Benchmarking Nearest Neighbor Search: Influence of Local Intrinsic Dimensionality and Result Diversity in Real-World Datasets

This paper reconsiders common benchmarking approaches to nearest neighbor search. It is shown that the concept of local intrinsic dimensionality (LID) allows to choose query sets of a wide range of diculty for real-world datasets. Moreover, the eect of dierent LID distributions on the running time performance of implementations is empirically studied. To this end, dierent visualization concepts are introduced that allow to get a more ne-grained overview of the inner workings of nearest neighbor earch principles. The paper closes with remarks
about the diversity of datasets commonly used for nearest neighbor search benchmarking. It is shown that such real-world datasets are not diverse: results on a single dataset predict results on all other datasets well.