Michael D. Ekstrand

Michael D. Ekstrand

LensKit is an open-source toolkit for building, researching, and learning about recommender systems. First released in 2010 as a Java framework, it has supported diverse published research, small-scale production deployments, and education in both MOOC and traditional classroom settings. In this paper, I present the next generation of the LensKit project, re-envisioning the original tool's objectives as flexible Python package for supporting recommender systems research and development. LensKit for Python (LKPY) enables researchers and students to build robust, flexible, and reproducible experiments that make use of the large and growing PyData and Scientific Python ecosystem, including scikit-learn, TensorFlow, and PyTorch. To that end, it provides classical collaborative filtering implementations, recommender system evaluation metrics, data preparation routines, and tools for efficiently batch running recommendation algorithms, all usable in any combination with each other or with other Python software. This paper describes the design goals, use cases, and capabilities of LKPY, contextualized in a reflection on the successes and failures of the original LensKit for Java software.

Alan Said, Maria Soledad Pera, Michael D. Ekstrand

In 2011, Xavier Amatriain sounded the alarm: recommender systems research was "doing it all wrong" [1]. His critique, rooted in statistical misinterpretation and methodological shortcuts, remains as relevant today as it was then. But rather than correcting course, we added new layers of sophistication on top of the same broken foundations. This paper revisits Amatriain's diagnosis and argues that many of the conceptual, epistemological, and infrastructural failures he identified still persist, in more subtle or systemic forms. Drawing on recent work in reproducibility, evaluation methodology, environmental impact, and participatory design, we showcase how the field's accelerating complexity has outpaced its introspection. We highlight ongoing community-led initiatives that attempt to shift the paradigm, including workshops, evaluation frameworks, and calls for value-sensitive and participatory research. At the same time, we contend that meaningful change will require not only new metrics or better tooling, but a fundamental reframing of what recommender systems research is for, who it serves, and how knowledge is produced and validated. Our call is not just for technical reform, but for a recommender systems research agenda grounded in epistemic humility, human impact, and sustainable practice.

Michael D. Ekstrand

Recommender systems research is concerned with many aspects of recommender system behavior and effects than simply its effectiveness, and simulation can be a powerful tool for uncovering these effects. In this brief position paper, I identify specific types of research that simulation is uniquely well-suited to address along with a hierarchy of simulation types.

Ngozi Ihemelandu, Michael D. Ekstrand

This paper calls attention to the missing component of the recommender system evaluation process: Statistical Inference. There is active research in several components of the recommender system evaluation process: selecting baselines, standardizing benchmarks, and target item sampling. However, there has not yet been significant work on the role and use of statistical inference for analyzing recommender system evaluation results. In this paper, we argue that the use of statistical inference is a key component of the evaluation process that has not been given sufficient attention. We support this argument with systematic review of recent RecSys papers to understand how statistical inference is currently being used, along with a brief survey of studies that have been done on the use of statistical inference in the information retrieval community. We present several challenges that exist for inference in recommendation experiment which buttresses the need for empirical studies to aid with appropriately selecting and applying statistical inference techniques.

Asia J. Biega, Fernando Diaz, Michael D. Ekstrand et al.

This paper provides an overview of the NIST TREC 2020 Fair Ranking track. For 2020, we again adopted an academic search task, where we have a corpus of academic article abstracts and queries submitted to a production academic search engine. The central goal of the Fair Ranking track is to provide fair exposure to different groups of authors (a group fairness framing). We recognize that there may be multiple group definitions (e.g. based on demographics, stature, topic) and hoped for the systems to be robust to these. We expected participants to develop systems that optimize for fairness and relevance for arbitrary group definitions, and did not reveal the exact group definitions until after the evaluation runs were submitted.The track contains two tasks,reranking and retrieval, with a shared evaluation.

Amifa Raj, Ashlee Milton, Michael D. Ekstrand

In this position paper, we argue for the need to investigate if and how gender stereotypes manifest in search and recommender systems.As a starting point, we particularly focus on how these systems may propagate and reinforce gender stereotypes through their results in learning environments, a context where teachers and children in their formative stage regularly interact with these systems. We provide motivating examples supporting our concerns and outline an agenda to support future research addressing the phenomena.

Michael D. Ekstrand, Anubrata Das, Robin Burke et al.

Recommendation, information retrieval, and other information access systems pose unique challenges for investigating and applying the fairness and non-discrimination concepts that have been developed for studying other machine learning systems. While fair information access shares many commonalities with fair classification, the multistakeholder nature of information access applications, the rank-based problem setting, the centrality of personalization in many cases, and the role of user response complicate the problem of identifying precisely what types and operationalizations of fairness may be relevant, let alone measuring or promoting them. In this monograph, we present a taxonomy of the various dimensions of fair information access and survey the literature to date on this new and rapidly-growing topic. We preface this with brief introductions to information access and algorithmic fairness, to facilitate use of this work by scholars with experience in one (or neither) of these fields who wish to learn about their intersection. We conclude with several open problems in fair information access, along with some suggestions for how to approach research in this space.

