Replication Data for: Revisiting Oblivious Top-k Selection with Applications to Secure k-NN Classification

The code implements a homomorphic Top-k selection algorithm and an application to secure k-nearest neighbors classification. More specifically, our method applies a new truncation technique to Batcher's odd-even sorting algorithm. All of this is done in an oblivious manner, because the underlying data is encrypted. Our experimental results show a speedup of up to 47 times (not accounting for difference in CPU) compared to a previous secure k-nearest neighbors classifier.