以社群探勘、偏最小平方結構方程與多準則決策分析歸納影響中小企業導入網路儲存裝置之關鍵因素

Derivations of Factors Influencing the Adoption of Network Attached Storage by Social Media Mining, PLS-SEM and MCDM Methods

童敬霖(Tung Ching-Lin)

指導教授 ： 黃啟祐

國立臺灣師範大學/科技與工程學院/工業教育學系科技應用管理碩士在職專班/碩士(2022年)

https://doi.org/10.6345/NTNU202205653

摘要

在大數據時代，儲存與備份資料能力，成為大部份企業有效運用數據資產的關鍵成功要素。網路儲存裝置（Networked Attached Storage，NAS）具備數據儲存之安全性、可擴展性與靈活性等特質，為可滿足中小企業數據儲存、備份與協作需求之最佳解決方案。雖然網路儲存裝置極為重要，然而，探討影響中小企業導入相關儲存裝置的研究極為有限。為了跨越此研究缺口，本論文擬透過社群媒體探勘與專家問卷，導入第二代整合科技接受模型（Unified Theory of Acceptance and Use of Technology 2，UTAUT 2），推衍影響中小企業導入網路儲存裝置之關鍵要素。
本研究首先以文字探勘（Text Mining）技術擷取社群網站 Dcard，與網路儲存裝置有關之貼文，並以基於隱含狄利克雷分布 (Latent Dirichlet Allocation，LDA) 之主題模型分析（Topic Modeling），探勘貼文中蘊含之主題，其次，以階層式集群分析法（Hierarchical Cluster Analysis）將主題分群後，將各主題歸入第二代整合科技接受模式之構面，並以偏最小平方結構方程模型（Partial Least Squares Structural Equation Modeling，PLS-SEM）驗證之。本研究同時將邀集專家，以基於決策實驗室分析法（Decision Making Trial and Evaluation Laboratory，DEMATEL）之分析網路流程（Analytic Network Process，ANP），又稱 DANP，推衍影響中小企業導入網路儲存裝置的最關鍵因素，並比較社群網路探勘與專家意見法兩者結果之差異。
依據社群網路探勘與結構方程模型驗證之結果，「行為意圖」和「價格價值」與中小企業導入網路儲存裝置的關連性最高。而以DANP彙整專家意見的結果，除了「行為意圖」和「價格價值」之外，「績效預期」亦將影響中小企業導入網路儲存裝置。
本研究結果可提供中小企業，作為成功佈署網路儲存裝置之基礎，而整合社群媒體探勘、偏最小平方結構方程模型與多準則決策分析（Multiple Criteria Decision Making，MCDM）之研究架構，可作為探討相關領域影響導入新科技之分析方法。

關鍵字

網路儲存裝置 ； 文字探勘 ； 偏最小平方結構方程模型 ； 多準則決策 ； 延伸型整合科技接受模式

並列摘要

In the era of big data, the capabilities of storing data have emerged as a key success factor for the efficient manipulation of data. Networked Attached Storage (NAS) is a hardware solution to address the data storage, backup, and collaboration needs of small and medium-sized enterprises (SMEs) for enhancing the security, scalability, and flexibility of data storage. Despite the significance of NAS, little research investigated the dominant factors related to the adoptions of NAS. To cross the research gap, this thesis aims to derive the dominant factors based on the Unified Theory of Acceptance and Use of Technology 2 (UTAUT 2) from the mining of social media as well as the collection of experts’ opinions by using the Multiple Criteria Decision Making (MCDM) based methods.
The research first extracted the posts related to network storage devices from the social networking site Dcard using Text Mining, and explored the topics embedded in the posts using a Topic Model technique based on the Latent Dirichlet Allocation (LDA). Then, the Hierarchical Cluster Analysis (HCA) was used to verify the structural equation model by using Partial Least Squares Structural Equation Modeling (PLS-SEM), after the topics were clustered according to the structure of the second-generation UTAUT 2 model. This research also invites experts to use the Analytic Network Process (ANP) based on the Decision Making Trial and Evaluation Laboratory (DEMATEL) to derive the most critical factor affecting the adoption of networked storage devices by SMEs.
According to the results of social network mining and structural equation model verification, "behavioral intention" and "price value" have the highest correlation with the introduction of network storage devices by SMEs. In addition to "behavioral intention" and "price and value", "performance expectation" will also affect the introduction of networked storage devices by SMEs. The findings of this research provide SMEs with a foundation for the successful deployment of networked storage devices, and the incorporation of social media analysis, partial least square structural equation modeling, and the MCDM research framework can be used as an analytic technique to examine the effects of the introduction of new technologies in related fields.

並列關鍵字

Networked Attached Storage ； Text Mining ； Partial Least Squares Structural Equation Modeling ； Multiple Criteria Decision Making ； Unified Theory of Acceptance and Use of Technology 2

參考文獻

Albalawi, R., Yeap, T. H., & Benyoucef, M. (2020). Using topic modeling methods for short-text data: A comparative analysis. Frontiers in Artificial Intelligence, 3, 42.

Alfadda, H. A., & Mahdi, H. S. (2021). Measuring students’ use of zoom application in language course based on the technology acceptance model (TAM). Journal of Psycholinguistic Research, 50(4), 883-900.

Allahyari, M., Pouriyeh, S., Assefi, M., Safaei, S., Trippe, E. D., Gutierrez, J. B., & Kochut, K. (2017). A brief survey of text mining: Classification, clustering and extraction techniques. arXiv preprint arXiv:1707.02919.

Büyüközkan, G., & Güleryüz, S. (2016). An integrated DEMATEL-ANP approach for renewable energy resources selection in Turkey. International journal of production economics, 182, 435-448.

Bai, X., Zhang, X., Li, K. X., Zhou, Y., & Yuen, K. F. (2021). Research topics and trends in the maritime transport: A structural topic model. Transport Policy, 102, 11-24. doi:https://doi.org/10.1016/j.tranpol.2020.12.013