Implemented a second way to instantiate the Bloom filter by specifying the desired false positive probability and expected number of input elements.
The optimal size and number of hash functions is then calculated automatically according to https://en.wikipedia.org/wiki/Bloom_filter.