Research Article
New Hybrid Features Selection Method: A Case Study on Websites Phishing
Table 3
The decrement in % between two successive scores based on the results of Table
2 for IG, CHI, and the proposed method.
| Features rank | IG drop % | CHI drop % | New method drop % | Average drop % |
| 1 & 2 | 0.044088176 | 0.107687706 | 0.07534097 | 0.075705617 | 2 & 3 | 0.742138365 | 0.713040563 | 0.737068273 | 0.730749067 | 3 & 4 | 0.069105691 | 0.068341121 | 0.059243017 | 0.065563276 | 4 & 5 | 0.048034934 | 0.15799373 | 0.016956892 | 0.074328519 | 5 & 6 | 0.568807339 | 0.469843634 | 0.552891323 | 0.530514099 | 6 & 7 | 0.021276596 | 0.004213483 | 0.011666616 | 0.012385565 | 7 & 8 | 0.195652174 | 0.205923836 | 0.210839348 | 0.204138453 | 8 & 9 | 0.027027027 | 0.037300178 | 0.009229999 | 0.024519068 | 9 & 10 | 0.694444444 | 0.662361624 | 0.680106538 | 0.678970869 | ⋯ | ⋯ | ⋯ | ⋯ | ⋯ |
|
|