University of Munich

Latest Content From University of Munich

Whitepaper: Computing Clusters of Correlation Connected Objects

by University of MunichJan 01, 2008

This research paper from University of Munich, proposes a method called 4C (Computing Correlation Connected Clusters) to identify local subgroups of the data objects sharing a uniform but arbitrarily complex correlation. The algorithm is based on a combination of Principal Components Analysis (PCA) and Density-BASed ClusteriNg (DBSCAN). The method has a determinate result and is robust against noise. A broad comparative evaluation demonstrates the superior performance of 4C over competing methods such as DBSCAN, CLIQUE and ORCLUS.

Whitepaper: Subspace Selection for Clustering High-Dimensional Data

by University of MunichJan 01, 2008

In high-dimensional feature spaces traditional clustering algorithms tend to break down in terms of efficiency and quality. This research paper from University of Munich presents a feature selection technique called SURFING (SUbspaces Relevant For clusterING) that finds all subspaces interesting for clustering and sorts them by relevance. The sorting is based on a quality criterion for the interestingness of a subspace using the k-nearest neighbor distances of the objects. A broad evaluation based on synthetic and real-world data sets demonstrates that SURFING is suitable to find all relevant subspaces in high dimensional, sparse data sets and produces better results than comparative methods.

Whitepaper: Logging Usage of AJAX Applications With the UsaProxy HTTP Proxy

by University of MunichJan 01, 2008

This research paper by University of Munich shows how to use the UsaProxy HTTP proxy to perform logging of user activity for AJAX web applications. UsaProxy is a special-purpose HTTP proxy which modifies HTML pages before forwarding them to the client browser. It adds JavaScript code which collects data about mouse movement, clicks, key presses and other types of interaction without affecting the user?s browsing experience in any way.

Whitepaper: Patterns of Free Revealing - Balancing Code Sharing and Protection in Commercial Open Source Development

by University of MunichJan 01, 2008

Commercial firms increasingly contribute to the development of Open Source Software (OSS). Commercial firms increasingly contribute to the development of OSS. However, a conflict often arises between the requirements of the General Public License to make ""Derived work"" available, and firms’ interest to protect their intellectual property embodied in the code. If there are ways to mitigate or solve this conflict, the conditions under which OSS will be an appealing solution to firms become much more general. This paper provides a quantitative empirical study of this conflict and the ways firms deal with it.

Whitepaper: Munich/MIT Survey: Development of Embedded Linux

by University of MunichJan 01, 2008

Most of the publicly available code for embedded Linux is developed by commercial firms, not by hobbyists. While it is true that also for standard Linux many contributions come from IBM and other large firms, the situations differ. IBM pursues the strategic goal of establishing Linux as a widely used operating system. In contrast, embedded Linux is an integral part of their products for hardware manufacturers, and the core market offering for specialized software firms. This raises the question if and how the development process of embedded Linux differs from that of other open source software.