Scalable extraction and visualization of multi-attribute logical and functional dependencies in tabular data
The research introduces LDTool and HLDTool to extract and visualize multi-attribute logical dependencies (LDs) and functional dependencies (FDs) from tabular data. LDTool extends dependency discovery beyond simple pairwise relationships, while HLDTool uses hypergraph-guided search to handle high-dimensional data efficiently. This framework provides a unified method for finding complex, non-deterministic relationships.