基于LDA的电影主题自动分类方法的研究  

Research on Automatic Classification Method of Movie Themes Based on LDA

作　　者：李璐 王妍 王艳娥 杨倩[1] LI Lu;WANG Yan;WANG Yane;YANG Qian(Institute of Science and Technology,Xi'an Siyuan University,Xi'an 710038,China;Fujian Wangneng Technology Development Co.,Ltd.,Fuzhou 350000,China)

机构地区：[1]西安思源学院理工学院,陕西西安710038 [2]福建网能科技开发有限责任公司,福建福州350003

出　　处：《计算机与网络》2023年第3期58-61,共4页Computer & Network

基　　金：一般专项科研计划项目(2022JK0515)。

摘　　要：针对传统采用人工方式对电影主题进行分类存在主观性强、分类标准不统一的问题,提出了一种基于LDA的电影主题自动分类方法,通过对电影简介数据进行建模,计算出电影主题的概率主题模型的联合分布公式,使用Gibbs采样算法求解联合分布公式,得出电影的主题分布及电影主题关键词的分布,并根据这2个分布完成电影主题的自动分类及类别的自动标识,使用电影简介数据对电影主题进行分类实验。实验结果表明,该方法能够对电影主题进行准确分类,精确度达到95%,从根本上消除了人工分类方法中存在的主观性强、分类标准不统一的问题。In order to solve the problem of strong subjectivity and inconsistent classification standards in traditional artificial methods of classifying movie themes,an automatic classification method of movie themes based on LDA is proposed.Through the modeling of movie introduction data,the joint distribution formula of the probability theme model of movie themes is calculated,and the joint distribution formula is solved by Gibbs sampling algorithm.Thus,the distribution of movie themes and their key words are obtained.According to this,the automatic classification of movie themes and the automatic identification of categories are completed.Finally,the movie theme classification experiments are carried out by using movie introduction data.The experimental results show that this method can classify movie themes with the accuracy to be 95%,eliminating the problems of strong subjectivity and inconsistent classification standards in artificial classification methods.

关 键 词：LDA GIBBS采样 电影主题 自动分类 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]