A Study of Review Hot Words Extraction Technology Based on the LSTM Web Review Validity Model

2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI) Pub Date : 2022-11-05 DOI:10.1109/CISP-BMEI56279.2022.9979915

Yue Li, Yuanhui Yu, Yaxian Su, Tao Yang, Xu Zhang, Jiandong Shi

{"title":"A Study of Review Hot Words Extraction Technology Based on the LSTM Web Review Validity Model","authors":"Yue Li, Yuanhui Yu, Yaxian Su, Tao Yang, Xu Zhang, Jiandong Shi","doi":"10.1109/CISP-BMEI56279.2022.9979915","DOIUrl":null,"url":null,"abstract":"The huge amount of online review text data brings a great challenge to the extraction of valid information and hot words extraction work. This paper addresses this problem and designs a study on hot words extraction based on a Bidirectional LSTM(Long short term memory) online review text validity model. Firstly, data pre-processing is performed on the data set of online review texts collected by crawlers, secondly, a validity model of online review texts based on LSTM neural network is established to filter the valid online review texts, and finally, hot words are extracted from the valid review texts to get the hot words containing valuable information. In this paper, we take hotel review text as an example to conduct experiments, and the experimental results prove that the accuracy of LSTM online review text validity model reaches 90%, the loss value reaches 0.2, and the screening of valid text for hot words extraction achieves good results.","PeriodicalId":198522,"journal":{"name":"2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CISP-BMEI56279.2022.9979915","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

The huge amount of online review text data brings a great challenge to the extraction of valid information and hot words extraction work. This paper addresses this problem and designs a study on hot words extraction based on a Bidirectional LSTM(Long short term memory) online review text validity model. Firstly, data pre-processing is performed on the data set of online review texts collected by crawlers, secondly, a validity model of online review texts based on LSTM neural network is established to filter the valid online review texts, and finally, hot words are extracted from the valid review texts to get the hot words containing valuable information. In this paper, we take hotel review text as an example to conduct experiments, and the experimental results prove that the accuracy of LSTM online review text validity model reaches 90%, the loss value reaches 0.2, and the screening of valid text for hot words extraction achieves good results.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

基于LSTM Web评论有效性模型的评论热词提取技术研究

海量的在线评论文本数据给有效信息的提取和热词提取工作带来了极大的挑战。针对这一问题，本文设计了一种基于双向LSTM(长短期记忆)在线评论文本有效性模型的热词提取研究。首先对爬虫收集的在线评论文本数据集进行数据预处理，然后建立基于LSTM神经网络的在线评论文本有效性模型对有效的在线评论文本进行过滤，最后从有效的在线评论文本中提取热词，得到包含有价值信息的热词。本文以酒店点评文本为例进行实验，实验结果证明LSTM在线点评文本有效性模型的准确率达到90%，损失值达到0.2，对有效文本的筛选进行热词提取取得了较好的效果。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)

自引率

0.00%

发文量

期刊最新文献

Point Cloud Registration based Track Correlation Method An End-to-end Image Feature Representation Model of Pulmonary Nodules Vision-Guided Speaker Embedding Based Speech Separation Analysis and Simulation of Interference Effects on CSK Modulation Systems Sentiment Analysis Based On Deep Residual Bidirectional Gated Recurrent Unit Neural Networks