Jo et al., "Extracting hierarchical backbones from bipartite networks" (2023)

Woo Seong Jo, Jaehyuk Park, Arthur Luhur, Beom Jun Kim, and Yong-Yeol Ahn, In preparation (2023)
arXiv | Code

@article{jo2022hierarchy,
    author = {Woo Seong Jo and Jaehyuk Park and Arthur Luhur and Beom Jun Kim and Yong-Yeol Ahn},
    title = {Extracting hierarchical backbones from bipartite networks},
    year = {2023},
    eprint = {2002.07239},
    archivePrefix = {arXiv},
    primaryClass = {cs.SI},
}

We propose a method for extracting hierarchical backbones from bipartite networks by leveraging asymmetries in conditional probabilities of connections. The method estimates both the importance and direction of hierarchical relationships between node pairs. Using semi-synthetic benchmarks, it outperforms existing methods at identifying planted hierarchy. Applications to empirical datasets—skills and individuals, as well as gene products and Gene Ontology terms—demonstrate the possibility of automatically extracting or augmenting ontology from data.