3

I have n number of roaring bitmaps, and I want to find out which element is most occurring in these n roaring bitmaps. Also, find out the count. What is the best and most efficient way to do this? I have millions of records in one bitmap, and traversing and storing the results in a hashmap is not feasible. Additionally, I am looking for elements that occur in more than n/2 bitmaps

I have tried Boyer-Moore Majority algorithms, but it is inefficient because I don't want to traverse all the elements.

Vishal Jangid
  • 358
  • 1
  • 10
  • Yes, in single roaring bitmap, i might have millions of elements. – Vishal Jangid Jan 17 '23 at 08:29
  • 1
    https://roaringbitmap.org/ – Vishal Jangid Jan 17 '23 at 08:57
  • It's unclear what you're asking for. You say you want to find the elements that occur in more than n/2 bitmaps, but you don't want to traverse all the elements. How in the world do you expect to determine the most common elements if you don't look at all of them? – Jim Mischel Jan 17 '23 at 18:43
  • We can ignore high cardinality sets and find the top k occurring element in the remaining sets. Once we get the top k elements, we can check those top k elements present in ignored sets. #justthinking – Vishal Jangid Jan 20 '23 at 08:53

0 Answers0