以CNN方法結合信任與忠誠度預測再購行為之研究;Applying CNN on trust and loyalty data to predict customer purchasing behavior

NCUIR > School of Management at National Central University > Graduate Institute of Business Administration > Electronic Thesis & Dissertation > Item 987654321/76851

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/76851

Title:	以CNN方法結合信任與忠誠度預測再購行為之研究;Applying CNN on trust and loyalty data to predict customer purchasing behavior
Authors:	鄭楊叡;Jheng, Yang-Ruei
Contributors:	企業管理學系
Keywords:	信任;忠誠度;再購行為;資料驅動;卷積類神經網路;trust;loyalty;data-driven;CNN;repurchase behavior
Date:	2018-07-06
Issue Date:	2018-08-31 11:51:09 (UTC+8)
Publisher:	國立中央大學
Abstract:	在過去大部分的研究都是採用問卷的方式進行研究。然而，在很多情況下參與者並不願意填寫很長的問卷，導致問卷收集無效。在本研究我們將採用資料驅動的方式從ERP系統中萃取出有價值的資訊。藉由確認信任與忠誠度的關係我們驗證了我們的信任與忠誠度的操作定義。此外，我們採用信任與忠誠度當我們的獨立變數以及藉由卷積類神經網路演算去預測顧客的消費者再購行為。此結果顯示與SVM和回歸方程相比，都有較高的準確率。;The prediction of customer’s repurchasing behavior provides competitive advantage to companies. In e-commerce, the trust and loyalty are key factors influencing repurchasing behavior. Therefore, the aim of this study was to use data driven based methodology to compute trust and loyalty and to predict repurchasing behavior with these values. In the past, most researches replied on questionnaires to collect customer loyalty and trust. The drawback of this approach is that participants may not be willing to fill the lengthy questions which resulted in low data collection rate and even low quality of data being collected. In this study, we extract the information from ERP systems and defined formula to compute trust and loyalties. The values of trust and loyalty are treated as independent variables and being applied to Convolutional Neural Network (CNN) to predict customer’s repurchase behavior. The accuracy of the prediction reached 84%
Appears in Collections:	[Graduate Institute of Business Administration] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	251	View/Open

社群 sharing

Loading...