recometrics: Evaluation Metrics for Implicit-Feedback Recommender Systems

Calculates evaluation metrics for implicit-feedback recommender systems that are based on low-rank matrix factorization models, given the fitted model matrices and data, thus allowing to compare models from a variety of libraries. Metrics include P@K (precision-at-k, for top-K recommendations), R@K (recall at k), AP@K (average precision at k), NDCG@K (normalized discounted cumulative gain at k), Hit@K (from which the 'Hit Rate' is calculated), RR@K (reciprocal rank at k, from which the 'MRR' or 'mean reciprocal rank' is calculated), ROC-AUC (area under the receiver-operating characteristic curve), and PR-AUC (area under the precision-recall curve). These are calculated on a per-user basis according to the ranking of items induced by the model, using efficient multi-threaded routines. Also provides functions for creating train-test splits for model fitting and evaluation.

Version: 0.1.6-3
Imports: Rcpp (≥ 1.0.1), Matrix (≥ 1.3-4), MatrixExtra (≥ 0.1.6), float, RhpcBLASctl, methods
LinkingTo: Rcpp, float
Suggests: recommenderlab (≥ 0.2-7), cmfrec (≥ 3.2.0), data.table, knitr, rmarkdown, kableExtra, testthat
Published: 2023-02-19
DOI: 10.32614/CRAN.package.recometrics
Author: David Cortes
Maintainer: David Cortes <david.cortes.rivera at>
License: BSD_2_clause + file LICENSE
NeedsCompilation: yes
CRAN checks: recometrics results


Reference manual: recometrics.pdf
Vignettes: Evaluating_recommender_systems


Package source: recometrics_0.1.6-3.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): recometrics_0.1.6-3.tgz, r-oldrel (arm64): recometrics_0.1.6-3.tgz, r-release (x86_64): recometrics_0.1.6-3.tgz, r-oldrel (x86_64): recometrics_0.1.6-3.tgz
Old sources: recometrics archive


Please use the canonical form to link to this page.