Abstract
Private searching on streaming data allows a user to collect potentially useful information from huge streaming sources of data without revealing his or her searching criteria. This technique can be used for airports, without knowing a classified “possible terrorists” list, to find if any of hundreds of passenger lists has a name from the “possible terrorists” list and if so his/hers itinerary. Current solutions for private searching on streaming data only support searching for “OR” of keywords or “AND” of two sets of keywords. In this paper, we extend the types of private queries to support searching on streaming data for an “OR” of a set of both single and conjunctive keywords. Our protocol is built on Boneh et al.’s result for the evaluation of 2-DNF formulas on ciphertexts. The size of our encrypted dictionary is O(|D|) only, which is much less than 〖|D|〗^2, the size of the encrypted dictionary if conjunctive keywords (A_i,B_i) (i=1,2,…,k) is treated as single keyword, where we assume A_i,B_i∈D (i=1,2,…,k).
