Informed recommender: Basing recommendations on consumer product reviews

被引：106

作者：

Aciar, Silvana

Zhang, Debbie

Simoff, Simeon

Debenham, John

机构：

[1] Univ Girona, Dept Elect Comp Sci & Automat Control, Girona, Spain

[2] Univ Technol Sydney, Fac Informat Technol, Dept Comp Syst, Broadway, NSW 2007, Australia

[3] Univ Technol Sydney, E Markets Res Grp, Broadway, NSW 2007, Australia

[4] Univ Technol Sydney, Fac Informat Technol, Dept Software Engn, Broadway, NSW 2007, Australia

来源：

IEEE INTELLIGENT SYSTEMS | 2007年 / 22卷 / 03期

关键词：

D O I：

10.1109/MIS.2007.55

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

An Informed recommender has been developed that uses prioritized consumer product reviews to make recommendations and maps each piece of review comment using text mining techniques. The recommender uses ontology that provides a controlled vocabulary and relationships among words to describe the consumer's skill level and experience with the product in the review comment in the system. The recommender process involves the useful information to be useful for the recommendation process. The recommender system makes a recommendation based on the ontology data, and therefore, recommendation quality depends on accurately mapping the proper knowledge from the semantic features in the review comments. When a user requests an evaluation of a particular product based on certain features, the overall feature quality is calculated from reviews containing the valuation of this feature.

引用

页码：39 / 47

页数：9

共 4 条

[1] Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions [J].

Adomavicius, G ;

Tuzhilin, A .

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2005, 17 (06) :734-749

[2]

Weiss SM, 2004, TEXT MINING PREDICTI

[3]

Wietsma R.T.A., 2005, P WORKSH PERV MOB IN, P15

[4]

Zhang D, 2006, LECT NOTES ARTIF INT, V3755, P176

← 1 →