Personalized Prefetching of Web Contents

OData support
Supervisor:
Dr. Maliosz Markosz
Department of Telecommunications and Media Informatics

Over the past few years, the rapid development of mobile phones and mobile internet services was perceptible. Their functionality was significantly expanded, along with the usage of these devices – it has become a part of our everyday life. Although one can witness the spread of mobile internet technologies, unfortunately it may not able to keep up with the real needs.

Sometimes mobile application users are struggling with the slowness while browsing a page on the world wide web. They have to wait a lot of time for loading a webpage if circumstances are not acceptable. This can be explained with the poor coverage, therefore the low bandwidth or the load of the server. All of these can greatly affect user experience. One of the solutions we can choose is the prefetching of these webpages.

In my thesis I examine the possibilities provided by the Android platform, and the final goal is to implement an application which shall increase the user’s satisfaction mentioned above. It shall preload the relevant content of each pages. The application works with a large set of keywords, which one can add or edit itself. It is necessary to sort out the one link on which the user clicks with the greatest chance from very many links. That is why I apply a model being based on conditional probability. It helps to create the prognosis using data collected according to past behaviour.

Downloads

Please sign in to download the files of this thesis.