Bloom Filters are data structures used to efficiently answer queries when we do not have enough "search key" space to handle all possible queries. In this case, the "search key" is hashed, marked and then used later to check if it was searched earlier or not.
Bloom Filters use hashing as this is an immutable function result, and marking the respective positions in the data structure guarantees that the next search for the same string will return true.
This data structure has an error rate when returning 'true', and we look into how the number of hash functions effect it's performance. In practice, Bloom Filters can be used to check for membership and to avoid 'One Hit Wonders'.
We talk about what are bloom filters, how do we use them and where can these filters be applied.
An example would be tinyUrl, which can check if a url has been previously generated using a bloom filter, and regenerate it if the answer is positive.
Bloom Filter: https://en.wikipedia.org/wiki/Bloom_filter
FaceBook Talk: https://www.facebook.com/Engineering/videos/432864835468/
One Hit Wonder: https://en.wikipedia.org/wiki/One-hit_wonder
My Social Media Links: