A Locality Sensitive Hashing Approach for Conceptual Classification

Abstract

The increasing volume of semantic content available in the Web, generally classified by concept hierarchies or simple ontologies, turns the searching and reasoning upon these data a great challenge. Generally, a search in Semantic Web may not be addressed to a specific document, but to a group of data classified in the same concept. Several structures used to distribute data, e.g. P2P networks, use hash values to identify these data, without maintaining the semantic values of the stored data. This paper contributes by proposing the creation of hash values that keep similar data stored near to each other in a P2P network, reducing the effort to retrieve similar data. The proposed hash values are derived from the data classification based on ontologies, using locality sensitive hashing (LSH) functions.

Improved locality-sensitive hashing method for the approximate nearest neighbor problem

“…The locality-sensitive hashing algorithm, as the effective method to deal with the nearest neighbor search (NNS) problem, is widely used in many applications, like image retrieval, [10][11][12][13] classification, [14,15] copy detection, [16,17] association rule mining, [18] and three-dimensional (3D) object indexing. [19] The most significant feature of LSH is that the construction of hash functions in LSH is dependent on diverse similarity metrics.…”

“…Choosing more appropriate hash functions to better partition the vector space plays an important role in the LSH's performance. [38] Cao et al [14] brought forward two improved LSH algorithms based on a weakly supervised learning technique. The first one selected the most appropriate hash functions based on sample pairs labeled with "similar" or "dissimilar".…”

Locality-Sensitive Hashing for Massive String-Based Ontology Matching

“…In this paper localitysensitive hashing (LSH) will be used to prune the number of pairs which have to be compared. LSH has been used earlier in the context of ontologies, for instance by de Paula et al [1] who used the LSH technique to group similar concepts together, mainly aiming at a faster retrieval speed. A recent article by Duan et al [2] also uses LSH for the matching of ontologies.…”