FINALLY: A Dataset Recommender for RecSys Experiments

Abstract

FINALLY is a web-based recommender system for dataset selection in RecSys experiments. It supports a decision often made manually or by convention: choosing datasets to evaluate a recommendation algorithm offline. With FINALLY, users choose seed datasets, filters, a target set size, and a recommendation strategy. FINALLY returns dataset recommendations that can be inspected through metadata and, where available, an Algorithm Performance Space (APS), and exported for experiments and publications. FINALLY supports APS-based diverse and non-diverse strategies as well as random selection, with diversity as the default. FINALLY is publicly accessible, operates on more than 90 recommender-system datasets, and builds on the APS Explorer code base with a recommendation-centered workflow. To the best of our knowledge, FINALLY is the first operational system that recommends configurable sets of datasets specifically for recommender-systems experiments.