I am learning about DB systems. I am currently learning about Data Mining and the problem of finding similar items.

Suppose that the universal set is {1, 2,...,10}, and signatures for sets are constructed using the following list of permutations: 1. (1,2,3,4,5,6;7,8,9,10) 2. (10,8,6,4,2,9,7,5,3,1) 3. (4;7,2,9,1,5,3,10,6,8)

Construct minhash signatures for the following sets: a) {3,6,9}. b) {2,4,6,8} c) {2,3,4} How does the estimate of the Jaccard similarity for each pair, derived from the signatures, compare with the true Jaccard similarity?