pyjedai.prioritization

pyjedai.prioritization#

Entity Matching Prioritization Module

Classes

BlockIndependentPM([similarity_function, ...])

Applies the matching process to a subset of available pairs progressively

EmbeddingsNNBPM([language_model, ...])

Utilizes/Creates entity embeddings, constructs neighborhoods via NN Approach and applies Progressive Matching

GlobalPSNM([weighting_scheme, window_size, ...])

Applies Global Progressive Sorted Neighborhood Matching

GlobalTopPM([weighting_scheme, ...])

Applies Progressive CEP, sorts retained comparisons and applies Progressive Matching

HashBasedProgressiveMatching([...])

Applies hash based candidate graph prunning, sorts retained comparisons and applies Progressive Matching

LocalPSNM([weighting_scheme, window_size, ...])

Applies Local Progressive Sorted Neighborhood Matching

LocalTopPM([weighting_scheme, ...])

Applies Progressive CNP, sorts retained comparisons and applies Progressive Matching

PESM([weighting_scheme, ...])

Applies Progressive Entity Scheduling Matching

ProgressiveMatching([similarity_function, ...])

Applies the matching process to a subset of available pairs progressively

RandomPM([similarity_function, tokenizer, ...])

Picks a number of random comparisons equal to the available budget

SimilarityBasedProgressiveMatching([...])

Applies similarity based candidate graph prunning, sorts retained comparisons and applies Progressive Matching

TopKJoinPM([similarity_function, ...])

Applies index based matching for ES, emits candidate pairs using defined budget/emission technique