Design change prediction based on social media sentiment analysis

IF 1.7 3区工程技术 Q3 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Ai Edam-Artificial Intelligence for Engineering Design Analysis and Manufacturing Pub Date : 2022-07-27 DOI:10.1017/S0890060422000129

E. C. Koh

{"title":"Design change prediction based on social media sentiment analysis","authors":"E. C. Koh","doi":"10.1017/S0890060422000129","DOIUrl":null,"url":null,"abstract":"Abstract The use of artificial intelligence (AI) techniques to uncover customer sentiment is not uncommon. However, the integration of sentiment analysis with research in design change prediction remains an untapped potential. This paper presents a method that uses social media sentiment analysis to identify opportunities for design change and the set of product components affected by the change. The method builds on natural language processing to determine change candidates from textual data and uses dependency modeling to reveal direct and indirect change propagation paths arising from the change candidates. The method was applied in a case example where 3665 YouTube comments on a diesel engine were analyzed. Based on the results, two engine components were recommended for design change with six others predicted as likely to be affected through change propagation. The findings suggest that the method can be used to aid decision quality in product planning through a better understanding of the change impact associated with the opportunities identified.","PeriodicalId":50951,"journal":{"name":"Ai Edam-Artificial Intelligence for Engineering Design Analysis and Manufacturing","volume":"36 1","pages":""},"PeriodicalIF":1.7000,"publicationDate":"2022-07-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ai Edam-Artificial Intelligence for Engineering Design Analysis and Manufacturing","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1017/S0890060422000129","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}

引用次数: 2

Abstract

Abstract The use of artificial intelligence (AI) techniques to uncover customer sentiment is not uncommon. However, the integration of sentiment analysis with research in design change prediction remains an untapped potential. This paper presents a method that uses social media sentiment analysis to identify opportunities for design change and the set of product components affected by the change. The method builds on natural language processing to determine change candidates from textual data and uses dependency modeling to reveal direct and indirect change propagation paths arising from the change candidates. The method was applied in a case example where 3665 YouTube comments on a diesel engine were analyzed. Based on the results, two engine components were recommended for design change with six others predicted as likely to be affected through change propagation. The findings suggest that the method can be used to aid decision quality in product planning through a better understanding of the change impact associated with the opportunities identified.

查看原文本刊更多论文

基于社交媒体情感分析的设计变更预测

摘要使用人工智能（AI）技术来揭示客户情绪并不罕见。然而，情感分析与设计变更预测研究的结合仍然是一个尚未开发的潜力。本文提出了一种使用社交媒体情绪分析来识别设计变更的机会以及受变更影响的产品组件集的方法。该方法建立在自然语言处理的基础上，从文本数据中确定候选变化，并使用依赖性建模来揭示由候选变化引起的直接和间接变化传播路径。该方法应用于一个案例，其中分析了3665条关于柴油发动机的YouTube评论。根据结果，建议对两个发动机部件进行设计变更，另外六个部件预计可能会受到变更传播的影响。研究结果表明，该方法可以通过更好地理解与所确定的机会相关的变化影响，来帮助产品规划中的决策质量。

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

Ai Edam-Artificial Intelligence for Engineering Design Analysis and Manufacturing 工程技术-工程：制造

CiteScore

4.40

自引率

14.30%

发文量

审稿时长

>12 weeks

期刊介绍： The journal publishes original articles about significant AI theory and applications based on the most up-to-date research in all branches and phases of engineering. Suitable topics include: analysis and evaluation; selection; configuration and design; manufacturing and assembly; and concurrent engineering. Specifically, the journal is interested in the use of AI in planning, design, analysis, simulation, qualitative reasoning, spatial reasoning and graphics, manufacturing, assembly, process planning, scheduling, numerical analysis, optimization, distributed systems, multi-agent applications, cooperation, cognitive modeling, learning and creativity. AI EDAM is also interested in original, major applications of state-of-the-art knowledge-based techniques to important engineering problems.