Discovery of Keys for Graphs [Extended Version]

TL;DR

This paper introduces a new algorithm for mining keys in graph databases by expanding frequent subgraphs, emphasizing properties like minimality and support, and demonstrates its efficiency on real-world data.

Contribution

It presents a novel algorithm for discovering graph keys through frequent subgraph expansion, incorporating properties like minimality and support, and validates its performance on real-world graphs.

Findings

Efficient key discovery demonstrated on real-world graphs

Algorithm leverages properties of minimality and support

Experimental results confirm practical applicability

Abstract

Keys for graphs uses the topology and value constraints needed to uniquely identify entities in a graph database. They have been studied to support object identification, knowledge fusion, data deduplication, and social network reconciliation. In this paper, we present our algorithm to mine keys over graphs. Our algorithm discovers keys in a graph via frequent subgraph expansion. We present two properties that define a meaningful key, including minimality and support. Lastly, using real-world graphs, we experimentally verify the efficiency of our algorithm on real world graphs.