Skip to main content

Table 10 Accuracy evaluation results—scalability test

From: Efficient spatial data partitioning for distributed \(k\)NN joins

  Executors Exact match Mis-match Missing Exact match Mis-match Missing
With non-spatial data* Without non-spatial data*
SpPart_kNN  VS   LocationSpark
BUS Records (119, 319) 10 \(80.72\%\) \(19.28\%\) \(0.01\%\) \(80.54\%\) \(19.46\%\) \(0.01\%\)
20 \(81.30\%\) \(18.70\%\) \(0.01\%\) \(80.43\%\) \(19.57\%\) \(0.01\%\)
30 \(81.26\%\) \(18.74\%\) \(0.01\%\) \(80.67\%\) \(19.33\%\) \(0.01\%\)
40 \(81.19\%\) \(18.81\%\) \(0.01\%\) \(80.40\%\) \(19.60\%\) \(0.01\%\)
50 \(81.50\%\) \(18.50\%\) \(0.01\%\) \(80.61\%\) \(19.39\%\) \(0.01\%\)
TAXI Records (119, 319) 10 \(81.98\%\) \(18.02\%\) \(0.00\%\) \(81.92\%\) \(18.08\%\) \(0.00\%\)
20 \(81.96\%\) \(18.04\%\) \(0.00\%\) \(81.93\%\) \(18.07\%\) \(0.00\%\)
30 \(82.00\%\) \(18.00\%\) \(0.00\%\) \(82.12\%\) \(17.88\%\) \(0.00\%\)
40 \(82.01\%\) \(17.99\%\) \(0.00\%\) \(82.01\%\) \(17.99\%\) \(0.00\%\)
50 \(81.97\%\) \(18.03\%\) \(0.00\%\) \(82.10\%\) \(17.90\%\) \(0.00\%\)
TLC Records(119, 319) 10 \(85.67\%\) \(14.33\%\) \(0.00\%\) \(85.62\%\) \(14.38\%\) \(0.00\%\)
20 \(85.67\%\) \(14.33\%\) \(0.00\%\) \(85.62\%\) \(14.38\%\) \(0.00\%\)
30 \(85.68\%\) \(14.32\%\) \(0.00\%\) \(85.63\%\) \(14.37\%\) \(0.00\%\)
40 \(85.69\%\) \(14.31\%\) \(0.00\%\) \(85.62\%\) \(14.38\%\) \(0.00\%\)
50 \(85.69\%\) \(14.31\%\) \(0.00\%\) \(85.61\%\) \(14.39\%\) \(0.00\%\)
  1. *Simba, STARK, GeoSpark results omitted for lack of support or exceeding 180 minutes of runtime