Abstract: With the rapid growth of data, near-duplicate documents bearing high similarity are abundant. Elimination of near-duplicates can reduce storage cost and improve the quality of search indexes ...