Improving Hamming-Distance Computation for Adaptive Similarity Search Approach

Abstract

In the modern context, the similarity is determined by content-preserving stimuli, retrieval of relevant ‘nearest neighbor' objects, and the way similar objects are pursued. Current similarity search in hamming-space-based strategies finds all the data objects within a threshold hamming-distance for a user query, though the number of computations for distance and candidate generation are key concerns from the many years. The hamming-space paradigm extends the range of alternatives for an optimized search experience. A novel counting-based similarity search strategy is proposed with an improved hamming-space (e.g., optimized candidate generation and verification function). The strategy adapts towards the lesser set of user query dimensions and subsequently constrains the hamming-space computations with each data object driven by generated statistics. The extensive evaluation asserts that the proposed counting-based approach can be combined with any pigeonhole principle-based similarity search to further improve its performance.