Web page multi-label classification for filtering content from the web

Abstract

In this paper, we describe a simple approach to filter unwanted web pages, according to their content. The result of this work is a demo of an application that is usable in realtime filtering and in non-real-time indexing of any given web pages. We describe a proposed technique step by step, while discussing possible alternative ways for each part. In the end we discuss the overall quality and proposed next steps that could lead to a fully usable business application.