Skip to main content

Table 7 The evolution of the error rate of the naïve bayesian network related to the records count in the database

From: A novel lightweight URL phishing detection system using SVM and similarity index

Records count 100 250 500 750 1000 1250 1500 1750 2000
Naive bayes (%) 52 53.96 36.8 36.70 34.40 36.74 33.60 38.12 35.8
Naive bayes—JW (%) 72 26.98 47.20 37.76 39.20 37.69 28.8 35.15 37.6
Naive bayes—H (%) 16 9.52 8 1.59 5.20 38.97 35.46 36.98 36
Naive bayes—LCS (%) 20 11.11 4.80 3.19 0.4 37.06 32.80 36.75 40.6
Naive bayes—L (%) 16 19.04 2.40 2.12 4 10.22 5.86 36.75 35
Naive bayes—NL (%) 44 50.79 40.80 27.12 36.40 37.06 39.2 25.79 32.80
Naive bayes—QG (%) 12 6.34 1.60 4.78 35.2 35.14 41.06 36.07 34.8
  1. 32.80% is the lowest error rate obtained in these tests while using the complete dataset