and Japan are mostly "special effects","details","Pixar","Disney","Hayao Miyazaki" and other themes,which show that there are large animation companies or studios in the United States and Japan,while the negative theme words in China are mostly "model" Imitation,boredom and low age show that there is still a lack of large animation production companies in China,which leads to a lack of excellent original works and excellent screenwriters.In this regard,the author suggests that the creation of Chinese animation films should be strengthened,large animation production companies with Chinese characteristics and styles,creative teams with original spirit and excellent original works should be cultivated,audiences should be expanded,derivatives of animation films should be developed,and relevant technical personnel should be trained to strengthen the protection of intellectual property rights,To provide a good creative environment for the creators;to formulate the domestic film industry development plan and support policies,cultivate the main body of the animation film and television industry market,broaden the animation film industry chain,actively develop and expand the animation film market,constantly enrich the cultural and spiritual life of the masses,expand cultural exchanges with foreign countries,and provide the influence of Chinese culture and the soft power of the country.
Keywords/Search Tags:Text Mining, movie reviews, text classification, LDA topic model
PDF Full Text Request
Related items