Abstract
Today’s rising acceptance of Internet Mail (Email) results in a growing amount of information used for nearly every field of application. A cause from globalization, internationalization and digitalization this information overload yields a proper categorization to focus upon. The problem of categorization to prioritize and focus important information often needs human interaction. The big amount of Email may overtax a human being and may result in causing information loss and business claim.
For assistance, modern Email systems try to organize the information using different approaches of content negotiation techniques. The problems of the topic of content negotiation lead to the conclusion that there are good tools and methods available already, but none of them fits all the scenarios. This thesis focuses on the techniques of content negotiation while discussing the pros and cons, using the results for an outcast of what could be possible when mixing the pros.
Overview
This thesis tries to visualize the complex topic of Content Negotiation in Internet Mail, step by step. Today’s upcoming of information intercommunion leads to an information flood which must be handled to avoid business or private claim. The categorization of information is an information management appendage. Categorization is easy for human beings – but gets more and more impossible while the information diversity grows. Computers have to help. The topic is visualized introducing the basic steps from static categorization up to methods using artificial intelligence techniques.
The practical part of this thesis analyzes each algorithm in different scenarios, using a test framework called Weka (WEKA – University of Waikato, New Zealand, EN, 2006: “Weka – Data Mining with Open Source Machine Learning Software in Java” ) to make the results comparable for review.