Content area
Full Text
* Recommender systems are among today's most successful application areas of artificial intelligence. However, in the recommender systems research community, we have fallen prey to a McNamara fallacy to a worrying extent: In the majority of our research efforts, we rely almost exclusively on computational measures such as prediction accuracy, which are easier to make than applying other evaluation methods. However, it remains unclear whether small improvements in terms of such computational measures matter greatly and whether they lead us to better systems in practice. A paradigm shift in terms of our research culture and goals is therefore needed. We can no longer focus exclusively on abstract computational measures but must direct our attention to research questions that are more relevant and have more impact in the real world. In this work, we review the various ways of how recommender systems may create value; how they, positively or negatively, impact consumers, businesses, and the society; and how we can measure the resulting effects. Through our analyses, we identify a number of research gaps and propose ways of broadening and improving our methodology in a way that leads us to more impactful research in our field.
Whenever we visit our favorite media streaming site, check for updates on social media, or shop online, it is highly likely that the content we see is personalized and tailored to our interests and needs. Recommender systems are the technology behind this automated adaptation and personalization, and they are among the most successful applications of artificial intelligence in practice. The broad successful commercial use of modern recommender systems dates to the late 1990s (Schafer, Konstan, and Riedl 1999). Amazon.com was among the early adopters, realizing that there is an enormous potential value in providing customers with automated recommendations. Specifically, they reported vastly improved click-through and conversion rates with personalized recommendations compared with situations where they presented nonpersonalized content (Linden, Smith, and York 2003). Today, recommendations have become a ubiquitous component of our online user experience, for example, on e-commerce sites, video, and music streaming platforms, and on social networks.
The huge success of recommender systems in practice has led to a continuously growing academic interest in this area, and recommender systems have become their own research field over the...