基于文本挖掘的纪录片传播影响因素分析

Analysis of Influencing Factors of Documentary Communication Based on Text Mining

摘要: 为了解影响纪录片传播的因素，将爬虫得到数据资源清洗后，经描述统计获得基础影响因素作为分类变量，使用Python得到较精准的中文分词结果，编写Gibbs算法建立LDA模型来进行分析得到不同纪录片适合的困惑度和主题数以及语义网络、词云图等结果。通过分析文本的结果得知人们的价值观与纪录片的传播可展现双向的影响作用，纪录片的拍摄会根据时代的主流与需求进行拍摄，而时代的主流与需求也是人们的价值观最直接的展现；再者，人们通过观看纪录片来了解当下的世界展现给人们的面目，同样这也是一种对人们思维的引领方向，影响较大的为人们的现实需求、纪录片的承载形式以及纪录片的真实性。

Abstract: After cleaning the basic influence factors as classification variables obtained from the descriptive statistics, in order to understand the factors affecting the spread of the documentary, the crawled data resources get more accuracy of Chinese word segmentation results using Python. Writing Gibbs algorithm analysis LDA model is set up to get a different degree of confusion and theme of the documentary for number and semantic network, such as word cloud image results. Through the analysis of the text, it is found that people’s values and the dissemination of documentaries can show a two-way influence. Documentary filming will be based on the mainstream and demand of The Times, which is also the most direct display of people’s values. In addition, people can under-stand what the current world shows to people by watching documentaries, which is also a guiding direction for people’s thinking and has a great influence on people’s practical needs, the carrying form of documentaries and the authenticity of documentaries.