情感效應對社群網絡推薦的影響;Effects of Sentiment on Recommendations in Social Network

NCUIR > School of Management at National Central University > Graduate Institute of Business Administration > Electronic Thesis & Dissertation > Item 987654321/80548

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/80548

Title:	情感效應對社群網絡推薦的影響;Effects of Sentiment on Recommendations in Social Network
Authors:	雷鴻村;Lei, Hong Tsuen
Contributors:	企業管理學系
Keywords:	電子商務;social networking site;social commerce;Plurk recommendation system
Date:	2019-07-03
Issue Date:	2019-09-03 14:44:23 (UTC+8)
Publisher:	國立中央大學
Abstract:	隨著社群網站的日益興起，在網路世界進行交流的人數急速增長，也逐漸成為將產品推薦給使用者的媒介之一。過去傳統微網誌(Microblog)的產品推薦方式，大多只在乎使用者個人喜好與興趣，卻忽略了其他可能的影響因素，例如使用者情緒字詞對推薦效果影響的研究。本研究採用情感字詞數據庫從微博貼文中提取情感相關方面字詞，然後用於判斷各種情感相關的字詞對產品推薦的影響。本研究發現包含有強烈情緒的貼文比僅含有溫和情緒的貼文獲得的點擊次數更多。另外，與包含多個情緒字詞貼文的推薦比僅包含一類情緒詞的貼文更有效。本研究還獲得另一個結果是，歸類為具有負向情緒字詞的貼文推薦比分類為具有正向情緒字詞的貼文獲得更多的點擊。此外，包含隱性情緒詞的貼文比在微博中包含顯性情緒詞的貼文獲得的點擊次數更多。本研究可以幫助使用Plurk或類似微博平台的產品或服務營銷人員確定如何將有限的財務資源集中在潛在的在線客戶上，以實現最大的電子商務銷售收入。 ;With the rise of social networking sites, the number of people communicating in the online world has grown rapidly, and SNS has gradually become one of the mediums for recommending products or services to users. In the past, the traditional Microblog product recommendation method mostly only concerned the user′s personal preferences and interests, but ignored other possible influencing factors, such as: the user′s emotional words in the post content. This thesis adopted the sentiment word database to extract sentiment-related aspects from microblog posts, which were then used to investigate the effectiveness of various sentiment-related components on product recommendations. This thesis found that the posts containing strong sentiments were discovered to receive more clicks than posts containing mellow sentiments. Recommendations associated with posts containing more than one positive sentiment words are more effective than posts containing only one such word. This thesis also demonstrated that posts classified as having negative polarity receive more clicks than those classified as having positive polarity. Additionally, posts containing implicit sentiment words receive more clicks than those containing explicit sentiment words in microblog. This thesis could assist product or service marketers who use Plurk or similar microblogging platforms to determine how to focus their limited financial resources on potential online customers to achieve maximum sales revenue.
Appears in Collections:	[Graduate Institute of Business Administration] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	203	View/Open

社群 sharing

Loading...