Research Methods and Infrastructure
I have a number of previous and ongoing projects to improve recommender systems research methods and infrastructure to support research. This has most notably resulted in the LensKit software, an open-source toolkit for recommender systems research, and more recently the new POPROX project to build online infrastructure for user-facing recommender systems research. POPROX is a new project to develop a news recommendation platform that will serve as shared infrastructure to support academic research on recommender systems with actual user responses. Itβs just kicking off, and should be ready for experiments in 2024. I am actively recruiting a Ph.D student for this project. LensKit is an open-source toolkit supporting recommender systems research and education. Originally released for Java in 2010, I rewrote it in Python in 2018. It has been used to support dozens of published papers. 2020. LensKit for Python: Next-Generation Software for Recommender Systems Experiments. In Proceedings of the 29th ACM International Conference on Information and Knowledge Management (CIKM β20, Resource track). ACM, pp.Β 2999β3006. Proc. CIKM β20 (Resource track). DOI 10.1145/3340531.3412778. arXiv:1809.03125. NSF PAR 10199450. No acceptance rate reported. Cited 45 times. Cited 63* times. 2016. Dependency Injection with Static Analysis and Context-Aware Policy. Journal of Object Technology 15(1) (February 2016), 1:1β31. DOI 10.5381/jot.2016.15.1.a1. Cited 14 times. 2014. Towards Recommender Engineering: Tools and Experiments in Recommender Differences. Ph.D thesis, University of Minnesota. 2014. Building Open-Source Tools for Reproducible Research and Education. At Sharing, Re-use, and Circulation of Resources in Cooperative Scientific Work, a workshop at CSCW 2014. 2011. Rethinking The Recommender Research Ecosystem: Reproducibility, Openness, and LensKit. In Proceedings of the Fifth ACM Conference on Recommender Systems (RecSys β11). ACM, pp.Β 133β140. Proc. RecSys β11. DOI 10.1145/2043932.2043958. Acceptance rate: 27% (20% for oral presentation, which this received). Cited 191 times. Cited 226 times. 2011. LensKit: A Modular Recommender Framework. Demo recorded in Proceedings of the 5th ACM Conference on Recommender Systems (RecSys β11). ACM, pp.Β 349-350. Proc. RecSys β11. DOI 10.1145/2043932.2044001. Cited 1 time. Cited 43 times. I have also done a variety of work on evaluating recommender systems, in addition to our work on fairness. 2023. Distributionally-Informed Recommender System Evaluation. Transactions on Recommender Systems (August 2023). TORS (August 2023). DOI 10.1145/3613455. arXiv:2309.05892. NSF PAR 10461937. 2021. Statistical Inference: The Missing Piece of RecSys Experiment Reliability Discourse. In Proceedings of the Perspectives on the Evaluation of Recommender Systems Workshop 2021 (RecSys β21). Proc. PERSPECTIVES @ RecSys β21. DOI 10.48550/arXiv.2109.06424. arXiv:2109.06424. Cited 5 times. Cited 3 times. 2021. Evaluating Recommenders with Distributions. At Proceedings of the RecSys 2021 Workshop on Perspectives on the Evaluation of Recommender Systems (RecSys β21). Proc. PERSPECTIVES @ RecSys β21. 2020. Evaluating Stochastic Rankings with Expected Exposure. In Proceedings of the 29th ACM International Conference on Information and Knowledge Management (CIKM β20). ACM, pp.Β 275β284. Proc. CIKM β20. DOI 10.1145/3340531.3411962. arXiv:2004.13157. NSF PAR 10199451. Acceptance rate: 20%. Nominated for Best Long Paper. Cited 121 times. Cited 115 times. 2020. Estimating Error and Bias in Offline Evaluation Results. Short paper in Proceedings of the 2020 Conference on Human Information Interaction and Retrieval (CHIIR β20). ACM, 5 pp. Proc. CHIIR β20. DOI 10.1145/3343413.3378004. arXiv:2001.09455. NSF PAR 10146883. Acceptance rate: 47%. Cited 8 times. Cited 7 times. 2018. Monte Carlo Estimates of Evaluation Metric Error and Bias. Computer Science Faculty Publications and Presentations 148. Boise State University. Presented at the REVEAL 2018 Workshop on Offline Evaluation for Recommender Systems, a workshop at RecSys 2018. DOI 10.18122/cs_facpubs/148/boisestate. NSF PAR 10074452. Cited 1 time. Cited 1 time. 2018. The Dagstuhl Perspectives Workshop on Performance Modeling and Prediction. SIGIR Forum 52(1) (June 2018), 91β101. DOI 10.1145/3274784.3274789. Cited 18 times. Cited 15 times. 2018. From Evaluating to Forecasting Performance: How to Turn Information Retrieval, Natural Language Processing and Recommender Systems into Predictive Sciences (Dagstuhl Perspectives Workshop 17442). Dagstuhl Manifestos 7(1) (November 2018), 96β139. DOI 10.4230/DagMan.7.1.96. Cited 15 times. Cited 18 times. 2017. Sturgeon and the Cool Kids: Problems with Random Decoys for Top-N Recommender Evaluation. In Proceedings of the 30th International Florida Artificial Intelligence Research Society Conference (Recommender Systems track). AAAI, pp.Β 639β644. (Recommender Systems track). No acceptance rate reported. Cited 9 times. Cited 13 times. 2011. RecBench: Benchmarks for Evaluating Performance of Recommender System Architectures. Proceedings of the VLDB Endowment 4(11) (August 2011), 911β920. Acceptance rate: 18%. Cited 9 times. Cited 21 times.Funding
POPROX
LensKit
Papers
Evaluation Practice
Papers