應用適應性類神經模糊理論於綠色產品之推薦

Application of Adaptive Neuro-Fuzzy theory to the Recommendation of Green Products

王生鼎(Sheng-Ting Wang)

指導教授 ： 李英聯

朝陽科技大學/理工學院/工業工程與管理系/碩士(2010年)

摘要

摘要
由於環境受污染的情況越來越嚴重，導致環保議題成為人類所關注的焦點。現今市面上綠色產品數量漸增，加上消費者環保意識提升，許多企業開始把環境相關元素加入產品研發和行銷策略中，使得綠色行銷成為了新興的行銷策略。推廣綠色產品除了讓消費者多一種選擇，也能減少對環境的負擔。
本研究欲了解消費者的綠色認知、態度及購買綠色產品偏好等行為，使消費者能經由推薦系統做適合的購買決策。研究步驟是先建置一個消費者調查的網站，內容包含綠色產品滿意度調查和受測者背景題項，其中背景題項作為群聚分析的依據，透過群聚分析將受測者分為若干群，以綠色產品滿意度調查所蒐集的資料，應用適應性類神經模糊推論系統分別為每一群集及每位受測者建構適應性類神經模糊模型，並運用產品屬性聯合分析方法，為各群集找出產品的重要屬性。
本研究發現為個別受測者建置之適應性類神經模糊模型，其預測值與實際值之間的差距不大且準確性高，可嘗試運用於網路推薦系統當中，作為個人化推薦使用。於產品屬性聯合分析，環境關心程度高的消費者，參與綠色活動踴躍並且重視產品的綠色程度，而對於環境關心程度低的消費者，參與綠色活動較為被動，對產品的綠色程度較不重視。

關鍵字

綠色行銷 ； 綠色消費者行為 ； 推薦系統 ； 群聚分析技術 ； 適應性類神經模糊 ； 聯合分析

並列摘要

Abstract
Owing the increasingly serious to environmental pollution, environmental issues have become the focus of attention. With the increasing quantity of green products driven by environment-savvy consumers, many companies begin to add elements of the environment-related R&D and marketing strategy, such that green marketing has become an important marketing strategy. Promotion of green products not only offers consumers more options, but also reduces the burden on the environment.
This study is to construct a green product recommendation system based upon the analysis of consumers’ green awareness, environmental attitude, and product preferences. A web-based survey of product preferences and consumers’ background is built. Consumers’ background is used as the basis for cluster analysis. Then, the adaptive neuro-fuzzy inference system (ANFIS) is applied to build individual- and cluster-based recommendation models to illustrate the applicability and capability of the ANFIS as the basis of recommendation system. In addition, Conjoint Analysis is employed to identify product preferences for every cluster.
The result shows that the gap between predicted and observed values is small in individual-based models with high accuracy. It could be used in personalized recommendation for online stores. Conjoint Analysis shows that the consumers who have high degree of environmental concern are active participants of green activities and prefer green products. However, the consumers who have low degree of environmental concern are less active in green activities and care less about green products.

並列關鍵字

Green Consumer Behavior ； Green Marketing ； Recommendation System ； Cluster Technology ； Adaptive Neuro-Fuzzy Inference System (ANFIS) ； Conjoint Analysis

參考文獻

陳弘庭，(2001)，模糊分群方法、語意變數、分群群數關係之研究─以市場區隔為例，碩士論文，國立成功大學 工業管理科學系。

Adomavicius, G., & Tuzhilin, A. (2005). Toward the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions. In IEEE, Vol 17(No.6).

Chamorro, A., & Bañegil, T. M. (2006). Green marketing philosophy: a study of Spanish firms with ecolabels. Corporate Social Responsibility and Environmental Management, 13(1), 11-24. doi:10.1002/csr.83

Chang, F., & Chang, Y. (2006). Adaptive neuro-fuzzy inference system for prediction of water level in reservoir. Advances in Water Resources, 29(1), 1-10. doi:10.1016/j.advwatres.2005.04.015

Cho, Y. H., Kim, J. K., & Kim, S. H. (2002). A personalized recommender system based on web usage mining and decision tree induction. Expert Systems with Applications, 23(3), 329-342. doi:10.1016/S0957-4174(02)00052-0