src.preprocessing.cluster_explainer

Generate a summary of the clusters.

Classes

ClusterExplainer([threshold])

Generate a summary and extracts the entities from the cluster.
class src.preprocessing.cluster_explainer.ClusterExplainer(threshold=10)[source]

Generate a summary and extracts the entities from the cluster.

Parameters:

threshold (int) – the number of important sentences in the cluster.

threshold: int = 10
rank_sentences(sentences)[source]

Rank the sentences based on the LexRank model.

Parameters:

sentences (Iterable[str]) – the filtered sentences in a cluster.

Return type:

list[str]

Returns:

the most important sentences in the cluster.

custom_transform(data, **transform_args)[source]

Generate a summary and extract the entities from the cluster.

Parameters:

  • data (DataFrame) – the input data.

  • transform_args (Never) – [UNUSED] the transformation arguments.

Return type:

DataFrame

Returns:

the transformed data.