Research & Projects

HelloWorld Goes Mobile, a scalable, encrypted, and decentralized social network for Google Android compatible mobile phones, is location aware and provides geolocating, mapping, friend-locating, as directions and distance features for friends, objects and services. The software relies on open standards (XML, OpenID and OpenSocial), public key infrastructures (private- and public key concepts), signatures, and Java-encryption technologies like RSA, PBEMD5 and AES.
Students: Artur Friesen, Kok-Chee Lim, Tobias Schröter and Julian Weigle. Winter 2008/2009.
Social Movie Map analyzes and visualizes the movie consumption patterns of MySpace users, focusing on the relationships between listed movies, user profiles, origin, age and gender. Social Movie Map is currently based on a data set of more than 6 million MySpace user profiles, collected by a distributed information retrieval system. The visualization is based on the Adobe Flex and open source visual analytics library “BirdEye RaVis”, the analysis is based on statistics, Bayesian and Levenshtein algorithms.
Students: Philipp Bender and Florian Wilhelm. Winter 2008/2009.
Bookstats analyzes the reading and literature consumption patterns within social networks based on the profile data of about 6 million MySpace users. The application uses Java, PHP, MySQL, Flash, ActionScript and Adobe Flex to generate charts and treemap visualizations in a web based interface, filtering results by age, country of origin, ethnicity, gender, and income. Reflecting the demographic change in media usage, from six million profiles only one million included favourite book titles, "Harry Potter" was featured in all cultures and religions right next to the Bible and the Qur'an.
Students: Sebastian Daum and Anika Faber. Winter 2008/2009.
WorldMusicTracker focuses on Myspace, analyzing music bands of different genres, audiences, popularity, and interactions across the globe. The analysis is based on a datasample of several million users, 600.000 bands, and 3 million comments. The application integrates uses MySQL, JSP, HTML, CSS, and Ampcharts for heatmaps and visualisations. WorldMusicTrackers provides solid evidence for the popularity of American, Canadian and British bands, fitting mostly in the genres Hip-Hop, Rap and R&B, and reveals that most user comments are created in the Winter, generally after 8pm.
Students: Thomas Kuhn and Eugen Schäfer. Winter 2008/2009.
HelloWorld, a distributed social network without formulated centers of gravity or walled gardens of the proprietary variety, is based on distributed identity management, strong encryption, open standards, micro formats, and ease of use. HelloWorld allows users to protect identities, relationships, interests and networks.
Students: Markus Ackermann, Krister Hymon, Benjamin Ludwig, and Kai Wilhelm. Summer 2008.
Crimeblips provides up-to-date crime maps and statistics for neighborhoods throughout Berlin (Germany). The application uses Bayesian algorithms to extract, filter and categorize relevant data; maps, visualizes, and analyzes crime incident patterns; and allows users to identify crime hot spots, trends and general patterns. The occurrences, events and data are extracted from public press releases of the City of Berlin Police Department. The interpretation of mapped and visualized results as well as further conclusions should be discussed with the local police department.
Students: Stefan Kovachev and Patrick Reichert. Summer 2008.
Tracing the Past: The Synagogues, explores the history and complexity of Jewish culture in Germany, focusing on the destruction of synagogues during the Kristallnacht of November 9-10 1938. The project provides an interactive map, visualizing the scale and dimension of damage and destruction, Jewish life present and past, combining GoogleMaps, charts, photographies, audio/visual material, eye witness reports, user generated data, newspaper references, wikipedia entries, and link collections.
Students: Sabine Bach, Bernd Eppler, Isabelle Schlitzer, Daniel von der Helm, Dennis Kleine, David Schneider, and Michael Stoll. Summer 2008.
StudiMaps, based on Google Maps, is a classic MashUp for the analysis and visualization of social relationships within a social network enjoying popularity amongst German students. The open source framework offers data, charts, and visualisations of geographical connections and provides interfaces to mapping services, driving directions and train schedules.
Students: Boris Hasselmann and Thomas Overbeck. Winter 2007/2008.
Gervis, an analysis and visualisation framework focusing on the intersection between politics and German industries, is based on Infrat$tat and Nepotist Analyzer. The application visualizes the relationships between companies listed at the stock exchange, board members, politicians, interests, and affiliations. The open source framework includes charts, statistics, salary information, revenue and employment tendencies as well as campaign contributions and individual donations reported to the Deutsche Bundestag.
