A comparison of several predictive algorithms for collaborative filtering on multi-valued ratings

Abstract

The basic objective of a predictive algorithm for collaborative filtering (CF) is to suggest items to a particular user based on his/her preferences and other users with similar interests. Many algorithms have been proposed for CF, and some works comparing sub-sets of them can be found in the literature; however, more comprehensive comparisons are not available. In this work, a meaningful sample of CF algorithms widely reported in the literature were chosen for analysis; they represent different stages in the evolutive process of CF, starting from simple user correlations, going through online learning, up to methods which use classification techniques. Our main purpose is to compare these algorithms when applied on multi-valued ratings.Experiments were conducted on three well-known datasets with different characteristics, using two protocols and four evaluation metrics, representing coverage, accuracy, reliability and agreement of predictions with respect to real values. Results from such experiments showed that the memorybased method is a good option because its results are more precise and reliable compared with the other methods. Online Learning methods exhibit a good level of accuracy with low variation, which makes them reliable models. On the other hand, Support Vector Machines generate predictions with acceptable agreement; however, their accuracy depends on the characteristics of the input data. Finally, DepenPermission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

IKT-gestützte Unternehmenskommunikation: Einsatz von Recommender-Systemen zum Management von Unternehmensportalen

“…In einem zweiten Schritt werden anhand dieses Modells Empfehlungen für jeden Nutzer berechnet. Bekannte Beispiele solcher Algorithmen sind das Cluster-Verfahren, das Bayesian Network Model (Breese et al 1998, S. 48) oder das Dependency Network (Calderón-Benavides et al 2004, S. 1038. Vorteile dieses Ansatzes liegen in der Skalierbarkeit, wobei dieser eine schlechte Prognosegüte gegenüber steht.…”

Identifying User and Group Information From Collaborative Filtering Datasets

“…The reason for choosing a traditional memory-based collaborative filtering approach is that it has been shown to perform well in comparison to many other collaborative filtering techniques. 4,9 Also, as can be seen from the descriptions of the approaches in Secs. 3.1 and 3.2, it is quite similar to the spreading activation approach outlined.…”

Collaborative Filtering on Data Streams

“…The off-line batch process is followed by most collaborative filtering algorithms to provide recommendations in real-time, but this produces an outdated model where the quality of the recommendations is low. Online algorithms in Collaborative Filtering [7] are more suitable for handling an incoming data stream since these are fast, incremental and there is no need to store all the previously seen examples. The first online algorithm applied to collaborative filtering was the Weighted Majority Prediction (WMP) [8], Delgado et al [9] extended this approach for multi-valued ratings.…”