Food recommender systems optimize towards a user’s current preferences. However, appetites may vary, in the sense that users might seek healthy recipes today and look for unhealthy meals tomorrow. In this paper, we propose a novel approach in the food domain to diversify recommendations across different lists to ‘serve’ different users goals, compiled in a multi-list food recommender interface. We evaluated our interface in a 2 (single list vs multiple lists) x 2 (without or with explanations) between-subject user study (N = 366), linking choice behavior and evaluation aspects through the user experience framework. Our multi-list interface was evaluated more favorably than a single-list interface, in terms of diversity and choice satisfaction. Moreover, it triggered changes in food choices, even though these choices were less healthy than those made in the single-list interface.

Full text in ACM Digital Library

There are various biases in recommender systems. Recognizing biases, as well as unfairness caused by problematic biases, is the first step of system optimization. Related studies on algorithmic biases are mainly from the perspective of either items or users. For the latter (we call it “algorithmic user bias”), existing works have considered algorithms’ accuracy performances measured by accuracy metrics like RMSE. However, algorithmic user biases in beyond-accuracy measurements have rarely been studied, even though beyond-accuracy oriented recommendation algorithms have been increasingly investigated, with the purpose of breaking through the personalization limits of traditional accuracy-oriented algorithms (such as the typical “filter bubble” phenomenon). To fill in the research gap, in this work, we employ a large-scale survey dataset collected from a commercial platform, in which more than 11,000 users’ ratings on the recommendation’s 5 performance objectives (i.e., relevance, diversity, novelty, unexpectedness, and serendipity) and 8 kinds of user characteristics (i.e., gender, age, big-5 personality traits, and curiosity) are available. We study user biases of four algorithms (i.e., HOT, Rel-CF, Nov-CF, and Ser-CF) in terms of those five measurements between user groups of the eight user characteristics. We further look into users’ behavior patterns like the preference of using more positive ratings, in order to interpret the observed biases. Finally, based on the observed algorithmic user bias and users’ behavior patterns, we analyze the possible factors leading to the biases and recognize problematic biases that may lead to unfairness.

Full text in ACM Digital Library