Text Summarization for Thai Food Reviews using Simplified Sentiment Analysis

2021 18th International Joint Conference on Computer Science and Software Engineering (JCSSE) Pub Date : 2021-06-30 DOI:10.1109/JCSSE53117.2021.9493839

P. Porntrakoon, Chayapol Moemeng, P. Santiprabhob

{"title":"Text Summarization for Thai Food Reviews using Simplified Sentiment Analysis","authors":"P. Porntrakoon, Chayapol Moemeng, P. Santiprabhob","doi":"10.1109/JCSSE53117.2021.9493839","DOIUrl":null,"url":null,"abstract":"The review of a customer who had experienced using the products and services helps others make the purchasing decision. Especially in the food and beverage domain, Thai reviewers frequently write long reviews in multiple dimensions, requiring a longer time to read and understand the reviews' opinions. The Thai sentiment analysis can analyze the review's sentiment in food, environment, and service dimensions and generate the review summary accordingly. Readers can save time in reading only the review summary and perceive the sentiments in the original review. In this paper, we propose a method to analyze the sentiment in Thai food review and generate a summary with the positive and negative sentiments in the review. The results show that our proposed method can generate 1,876 review summaries from 4,000 original reviews, which is equivalent to 46.68%. The total number of words in the review summaries is only 5.13% of the original reviews, with an average accuracy of 65.25% in Type 1 and 85.05% in Type 2.","PeriodicalId":437534,"journal":{"name":"2021 18th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 18th International Joint Conference on Computer Science and Software Engineering (JCSSE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/JCSSE53117.2021.9493839","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 1

Abstract

The review of a customer who had experienced using the products and services helps others make the purchasing decision. Especially in the food and beverage domain, Thai reviewers frequently write long reviews in multiple dimensions, requiring a longer time to read and understand the reviews' opinions. The Thai sentiment analysis can analyze the review's sentiment in food, environment, and service dimensions and generate the review summary accordingly. Readers can save time in reading only the review summary and perceive the sentiments in the original review. In this paper, we propose a method to analyze the sentiment in Thai food review and generate a summary with the positive and negative sentiments in the review. The results show that our proposed method can generate 1,876 review summaries from 4,000 original reviews, which is equivalent to 46.68%. The total number of words in the review summaries is only 5.13% of the original reviews, with an average accuracy of 65.25% in Type 1 and 85.05% in Type 2.

查看原文本刊更多论文

使用简化情感分析的泰国食物评论文本摘要

对有使用产品和服务经验的顾客的评价有助于其他人做出购买决定。尤其是在食品和饮料领域，泰国的评论者经常写多维度的长篇评论，需要更长的时间来阅读和理解评论者的意见。泰国情绪分析可以分析评论在食物、环境和服务维度上的情绪，并生成相应的评论摘要。读者可以节省时间，只阅读评论摘要，并感知原始评论的情绪。在本文中，我们提出了一种方法来分析泰国菜评论中的情绪，并产生一个总结与评论中的积极和消极的情绪。结果表明，我们提出的方法可以从4000篇原始评论中生成1876篇评论摘要，相当于46.68%。评论摘要的总字数仅为原始评论的5.13%，类型1的平均准确率为65.25%，类型2的平均准确率为85.05%。

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分钟内获得全文求助全文

来源期刊

2021 18th International Joint Conference on Computer Science and Software Engineering (JCSSE)

自引率

0.00%

发文量