Skip to main content

Table 5 Four-category analysis on real data sets (N = 1,083,878)

From: Efficient algorithms for fast integration on large data sets from multiple sources

   Type I Type II Type III Type IV
ED name constant t = 1 93.0% 2.2% 0.0% 4.8%
ED all t = 1 97.7% 2.1% 0.0% 0.2%
PD name - 91.6% 1.7% 0.0% 6.7%
PDED t = 1 98.4% 1.3% 0.0% 0.3%
ED name proportional t = 0.1 93.1% 2.2% 0.0% 4.7%
ED all t = 0.1 98.1% 0.1% 0.0% 0.4%
PD name - 91.6% 1.7% 0.0% 6.7%
PDED t = 0.1 98.1% 1.3% 0.0% 0.6%