Design of Kids-specific URL Classifier using Recurrent Convolutional Neural Network

• DOI: 10.1016/j.procs.2020.03.260
• Publication Date: 2020-04-16T15:44:04Z
• Authors: R. Rajalakshmi; Hans Tiwari; Jay Patel; Ankit Kumar; R Karthik.
• Keywords: Benchmark (surveying)

The use of digital devices has increased exponentially in the last decade. Especially, young children spend most their time surfing for various reasons such as homework. assignments and projects etc. Parental Control is highly important monitoring browsing behavior children. Though several content-based web page classification approaches are available, it requires entire contents purpose. This leads to wastage bandwidth due unnecessary downloads. exponential growth internet demands URL based classifiers adapt dynamic web, make swift decisions on fly. To address this problem, a deep learning approach been proposed research that can extract features only from page. learn patterns Kids-specific sites automatically, Convolutional Neural Network (CNN) combined with Bidirectional Gated Recurrent Unit (BGRU) rich context aware well preserve sequence information URL. By conducting experiments benchmark collection Open Directory Project (ODP), shown an accuracy 82.04% be achieved.