Table 7 Performance with different attribute embedding methods.

From: An efficient learning based approach for automatic record deduplication with benchmark datasets

Dataset

F1-score

Avg

Idf

Astm

all comp

Amazon-google

0.66

0.73

0.81

0.84

Amazon-Walmart dataset

0.84

0.812

0.855

0.855

dblp-scholar

0.97

0.955

0.975

0.98

Back to article page