Comparative Study of Data Mining Classification Techniques for Detection and Prediction of Phishing Websites

Abstract

Data mining is the process of discovering or extracting information from large amount of data that are stored in databases or datasets such as phishing dataset. Phishing is a vital web security problem that involves simulating legitimate websites to mislead online users in order to steal their sensitive information. This paper aims to detect and predict the type of the website to either legitimate or phishing class label. It investigates different data mining classifiers that are applied to the phishing dataset aiming to determine the effective ones in terms of classification performance. The comparison between nine classifiers with help of rapid miner software was conducted. Here, for comparing the result, five different metrics were used including accuracy, precision, recall, sensitivity and F-Measure. In this study, it has been able to identify the classifiers that precisely recognize fake websites especially with respect to the evolutionary nature of the information attacks.

Hyperteroid Disease Analysis With Backpropagation Artificial Neural Network

“…We used three machine learning algorithms in the classification process in this study, namely the K-NN, SVM, and Naive Bayes algorithms. In the K-NN algorithm, the K value (number of nearest neighbors) plays an essential role in the performance of the classification results; since it determines how much data should have similar characteristics [9]. As shown in the optimization of the K-NN algorithm using the certainty factor in determining students' careers, with 12 K values (K = 1 to K = 12), this study shows that the highest performance came from the K values of 3 and 4, while the worst from the K values of 12 [10].…”

A WGFS Based Approach to Extract Factors Influencing the Marketing of Korean Language in GCC

“…Accuracy could not be sufficient measurement to assess the classification results. Al-Shalabi [38] explained the importance of other measures for the classification quality including recall and precision.…”