Situation awareness for recommender systems
- PDF / 1,922,403 Bytes
- 24 Pages / 439.37 x 666.142 pts Page_size
- 102 Downloads / 209 Views
Situation awareness for recommender systems Christian Richthammer1 · Günther Pernul1
© Springer Science+Business Media, LLC, part of Springer Nature 2018
Abstract One major shortcoming of traditional recommender systems is their inability to adjust to users’ short-term preferences resulting from varying situation-specific factors. To address this, we propose the notion of situation-aware recommender systems, which are supposed to autonomously determine the users’ current situation based on a multitude of contextual side information and generate truly personalized recommendations. In particular, we develop a situation awareness model for recommender systems, include it in a situation-aware recommendation process, and derive generic design steps for the design of situation-aware recommender systems. The feasibility of these concepts is demonstrated by directly employing them for the development and implementation of a music recommender system for everyday situations. Moreover, their meaningfulness is shown by means of an empirical user study. The outcomes of the evaluation indicate a significant increase in user satisfaction compared to traditional (i.e. non-situation-aware) recommendations. Keywords Recommender systems · Situation awareness · Context awareness · Contextual side information · Situation-aware recommender systems
1 Introduction The rise of the World Wide Web has made sharing and accessing various kinds of information easier and faster than ever before, resulting in considerable benefits for its users (including both content providers and consumers). Long since, however, this trend has reached a point where the increasing amounts of alternatives and information overwhelm the users in the course of their decision making processes [26]. Recommender systems are intended to solve this phenomenon of information overload by making users aware of only those items they are probably interested in [13, 14]. However, one basic assumption underlying most recommender systems still is that users’ preferences do not change very fast [7]. Users’ general interests * Christian Richthammer [email protected] 1
University of Regensburg, Universitätsstraße 31, 93053 Regensburg, Germany
13
Vol.:(0123456789)
C. Richthammer, G. Pernul
may in fact be relatively stable but their preferences may also be influenced by many additional situation-specific factors, which are generally referred to as “context” [7]. Thus, users’ preferences in a specific situation may differ greatly from their general interests. Being unable to take this into account (i.e. to adjust to short-term preferences) constitutes a major shortcoming of traditional recommendation techniques. Against this background, the area of context-aware recommender systems has been established. Obviously, the main goal of the research on context-aware recommender systems is to increase the quality of recommendations, and thus the users’ satisfaction with the service, by taking advantage of the correlations between contextual side information and rat
Data Loading...
