SAR: Smart Adaptive Recommendations

'Smart Adaptive Recommendations' (SAR) is the name of a fast, scalable, adaptive algorithm for personalized recommendations based on user transactions and item descriptions. It produces easily explainable/interpretable recommendations and handles "cold item" and "semi-cold user" scenarios. This package provides two implementations of 'SAR': a standalone implementation, and an interface to a web service in Microsoft's 'Azure' cloud: <>. The former allows fast and easy experimentation, and the latter provides robust scalability and extra features for production use.

Version: 1.0.3
Depends: R (≥ 3.3)
Imports: AzureRMR, AzureStor, dplyr (≥ 0.7), httr, jsonlite, Matrix, R6, parallel, Rcpp (≥ 0.12), RcppParallel
LinkingTo: Rcpp, RcppArmadillo, RcppParallel
Suggests: testthat
Published: 2020-10-07
DOI: 10.32614/CRAN.package.SAR
Author: Hong Ooi [aut, cre], Microsoft Product Recommendations team [ctb] (source for MS sample datasets), Microsoft [cph]
Maintainer: Hong Ooi <hongooi73 at>
License: MIT + file LICENSE
NeedsCompilation: yes
SystemRequirements: GNU make
Materials: README NEWS
CRAN checks: SAR results


Reference manual: SAR.pdf


Package source: SAR_1.0.3.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): SAR_1.0.3.tgz, r-oldrel (arm64): SAR_1.0.3.tgz, r-release (x86_64): SAR_1.0.3.tgz, r-oldrel (x86_64): SAR_1.0.3.tgz
Old sources: SAR archive


Please use the canonical form to link to this page.