Situation awareness for recommender systems

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 situationaware 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 situationaware 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.

Something’s Missing? A Procedure for Extending Item Content Data Sets in the Context of Recommender Systems

“…Recommender systems have become a highly relevant category of decision support systems (Power et al 2015). In particular, in e-commerce, recommender systems are often necessary as users regularly need to make decisions for purchase, consumption or utilization of items (e.g., songs, movies, restaurants or hotels) from a plethora of possible alternatives available in information systems (IS) on e-commerce platforms (Kamis et al 2010;Levi et al 2012;Richthammer and Pernul 2018;Tang et al 2017;Vargas-Govea et al 2011).…”

“…This is, users often do not have the skills and experience to adequately evaluate the large number of available alternatives for making their choice (Ricci et al 2015b;Scholz et al 2017). The resulting problem leaves users of e-commerce IS unable to make effective decisions due to this large volume of information (e.g., items) to which users are exposed to (Hasan et al 2018;Lu et al 2015;Richthammer and Pernul 2018;Scholz et al 2017). In order to address the problem of information overload, the literature suggests for IS providers in ecommerce to incorporate decision support systems, in particular recommender systems, to assist users in their decisionmaking (Bunnell et al 2019;Karimova 2016;Lu et al 2015).…”

“…This data as well as the recommender systems commonly present on such e-commerce platforms aim at guiding users to their individually best item choice, improving user stickiness and increasing platform revenue (Zhou 2020). Such supporting systems are mandatory as customers regularly need to make a choice between a plethora of items (e.g., songs, movies, restaurants, hotels) on ecommerce platforms (Kamis et al 2010;Levi et al 2012;Richthammer and Pernul 2018;Tang et al 2017;Vargas-Govea et al 2011). It is thus hardly surprising that recommender systems in particular have been established as one of the most powerful and popular tools in the field of ecommerce in recent years (Ricci et al 2015a;Scholz et al 2017;Smith and Linden 2017).…”

Music recommendation algorithms based on knowledge graph and multi-task feature learning

“…However, this information explosion is both an opportunity and a challenge for people. Such an information explosion provides access to more information for analysis and use; the challenge is how this information can be accessed effectively for use and how the useful information that existed in a large amount of information can be extracted in a reasonable way 5 . Taking music recommendations as an example, a music library contains various categories, including themes, scenes, moods, eras, genres, and languages.…”

Music Recommendation Algorithms Based On Knowledge Graph and Multi-task Feature Learning

“…The advantage brings more information for us to analyze and use. The challenge is how to effectively access the information for using and digging out the useful information that existed in the large amount of information with a reasonable way [5]. Taking music recommendation as an example, the music library contains various categories including themes, scenes, moods, eras, genres and languages, and themes are divided into various sub-categories such as KTV gold songs, internet songs, love songs and DJ songs.…”