textnets.network.bipartite_rank

textnets.network.bipartite_rank(net: Textnet, normalizer: Literal['HITS', 'CoHITS', 'BGRM', 'BiRank'], alpha: float = 0.85, beta: float = 0.85, max_iter: int = -1, tolerance: float = 0.0001) → Series[source]

Calculate centralities of nodes in the bipartite network.

Parameters:

  • normalizer (string) – The normalizer to use: HITS, CoHITS, BGRM, or BiRank. See reference for details.

  • alpha (float, optional) – Damping factor for the rows and columns.

  • beta (float, optional) – Damping factor for the rows and columns.

  • max_iter (int, optional) – Maximum number of iterations to run before reaching convergence (default: -1, meaning iterate until the errors are within the specified tolerance).

  • tolerance (float, optional) – Error tolerance when checking for convergence.

Raises:

ValueError – If an invalid normalizer is specified.

Returns:

The BiRank for both sets of nodes indexed by node label.

Return type:

pandas.Series

Notes

Adapted from the implementation by Yang et al. [2020].

References

[He et al., 2017]