deepBF: Malicious URL detection using learned Bloom Filter and evolutionary deep learning

Malicious URL detection is an emerging research area due to the continuous modernization of various systems, for instance, Edge Computing. This article presents a novel malicious technique called deepBF (deep learning and Bloom Filter). presented in two-fold. Firstly, we propose learned Filter using 2-dimensional Filter. We experimentally decide best non-cryptography string hash function. Then, derive modified function from selected by introducing biases hashing method compared among functions. The other variants diverse It also with filters, particularly counting Filter, Kirsch et al., Cuckoo use cases. cases unearth weakness strengths filters. Secondly, mechanism Deep Learning. apply evolutionary convolutional neural network identify URLs. trained tested datasets. output accuracy. Moreover, Filters continuously update upon encountering new URL. Otherwise, can answer queries avoid unnecessary load on Furthermore, have achieved many conclusions our experimental evaluation results are able reach conclusive decisions, which article.