Locality-sensitive hashing scheme based on dynamic collision counting
Junhao Gan, Jianlin Feng, Qiong Fang, Wilfred Ng
Abstract
Junhao Gan, Jianlin Feng, Qiong Fang, Wilfred Ng
Abstract
Locality-Sensitive Hashing (LSH) and its variants are well-known methods for solving the c-approximate NN Search problem in high-dimensional space. Traditionally, several LSH functions are concatenated to form a "static" compound hash function for building a hash table. In this paper, we propose to use a base of m single LSH functions to construct "dynamic" compound hash functions, and define a new LSH scheme called Collision Counting LSH (C2LSH). If the number of LSH functions under which a data object o collides with a query object q is greater than a pre-specified collision threhold l, then o can be regarded as a good candidate of c-approximate NN of q. This is the basic idea of C2LSH.
OpenAlex reports 239 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Locality-Sensitive Hashing (LSH) and its variants are well-known methods for solving the c-approximate NN Search problem in high-dimensional space. Traditionally, several LSH functions are concatenated to form a "static" compound hash function for building a hash table. In this paper, we propose to use a base of m single LSH functions to construct "dynamic" compound hash functions, and define a new LSH scheme called Collision Counting LSH (C2LSH). If the number of LSH functions under which a data object o collides with a query object q is greater than a pre-specified collision threhold l, then o can be regarded as a good candidate of c-approximate NN of q. This is the basic idea of C2LSH.
Key concepts: Hash function, Locality-sensitive hashing, Hash table, Computer science, Collision, Double hashing, Object (grammar), Dynamic perfect hashing