Below you can find some screenshots of the Deep Email Miner v1 in action.
All screenshots were made using Fedora Linux and the Gnome Desktop.
The standard view. Each employee is displayed as a vertex in the
social network. An edge between two vertices indicates the
existence of email traffic between them, the edge's color its
extent.
In the clustering mode, edges are removed in the order of their
edge betweenness centrality.
The available statistics quantify the structure of the email
corpus.
The email set viewer displays all emails for a
selected edge or the internal/external incoming/outgoing emails
of an employee. They can be displayed via a double-click on the
email subject.
In the filtering mode, different edge and email properties can
be selected. The recipient type (TO/CC/BCC), the date and the
number of emails an edge must contain to be displayed.
Large and complex networks can easily be made more accessible
using this mechanism.
Ranking algorithms rank the vertices and edges on according to
different measures including email data and social network analysis
functions.
An integrated thread detection mechanism bundles single emails
with different senders to email threads.
The email labelling mechanism builds the foundation for the text
mining functionality.
All email data has to be loaded from the database.