A factual and perceptional framework for assessing diversity effects of online recommender systems

Abstract

Purpose The purpose of this paper is to explore the effects of online recommender systems (RS) on three types of diversity: algorithmic recommendation diversity, perceived recommendation diversity and sales diversity. The analysis distinguishes different recommendation algorithms and shows whether user perceptions match the actual effects of RS on sales. Design/methodology/approach An online experiment was conducted using a realistic shop design, various recommendation algorithms and a representative consumer sample to ensure the generalizability of the findings. Findings Recommendation algorithms show a differential impact on sales diversity, but only collaborative filtering can lead to higher sales diversity. However, some of these effects are subject to how much information firms have about users’ preferences. The level of recommendation diversity perceived by users does not always reflect the factual diversity effects. Research limitations/implications Recommendation and consumption patterns might differ for other types of products; future studies should replicate the study with search or credence goods. The authors also recommend that future research should move from taking a unidimensional measure for the assessment of diversity and employ multidimensional measures instead. Practical implications Online shops need to conduct a more comprehensive assessment of their RS’ effect on diversity, taking into account not only the effects on their sales distribution, but also on users’ perceptions and faith in the recommendation algorithm. Originality/value This study offers a framework for assessing different forms of diversity in online RS. It employs various recommendation algorithms and compares their impact using not just one but three different types of diversity measures. This helps explaining some of the contradictious findings from the previous literature.

Filter bubbles in recommender systems: Fact or fallacy—A systematic review

“…Instead, they observed that as users follow recommendations, their purchasing behavior becomes more similar to that of other users, as indicated by purchase similarity. Similarly, Matt et al (2014) found that perceived suggestion serendipity has a significant positive impact on both perceived preference fit and user satisfaction. Their findings suggest that simply increasing the number of innovative recommendations is not enough.…”

Recommender systems effect on the evolution of users’ choices distribution

“…In online experiments, a web platform offering products to users, while alternative RSs make recommendations to users, is implemented. Then, the measurable effect of the used RSs on the actual users' choices is compared (Matt et al, 2013;Zhu et al, 2018;Senecal et al, 2005). Matt et al (2013) designed a website offering music tracks to users.…”

“…Then, the measurable effect of the used RSs on the actual users' choices is compared (Matt et al, 2013;Zhu et al, 2018;Senecal et al, 2005). Matt et al (2013) designed a website offering music tracks to users. They randomly assigned users to five distinct groups.…”

“…Fortunately, nowadays, there is a growing attention to better understand the effects of RSs on users' decision making, also because it has been shown that RSs suffer from various types of biases, e.g., the popularity bias (Mansoury et al, 2020;Abdollahpouri et al, 2020;Yalcin & Bilge, 2021;Yao et al, 2021;Fleder & Hosanagar, 2009;Matt et al, 2013). Hence, the effect of RSs on users' choice behaviour has become a topic, studied by either conducting online user studies, i.e., by observing the effect of RSs on the choices of real users (Matt et al, 2013;Zhu et al, 2018;Lee & Hosanagar, 2014, or by designing algorithmic offline simulations of the users' choices in the presence of an RS (Fleder & Hosanagar, 2007Szlávik et al, 2011;Yao et al, 2021;Nadolski et al, 2009;Hazrati et al, 2020). While observing the choices of real subjects is probably the most reliable method, it also brings the complexity and cost of building an operational RS and running (many of) such evaluations.…”

Interpretable video tag recommendation with multimedia deep learning framework

“…To assist users and platforms in tagging UGCs, tag recommender systems have been constructed. Recommender systems strive to simplify search processes and reduce information overflow by identifying the information that best suits user-specific preferences (Matt et al , 2019). Tag recommender systems aim at recommending informative tags for UGCs and making the tag selection process easier (Lu et al , 2015).…”