Michael D. Ekstrand

Sturgeon and the Cool Kids: Problems with Random Decoys for Top-N Recommender Evaluation

2017 · Scholar Works (Boise State University)

Top-N evaluation of recommender systems, typically carried out using metrics from information retrieval or machine learning, has several challenges. Two of these challenges are popularity bias, where the evaluation intrinsically favors algorithms that recommend popular …

Monte Carlo Estimates of Evaluation Metric Error and Bias

2018

Traditional offline evaluations of recommender systems apply metrics from machine learning and information retrieval in settings where their underlying assumptions no longer hold. This results in significant error and bias in measures of top-N recommendation …

Dependency Injection with Static Analysis and Context-Aware Policy.

2016 · The Journal of Object Technology

The dependency injection design pattern improves the configurability, testability, and maintainability of object-oriented applications by decoupling components from both the concrete implementations of their dependencies and the strategy employed to select those implementations.In recent years, …

Exploring Author Gender in Book Rating and Recommendation

2018 · arXiv (Cornell University)

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 …

Building Human Values into Recommender Systems: An Interdisciplinary Synthesis

2023 · ACM Transactions on Recommender Systems

Recommender systems are the algorithms which select, filter, and personalize content across many of the world's largest platforms and apps. As such, their positive and negative effects on individuals and on societies have been extensively …

Evaluating Stochastic Rankings with Expected Exposure

2020

We introduce the concept of 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 …

LensKit for Python: Next-Generation Software for Recommender Systems Experiments

2020 · ScholarWorks (Boise State University)

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 …