Publication details

 

Data Preparation for User Profiling from Traffic Log

Basic information
Original title:Data Preparation for User Profiling from Traffic Log
Author:Marek Kumpošt
Further information
Citation:KUMPOŠT, Marek. Data Preparation for User Profiling from Traffic Log. In Proceedings of The International Conference on Emerging Security Information, Systems, and Technologies (SECURWARE 2007). Los Alamitos, CA, USA : IEEE Computer Society, 2007. ISBN 0-7695-2989-5, pp. 89-94. 14.10.2007, Valencia, Spain.
Original language:English
Field:Informatika
Type:Article in Proceedings
Keywords:user profiling; data mining; traffic log; privacy protection; user profiles

User profiling is a technique that services use to provide customizable content to their users. These services operate with a user profile that reflects users' behaviour in the past. Profiling gives us some evidence about users' behaviour (information about their interests and social background). Instead of requiring users to state their interests, a service could learn this information from users' past activities. The main question is how much private information can be derived from context information that is related to an individual. The second question is to what extent a behaviour pattern can be used to pinpoint an individual among others.