The crucial and criminal activities in Internet occurs due to malicious or suspicious websites. Therefore, it turns to be an challenge approach to keep away the end users from moving on to the malicious websites. In this paper we propose a technique called optimal RT-PFL to classify the malicious URLs from non-malicious URLs detected in the websites. Here the data set should be encoded into both the lexical as well as the host-based features related to the URL in order to generate the feature components. Certain features is extracted by the feature extraction process. Optimal features of URLs is selected based upon the proposed feature selection approach namely Gray Wolf Optimizer based Rough Set Theory algorithm. This proposed algorithm can productively identify minimal attribute reduction from the highly efficient dataset which in turn improves the classification systems performance. The chosen URL should be admitted towards the classifier to foresee whether the admitted URL is benign or it is malicious. The classification of URLs depends on the newly proposed particle filtering based fuzzy logic approach. The subsequent classifiers gains higher accuracy by identifying huge amount of malicious URLs from the malevolent sites.
CITATION STYLE
Rajitha, K., & Vijayalakshmi, D. (2018). Suspicious URLs filtering using optimal RT-PFL: A novel feature selection based web URL detection. In Smart Innovation, Systems and Technologies (Vol. 78, pp. 227–235). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-10-5547-8_24
Mendeley helps you to discover research relevant for your work.