Latent table discovery by semantic relationship extraction between unrelated sets of entity sets of structured data sources

TL;DR

This paper proposes a method to discover hidden semantic relationships between unrelated data sets to improve database querying efficiency and flexibility across heterogeneous and distributed systems.

Contribution

It introduces a novel approach for extracting and storing tacit semantic relationships between unrelated entity sets in structured data sources.

Findings

Significant improvement in query performance demonstrated

Semantic relationship extraction enables more flexible querying

Applicable to heterogeneous and distributed data sources

Abstract

Querying is one of the basic functionality expected from a database system. Query efficiency is adversely affected by increase in the number of participating tables. Also, querying based on syntax largely limits the gamut of queries a database system can process. Syntactic queries rely on the database table structure, which is a cause of concern for large organisations due to incompatibility between heterogeneous systems that store data distributed across geographic locations. Solution to these problems is answered to some extent by moving towards semantic technology by making data and the database meaningful. In doing so, relationship between sets of entity sets will not be limited only to syntactic constraints but would also permit semantic connections nonetheless such relationships may be tacit, intangible and invisible. The goal of this work is to extract such hidden relationships between unrelated sets of entity sets and store them in a tangible form. A few sample cases are provided to vindicate that the proposed work improves querying significantly.