基于文本挖掘的豆瓣电影评论的LDA主题模型分析
Analysis of LDA Theme Model for Douban Movie Reviews Based on Text Mining

网络评论的兴起，让普罗大众都能参与到对电影的解读和评价中来，撰写影评不再是专业影评人所独有的一项活动。电影网络评论作为对电影作品的评价，既是观众对电影好坏的直观反映，也是其观影体验的直接体现。对网络影评的分析有助于直接了解观众对电影的观感和评价，进而加深对经典电影大火基因的理解，并为电影制作提供相应借鉴。电影《让子弹飞》上映十三年后仍被奉为经典与神作，片中的经典桥段和台词一直被人们所津津乐道。本文通过对该部影片豆瓣影评的文本挖掘和LDA主题模型分析，探究观众评价中对影片的关注点和评论视角，基于高频主题词，挖掘深层主题，有助于客观整体评价电影和反映观众的真实感受，寻找该片持续大火的传播基因与密码。
The rise of online reviews has enabled the general public to participate in the interpretation and evaluation of movies, and writing reviews is no longer a unique activity for professional film critics. Film online reviews, as an evaluation of film work, are not only an intuitive reflection of the quality of the movie by the audience, but also a direct reflection of their viewing experience. The analysis of online film reviews helps to directly understand the audience’s perception and evaluation of movies, deepen their understanding of the genes of classic movies, and provide corresponding references for film production. The movie “Let the Bullets Fly” is still regarded as a classic and masterpiece thirteen years after its release, and the classic scenes and lines in the film have always been talked about by people. This article explores the focus and perspective of audience evaluation on the film through text mining and LDA theme model analysis of Douban film reviews. Based on high-frequency theme words, it explores deep themes, which helps to objectively evaluate the film as a whole and reflect the audience's true feelings, and to find the spreading genes and codes of the film’s sustained popularity.