Amifa Raj, Michael D. Ekstrand

Ranked lists are frequently used by information retrieval (IR) systems to present results believed to be relevant to the users information need. Fairness is a relatively new but important aspect of these rankings to measure, joining a rich set of metrics that go beyond traditional accuracy or utility constructs to provide a more holistic understanding of IR system behavior. In the last few years, several metrics have been proposed to quantify the (un)fairness of rankings, particularly with respect to particular group(s) of content providers, but comparative analyses of these metrics -- particularly for IR -- is lacking. There is limited guidance, therefore, to decide what fairness metrics are applicable to a specific scenario, or assessment of the extent to which metrics agree or disagree applied to real data. In this paper, we describe several fair ranking metrics from existing literature in a common notation, enabling direct comparison of their assumptions, goals, and design choices; we then empirically compare them on multiple data sets covering both search and recommendation tasks.

Fernando Diaz, Bhaskar Mitra, Michael D. Ekstrand et al.

We introduce the concept of \emph{expected exposure} as the average attention ranked items receive from users over repeated samples of the same query. Furthermore, we advocate for the adoption of the principle of equal expected exposure: given a fixed information need, no item should receive more or less expected exposure than any other item of the same relevance grade. We argue that this principle is desirable for many retrieval objectives and scenarios, including topical diversity and fair ranking. Leveraging user models from existing retrieval metrics, we propose a general evaluation methodology based on expected exposure and draw connections to related metrics in information retrieval evaluation. Importantly, this methodology relaxes classic information retrieval assumptions, allowing a system, in response to a query, to produce a \emph{distribution over rankings} instead of a single fixed ranking. We study the behavior of the expected exposure metric and stochastic rankers across a variety of information access conditions, including \emph{ad hoc} retrieval and recommendation. We believe that measuring and optimizing expected exposure metrics using randomization opens a new area for retrieval algorithm development and progress.

Asia J. Biega, Fernando Diaz, Michael D. Ekstrand et al.

The goal of the TREC Fair Ranking track was to develop a benchmark for evaluating retrieval systems in terms of fairness to different content providers in addition to classic notions of relevance. As part of the benchmark, we defined standardized fairness metrics with evaluation protocols and released a dataset for the fair ranking problem. The 2019 task focused on reranking academic paper abstracts given a query. The objective was to fairly represent relevant authors from several groups that were unknown at the system submission time. Thus, the track emphasized the development of systems which have robust performance across a variety of group definitions. Participants were provided with querylog data (queries, documents, and relevance) from Semantic Scholar. This paper presents an overview of the track, including the task definition, descriptions of the data and the annotation process, as well as a comparison of the performance of submitted systems.

Mucun Tian, Michael D. Ekstrand

Offline evaluations of recommender systems attempt to estimate users' satisfaction with recommendations using static data from prior user interactions. These evaluations provide researchers and developers with first approximations of the likely performance of a new system and help weed out bad ideas before presenting them to users. However, offline evaluation cannot accurately assess novel, relevant recommendations, because the most novel items were previously unknown to the user, so they are missing from the historical data and cannot be judged as relevant. We present a simulation study to estimate the error that such missing data causes in commonly-used evaluation metrics in order to assess its prevalence and impact. We find that missing data in the rating or observation process causes the evaluation protocol to systematically mis-estimate metric values, and in some cases erroneously determine that a popularity-based recommender outperforms even a perfect personalized recommender. Substantial breakthroughs in recommendation quality, therefore, will be difficult to assess with existing offline techniques.

Alexandra Olteanu, Jean Garcia-Gathright, Maarten de Rijke et al.

The proceedings list for the program of FACTS-IR 2019, the Workshop on Fairness, Accountability, Confidentiality, Transparency, and Safety in Information Retrieval held at SIGIR 2019.

Michael D. Ekstrand, Joseph A. Konstan

As the field of recommender systems has developed, authors have used a myriad of notations for describing the mathematical workings of recommendation algorithms. These notations ap-pear in research papers, books, lecture notes, blog posts, and software documentation. The dis-ciplinary diversity of the field has not contributed to consistency in notation; scholars whose home base is in information retrieval have different habits and expectations than those in ma-chine learning or human-computer interaction. In the course of years of teaching and research on recommender systems, we have seen the val-ue in adopting a consistent notation across our work. This has been particularly highlighted in our development of the Recommender Systems MOOC on Coursera (Konstan et al. 2015), as we need to explain a wide variety of algorithms and our learners are not well-served by changing notation between algorithms. In this paper, we describe the notation we have adopted in our work, along with its justification and some discussion of considered alternatives. We present this in hope that it will be useful to others writing and teaching about recommender systems. This notation has served us well for some time now, in research, online education, and traditional classroom instruction. We feel it is ready for broad use.

Michael D. Ekstrand, Daniel Kluver

Collaborative filtering algorithms find useful patterns in rating and consumption data and exploit these patterns to guide users to good items. Many of the patterns in rating datasets reflect important real-world differences between the various users and items in the data; other patterns may be irrelevant or possibly undesirable for social or ethical reasons, particularly if they reflect undesired discrimination, such as discrimination in publishing or purchasing against authors who are women or ethnic minorities. In this work, we examine the response of collaborative filtering recommender algorithms to the distribution of their input data with respect to a dimension of social concern, namely content creator gender. Using publicly-available book ratings data, we measure the distribution of the genders of the authors of books in user rating profiles and recommendation lists produced from this data. We find that common collaborative filtering algorithms differ in the gender distribution of their recommendation lists, and in the relationship of that output distribution to user profile distribution.