Students: Christina Dombrowski, Stefan Schreiber and Daniela Natale. Winter 2007/2008.
Infra$tat visualizes the relationships between companies, board of directors and supervisory boards of registered companies in Germany, providing a detailed picture of the interrelationships of the Deutschland AG. The framework relies on data provided by commercial content providers and visualizes the data with the visualization toolkit Prefuse. The framework is currently expanded to include political affiliations and campaign contributions.
Students: Michael Kabdebo, Marc Rohe, and Florian Bielsky. Summer 2007.
Facelift is a framework for the visualization and analysis of online social networking services. The framework displays communities as node-link diagrams and provides several search, filtering, clustering and analysis features. The current release provides an implementation for the online community Hi5.com, allowing members to download, analyze and visualize member, friend, user, and relationship information.
Students: Patrick Opitz and Sabrina Hunsicker. Summer 2007.
Last.forward analyzes and visualizes Last.fm, a web service and social network based on music, artists and user preferences. The free software framework Last.forward analyzes the relationships between individual users, friends, groups, and tastes and visualizes the results with the visualization toolkit Prefuse.
Students: Thomas Knoll, Christian Endlich, and Heike Scherer. Summer 2007.
StudiAnalyse, a free software project that analyzes and visualizes a German social network similar to Facebook, that was launched in October 2005 and still enjoys popularity amongst students in Germany
Students: Christoph Gerstle and Florian Moritz. Winter 2006/2007.
Nepotist Analyzer visualizes the relationships between commercial and political interests in Germany, providing a detailed picture of the Deutschland AG. The framework relies on data provided by commercial content providers and visualizes the data within a modular XML/SVG based visualization module.
Students: Charlene Beavers and Rainer Pfister. Winter 2006/2007.
SpamCan is a web based framework, allowing users to attract, analyze and locate spam, harvester, and spammer. Spam Emails are categorized into several categories using Bayesian filter.
Students: Katharina Gerhardt, Ralf Wagner, and Yvonne Schmittler. Winter 2006/2007.
Visual Ciao visualizes the social network of Ciao.de, a consumer centric community, providing a detailed analysis of age and gender distribution, experience, credits and community participation. The visualizations of Visual Ciao, an open source application written in Java, can be customized to explore specific sub communities, themes or individual members.
Students: Heidi Boeckh and Rainer Storck. Winter 2006/2007.
PeerLo is a plugin for the bittorrent-client "Azureus". With this plugin it is possible to visualize the swarm of users trying to download a particular torrent file. The plug-in identifies the location of each swarm member and visualizes the entire network on a map. Peerlo also provides some statistics about the origin and the downloading/uploading behavior of individual clients or networks.
Students: Andreas Augustin and Christian Becker. Winter 2006/2007.
Vinex (virtual network explorer) reduces the complexity of social networks by visualizing their structures. Based on data collected by web crawlers, Vinex offers the possibility to explore relationships between social networks, groups, and individual community members as well as interests, coordinates, profiles, addresses, email addresses, and themes.
Students: Pascal Jaubert and Frank Langhans. Summer 2006.
WebLoupe explores, analyses and visualizes link structures and websites. Filter criteria for charts, tree maps and visualisations include search depth, domain limitations, specific file sizes, ping/connection times, Mime types, keywords, and online/offline searches.
Students: Matthias Kahlau and Katja Langholz. Winter 2004/2005.
Wlandscape identifies, analyses and visualizes wireless network information, focusing on SSID, channel, signal strength, status, location, and coverage. Networks and network information are visualized on a topographical map.
Students: Benjamin Angel, Niko Bender, Jochen Bräuer, Christoph Müller, Steffen Schuler, Marcus Schug, Felix Noz, Thomas Müller, Steffen Schreiner and Christian Graß. Summer 2003.

Projects

HelloWorld Goes Mobile HelloWorld Goes Mobile
Social Movie Map Social Movie Map
Bookstats Bookstats
Bookstats Bookstats
HelloWorld HelloWorld
Crimeblips - Berlin Crime Map Crimeblips - Berlin Crime Map
Synagogues - Tracing the Past Synagogues - Tracing the Past
StudiMaps StudiMaps
GerVis Deutschland AG GerVis Deutschland AG
infra$tat infra$tat
facelift facelift
lastforward lastforward
studiAnalyse studiAnalyse
Nepotist Analyzer Nepotist Analyzer
SpamCan SpamCan
Visual Ciao Visual Ciao
PeerLo PeerLo
Vinex Vinex
Web Loupe Web Loupe
Wlandscape Wlandscape