Blog

Thoughts, tech notes, in-depth articles and walkthroughs.

Architecture Business Cloud Engineering General Research Talks Thoughts

August 01, 2015 2 mins read

High Dimensional Data Visualization

I attended a seminar on large scale visualization and wrote a paper on high dimensional data visualization, comparing different techniques. A key concept of high dimensional data visualization is dimensionality reduction, in order to reduce the number of dimensions to be visualized. The paper covers linear (principal component analysis) and non-linear (local linear embedding, ISOMAP, t-SNE) dimensionality reduction techniques. The paper has been peer-reviewed in class.

The title photo is a visualization of the MNIST dataset - a database of handwritten digits. The R language was used to create the plots.
Read more

June 01, 2015 1 mins read

Reference Models for Information Management in a Smart Factory

Abstract

In the course of the fourth industrial revolution manufacturing plants are becoming increasingly intelligent and companies have to face the challenge of this increasing information flow. At present, industry and research are striving to find a viable path to enable the vision of the intelligent factory of the future. A key challenge is information management in a highly networked factory, which is also typically networked with its entire supply chain. This work presents four reference models for information management and draws a comparative analysis of the state of research.

March 01, 2015 2 mins read

Influence of slicing tools on quality of 3D printed parts

Project Background

In a team of three students we had the chance to write a seminar paper examining the influence of slicing tools on the quality of 3D printed parts. After finishing the seminar paper, we decided to turn the findings into a research paper, which meant to shorten and rewrite a majority of our original work. We submitted the paper to a special issue of the Computer-Aided Design and Applications journal and the paper got accepted.

November 01, 2014 2 mins read

Bachelor's Thesis: Systematic Architecture Level Fault Diagnosis Using Statistical Techniques

Abstract

In the past various spectrum-based fault localization (SBFL) algorithms have been developed to pinpoint a fault location given a set of failing and passing test executions. Most of the algorithms use similarity coefficients and have only been evaluated on established benchmark programs like the Siemens set or the space program from the Software-artifact Infrastructure Repository. In addition to that, SBFL has not been applied by developers in practice yet. This study evaluates the feasibility of applying SBFL to a real-world project, namely AspectJ. From an initial set of 110 manually classified faulty versions, a maximum of seven bugs can be found after examining the 1000 most suspicious lines produced by various SBFL techniques. To explain the result, the influence of the program size is examined using different metrics and evaluations. In general, the program size has a slight influence on some metrics, but is not the primary explanation for the results. The results seem to originate from the metrics currently used throughout the research community to assess SBFL performance. The study showcases the limitations of SBFL with the help of different performance metrics and the insights learned during manual classification. Moreover, additional performance metrics that are better suited to evaluate the fault localization performance are proposed.