以文字探勘與偏最小平方法結構方程推衍影響消費者接受汽車自動駕駛系統之關鍵要素

No Thumbnail Available

Date

2021

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

現在的車輛已開始將自動駕駛系統(Automated Driving Systems, ADS)的相關功能加進汽車安全輔助系統功能中,為車輛駕駛人及其他用路人創造出更安全且便利的道路生活環境。自動駕駛系統以機器分析判斷,可協助車輛避免危害,以及處理駕駛人所無法顧及突發狀況,甚至自動駕駛系統能直接介入人員駕駛中的車輛,進行控制,來防止意外發生。現階段自動駕駛系統仍處於發展萌芽期,仍有賴更多的創新研發技術來提升自動駕駛車輛的完整性,最終勢將發展成為無需人類駕駛之運載工具,所以消費者對自動駕駛系統車輛之信賴程度,將影響未來自駕車之研究與發展。為廣範擷取感受自動駕駛車輛的相關數據,運用大數據技術分析的方式從母體進行分析,對於次世代自動駕駛系統之開發,至關重要,但少有學者從事相關研究。為跨越此研究缺口,本研究擬導入文字探勘,由社群網路挖掘特定議題或產品、功能或服務之使用者意見,挖掘大眾對目標議題之認知及經驗,然後將這些資訊彙整後摘整,轉換成結構化資訊,進行分析。因此,本研究擬探勘社群網站,搭配與自動駕駛系統技術有關之關鍵字,進行資料採集、擷取特徵,並利用「隱含Dirichlet配置模型(Latent Dirichlet Allocation, LDA)」建立文件主題模型,將主題歸入整合科技接受模型理論之構面後,作為理論架構,並發展對應之假設。最後導入偏最小平方法結構方程模型式(Partial Least Squares Structural Equation Modelling, PLS-SEM)驗證之。本研究將以探勘臺灣社群媒體網站之結果,驗證分析架構之可行性,所得消費者選擇使用自動駕駛系統車輛的行為意圖,包括希望帶來更好的正面效果,使車輛更容易操控,以及發展對內嵌自動駕駛系統之車輛有利的駕駛環境,將正向影響消費者的實際使用行為。研究結論可作為發展內嵌自動駕駛系統車輛設計開發及訂定行銷策略之參據,驗證完善之分析架構,也可以作為擷取影響接受其他突破性或漸進式創新關鍵要素之用。
Automated Driving Systems (ADS) are now being added to vehicle safety assistance systems to create a safer and more convenient road environment for vehicle drivers and other road users. ADS uses machine analysis and judgment to help vehicles avoid hazards and deal with unexpected situations that drivers cannot take into account. Even ADS can directly intervene in the vehicle being driven by a person to control it to prevent accidents. Now the ADS is still in the budding stage of development. It still depends on more innovative research and development technologies to enhance the integrity of self-driving vehicles. It will develop into a vehicle that does not need to be driven by human eventually. Therefore, the level of consumer trust in self-driving vehicles will influence the future research and development of self-driving vehicles.In order to obtain a wide range of data related to the feeling of self-driving vehicles. The use of big data technology to analyze the data from the parent is very important for the development of next-generation automated driving systems, but few scholars have conducted related research. In order to bridge this research gap, this research intends to use textual prospecting techniques to mine user opinions on specific issues or products, features, or services in social networks. In this way, we can obtain the public's knowledge and experience of the target issue. This information is then aggregated and extracted into structured information, and finally analyzed. Therefore, this research intends to conduct social media mining in the website with keywords related to ADS technology to collect data, extract features, and build a document topic model using Latent Dirichlet Allocation (LDA). The topics are then grouped into Unified Theory of Acceptance and Use of Technology (UTAUT) as a theoretical framework, and corresponding assumptions are developed. Finally, Partial Least Squares Structural Equation Modelling (PLS-SEM) is used for validation. In this research, the results of mining Taiwanese social media sites are used to verify the feasibility of the framework. The behavioral intent of consumers choosing to use ADS vehicles was obtained. These include the desire for better positive outcomes, easier handling of the vehicle, and the development of a driving environment conducive to ADS vehicles, which will positively influence the actual usage behavior of consumers. The findings of the research can be used as a reference for the development of embedded ADS vehicle design and marketing strategies. The validated analytical framework can also be used to influence the acceptance of other key elements of radical or incremental innovation.

Description

Keywords

自動駕駛系統, 整合科技接受模型, 文字探勘, 偏最小平方法結構方程式, Automated Driving Systems, UTAUT, Text Mining, PLS-SEM

Citation

Collections