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

Abstract

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.

8 pages; accepted for publication in CIKM 2020

References in corpus (1)

• Estimating Error and Bias in Offline Evaluation Results

Cited by in corpus (12)

• Evaluating Stochastic Rankings with Expected Exposure
• From Clicks to Carbon: The Environmental Toll of Recommender Systems
• Distributionally-Informed Recommender System Evaluation
• It's Not You, It's Me: The Impact of Choice Models and Ranking Strategies on Gender Imbalance in Music Recommendation
• New Insights into Metric Optimization for Ranking-based Recommendation
• Candidate Set Sampling for Evaluating Top-N Recommendation
• Mitigating Mainstream Bias in Recommendation via Cost-sensitive Learning
• Recommender Systems Algorithm Selection for Ranking Prediction on Implicit Feedback Datasets
• Inference at Scale Significance Testing for Large Search and Recommendation Experiments
• Informfully Recommenders -- Reproducibility Framework for Diversity-aware Intra-session Recommendations
• User and Recommender Behavior Over Time: Contextualizing Activity, Effectiveness, Diversity, and Fairness in Book Recommendation
• Extending MovieLens-32M to Provide New Evaluation Objectives