Recommender Systems

Leimstoll/Stormer

Collaborative Recommender Systems

COLLABORATIVE RECOMMENDER SYSTEMS FOR ONLINE SHOPS

Uwe Leimstoll University of Applied Sciences Northwestern Switzerland uwe.leimstoll@fhnw.ch

Henrik Stormer University of Fribourg, Switzerland henrik.stormer@unifr.ch

Abstract

Recommender systems are often used in electronic shops in order to suggest similar or related products, potentially interesting products for a given customer or a set of products for a marketing campaign. Most recommender systems use the collaborative filtering method in order to provide the personalization information. The collaborative filtering method is a very efficient and convenient way of achieving personalization as there is no need to introduce semantic information about the products or to manually link products and users together. In the last years, a number of optimizations for collaborative filtering techniques have been developed. This paper collects the ideas and shows which of them could be integrated successfully in order to optimize a collaborative recommender system for online shops.

Keywords

Recommender Systems, Collaborative Filtering, Recommendations, Reference Process, Online Shops

Introduction

In the past years, the number of personalization applications has strongly increased, especially in the field of electronic commerce where personalization becomes an important success factor (Manber et al. 2000, Schubert/Koch 2002). The term personalization means the filtering of information for each particular person in order to provide customers a customized or personalized interaction with a company’s products, services, web site and employees (Deitel et al. 2001). The personalization concept is a fundamental requirement for online shops. In contrast to traditional shops, electronic shops cannot provide the personal contact and the individual consultation, which are important means of the customer relationship management. Recommender systems are one...