<ul><li>A novel clustering method called CMDI is proposed, which incorporates two-dimensional structural information into the graph-based clustering process.</li><li>CMDI reformulates graph partitioning as an abstract clustering problem, leveraging maximum decoding information to minimize uncertainty associated with random visits to vertices.</li><li>Empirical evaluations on three real-world datasets show that CMDI outperforms classical baseline methods, exhibiting a superior decoding information ratio (DI-R).</li><li>CMDI demonstrates heightened efficiency, especially when considering prior knowledge (PK), making it a valuable tool in graph-based clustering analyses.</li></ul>

A Clustering Method with Graph Maximum Decoding Information

Discover more