個人化新產品推薦系統：行為基礎與認知基礎模式之比較與整合

產品推薦系統是落實顧客關係管理之一對一行銷決策支援系統。在過去的研究中，產品推薦系統大致可分成兩類，一是合作篩選系統（collaborative filtering），根據產品間之相關性進行推薦；另是內容篩選系統（content filtering），根據顧客自我顯現（self-explicated）的偏好結構進行產品推薦。這兩種推薦系統皆屬於總合層次，前者需要大量購買紀錄方能得到較穩定的產品相關結構，後者則需要集合相似的顧客資料方能得到穩定的偏好結構，故皆無法充分反映顧客的異質性。此外，合作篩選系統無法分析顧客選購產品的理由或偏好結構，故無法進行新產品的推薦，只能就現有產品進行關連性銷售（cross-selling）傳統的內容篩選系統無法探討個人偏好結構與人格特質的關餘，故無法進一步應用於對新客戶的產品推薦，只能針對具有購買紀錄的現有顧客進行產品推薦。有鑑於此，本研究擬以個人偏好結構（individual preference structure）為基礎，設計一套同時適用於新產品與新顧客的產品推薦系統，期使顧客關係管理之觀念能真正落實於企業之日常操作系統。本研究根據顧客的主觀認知與客觀行為，分別建立認知基礎與行為基礎的個人化偏好結構，試圖比較二者是否具一致性，從而評估以主觀認知為主的產品推薦系統的有效性。在主觀認知方面，為降低受訪者的資訊處理負擔，本研究採取自我顯現偏好，以問卷衡量之；在客觀行為方面，本研究以相同受訪者的交易紀錄及虛擬產品構成該位顧客的產品考慮集合，再以層級貝氏Probit模式估計個人化的偏好結構。實證結果顯示，想知基礎和行為基礎的偏好結構不具一致性，這也帶出了以往利用問卷方式去衡量購買態度，進而預測購買行為的方式將會發生「適用性」及「預測效度」的問題。

關鍵字

新產品推薦系統；偏好結構分析；層級貝氏模式

並列摘要

Product recommendation system is a one-to-one marketing decision supporting system, which put customer relationship management into practice. In the past research, product recommendation systems fall into two classes. One is so-called collaborative filtering, which makes recommendations depending on correlation structure of all products. The other is known as content filtering, which makes recommendations on the basis of consumer's self-explicated preference structure for product attributes. These two systems both are at aggregate level. The former needs dense purchases history data to get more stable correlation structure of products; the latter also needs pooling data sets of homogeneous customers to get more stable estimation of preference structure. Besides, for collaborative filtering providing few reasons for a recommendation and little information about preference structure of customers, it lacks the ability to make entirely new product recommendation but just make cross-selling among exiting pro ducts. The traditional content filtering does not analyze the relationship between individual preference structure and personality, so it cannot be applied to make recommendation for entirely new customers who didn't provide any preference information. Therefore, this paper will design a product recommendation system suitable for both new items and new customers at individual level such that we can put the concept of customer relationship management into regular business operation systems. We study recognition-based and behavior-based individual preference structure respectively depending on subject recognition and objective behaviors of consumers. We try to compare the two kinds of preference structures and then evaluate the validity of the recognition-based product recommendation system. The empirical results show that the two preference structures are not consistent with each other. It points out the problem of validity of questionnaire survey to measure consumers' subjective attitude and moreover to predict their future purchase behaviors.

並列關鍵字

product recommendation system ； preference structure analysis ； hierarchical bayes models

參考文獻

Allenby, Greg M.,Arora, Neeraj,Ginter, James L.(1995).Incorporating Prior Knowledge into the Analysis of Conjoint Studies.Journal of Marketing Research.32,152-162.

Google Scholar

Ansari, Asim,Essegaier, Skander,Kohli, Rajeev(2000).Internet Recommendation Systems.Journal of Marketing Research.37,363-375.

Google Scholar

Bucklin, Randolph E.,Srinivasan, V.(1991).Determining Interbrand Substitutability Through Survey Measurement of Consumer Preference Structures.Journal of Marketing Research.28,58-71.

Google Scholar

Fishbein, Martin(1963).An Investigation of the Relationships Between Beliefs About an Object and the Attitude Toward that Object.Human Relations.16,233-240.