TU Darmstadt / ULB / TUbiblio

Detection of Category Breaches in E-Mail Threads Using RSS Data Feeds

Kirchbuchner, Florian (2012):
Detection of Category Breaches in E-Mail Threads Using RSS Data Feeds.
Technische Universität Darmstadt, [Master Thesis]

Abstract

The use of a single personal e-mail address, by users of both private and business communication, is becoming more and more common. Sometimes private e-mails are sent to business contacts by mistake, and vice versa; the process of separating and distributing such e-mails often creates problems. Again, often e-mail threads are focused on specific categories, which can be used for recognising changes in the content. For example, business contacts usually exchange project details in their e-mail messages, but not private information. In this thesis, we will consider different approaches towards categorising e-mails and present a way to use the categorisation for monitoring outgoing e-mail traffic. The goal is to enhance the security of e-mail communication. As a proof-of-concept, we carried out an extension for Mozilla Thunderbird, which categorises outbound e-mails and warns the user if it detects a change in the category. The individual categories are recognised with the help of speech models, based on RSS data feeds. The use of language models and the modular design also has interesting applications for future development within this field.

Item Type: Master Thesis
Erschienen: 2012
Creators: Kirchbuchner, Florian
Title: Detection of Category Breaches in E-Mail Threads Using RSS Data Feeds
Language: ["languages_typename_1" not defined]
Abstract:

The use of a single personal e-mail address, by users of both private and business communication, is becoming more and more common. Sometimes private e-mails are sent to business contacts by mistake, and vice versa; the process of separating and distributing such e-mails often creates problems. Again, often e-mail threads are focused on specific categories, which can be used for recognising changes in the content. For example, business contacts usually exchange project details in their e-mail messages, but not private information. In this thesis, we will consider different approaches towards categorising e-mails and present a way to use the categorisation for monitoring outgoing e-mail traffic. The goal is to enhance the security of e-mail communication. As a proof-of-concept, we carried out an extension for Mozilla Thunderbird, which categorises outbound e-mails and warns the user if it detects a change in the category. The individual categories are recognised with the help of speech models, based on RSS data feeds. The use of language models and the modular design also has interesting applications for future development within this field.

Uncontrolled Keywords: E-Mail, Privacy, Category, Classification
Divisions: 20 Department of Computer Science > Security, Usability and Society
LOEWE > LOEWE-Zentren > CASED – Center for Advanced Security Research Darmstadt
Profile Areas > Cybersecurity (CYSEC)
LOEWE > LOEWE-Zentren
20 Department of Computer Science
Profile Areas
LOEWE
Event Location: Darmstadt
Date Deposited: 31 Dec 2016 11:42
Identification Number: TUD-CS-2012-0236
Export:
Suche nach Titel in: TUfind oder in Google

Optionen (nur für Redakteure)

View Item View Item