Our previous work demonstrated the possibility of distinguishing several kinds of applications with accuracy of over 99%. Today, most of the traffic is generated by web browsers, which provide different kinds of services based on the HTTP protocol: web browsing, file downloads, audio and voice streaming through third-party plugins, etc. This paper suggests and evaluates two approaches to distinguish various HTTP content: distributed among volunteers' machines and centralized running in the core of the network. We also assess accuracy of the global classifier for both HTTP and non-HTTP traffic. We achieved accuracy of 94%, which supposed to be even higher in real-life usage. Finally, we provided graphical characteristics of different kinds of HTTP traffic.