LGM: Mining Frequent Subgraphs from Linear Graphs

TL;DR

The paper introduces LGM, an efficient algorithm for mining frequent subgraphs in linear graphs, which are common in biological and linguistic data, capable of detecting both connected and disconnected patterns.

Contribution

LGM leverages vertex order and reverse search to efficiently mine both connected and disconnected frequent subgraphs in linear graphs, improving over traditional methods.

Findings

LGM efficiently mines frequent subgraphs in protein contact maps.

LGM detects both connected and disconnected patterns.

Experiments demonstrate its utility and efficiency.

Abstract

A linear graph is a graph whose vertices are totally ordered. Biological and linguistic sequences with interactions among symbols are naturally represented as linear graphs. Examples include protein contact maps, RNA secondary structures and predicate-argument structures. Our algorithm, linear graph miner (LGM), leverages the vertex order for efficient enumeration of frequent subgraphs. Based on the reverse search principle, the pattern space is systematically traversed without expensive duplication checking. Disconnected subgraph patterns are particularly important in linear graphs due to their sequential nature. Unlike conventional graph mining algorithms detecting connected patterns only, LGM can detect disconnected patterns as well. The utility and efficiency of LGM are demonstrated in experiments on protein contact maps.