首页|基于SVM和LSTM的在线剧评分析模型

基于SVM和LSTM的在线剧评分析模型

扫码查看
海量的观影评论包含着广大观众对影视作品各方面的偏好,可以为影视剧的拍摄和宣传提供决策支持.提出影视剧在线评论分析模型:利用Python爬取传媒平台的评论信息,经过数据预处理和分词,分别采用SVM方法和LSTM方法对短评文本和长评文本进行情感极性分析,运用统计和可视化方法研究评论词语、语义网络关系、情感倾向演化、文本内容特征和地域热度分布.以公安剧《狂飙》为例进行实证分析,结果表明所提模型可以合理揭示总体情感热度演变规律,发现观众发表评论的内容偏好、行为规律和地域特征.
Online drama review analysis model based on SVM and LSTM
A vast amount of movie reviews encompass the preferences of a broad audience toward various aspects of film and television works,which can provide decision support for the production and promotion of film and television dramas.This paper pro-poses an analysis model for online comments on TV series:Python is used to crawl comments from media platforms.After data pre-processing and segmentation,the SVM method and LSTM method are used to perform sentiment analysis on short and long com-ment texts respectively.Statistical and visualization methods are used to study comment words,semantic network relationships,emotional tendency evolution,text content features,and geographical popularity distribution.Taking the police drama"The Knock-out"as an example for empirical analysis,the results show that the proposed model can reasonably reveal the overall evolution of emotional popularity,and discover preferences,behavioral patterns,and geographical features of audience comments.

online reviewssentiment analysistext miningvisual analysis

盛蒙蒙、马溯、顾孟钧、沈立峰

展开 >

浙江警察学院计算机与信息安全系,杭州 310018

中国电信股份有限公司浙江分公司,杭州 310000

浙江省杭州市公安局拱墅区分局,杭州 310000

在线评论 情感分析 文本挖掘 可视化分析

浙江省教育厅规划项目浙江省"十四五"研究生课程思政示范课程项目

2023SCG318JG2023002

2024

现代计算机
中大控股

现代计算机

影响因子:0.292
ISSN:1007-1423
年,卷(期):2024.30(2)
  • 13