Skip to main content

Table 5 The evolution of the error rate of the SVM 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
SVM (%) 40 41.26 34.40 36.70 28 30.35 28.26 23.97 26
SVM—JW (%) 24 19.04 17.60 16.48 12 12.77 8 12.32 12.20
SVM—H (%) 8 11.11 5.60 2.12 4 2.55 5.60 5.02 4.20
SVM—LCS (%) 12 9.52 4.80 2.65 3.20 3.19 2.66 3.42 5.20
SVM—L (%) 0 4.76 3.20 1.06 2.80 2.23 2.93 5.93 5.60
SVM—NL (%) 20 39.68 16 11.70 16.8 15.01 15.73 14.84 13.4
SVM—QG (%) 0 7.93 2.40 2.65 4.80 5.75 3.20 4.10 5
  1. 4.20% is the lowest error rate obtained in these tests while using the complete dataset