Home | Sitemap | Index | Contact | Legals | KIT

Publications of Matthias Huber

Refereed conference/Workshop papers

[1] Matthias Huber, Christian Henrich, J"orn M"uller-Quade, and Carmen Kempka. Towards secure cloud computing through a separation of duties. In Informatik 2011: Informatik schafft Communities, Beiträge der 41. Jahrestagung der Gesellschaft für Informatik e.V. (GI), 4.-7.10.2011, Berlin (Abstract Proceedings), Hans-Ulrich Heiß, Peter Pepper, Holger Schlingloff, and Jörg Schneider, editors, 2011, volume 192 of LNI. GI. 2011. [ bib ]
[2] Matthias Huber and Jörn Müller-Quade. Methods to secure services in an untrusted environment. In Software Engineering 2011: Fachtagung des GI-Fachbereichs Softwaretechnik, 21.-25. Februar 2011 in Karlsruhe, Ralf Reussner, Matthias Grund, Andreas Oberweis, and Walter F. Tichy, editors, 2011, volume 183 of LNI, pages 159-170. GI. 2011. [ bib ]
[3] Dirk Achenbach, Matthias Gabel, and Matthias Huber. Mimosecco: A middleware for secure cloud storage. In Improving Complex Systems Today, Daniel D. Frey, Shuichi Fukuda, and Georg Rock, editors, Advanced Concurrent Engineering, pages 175-181. Springer London, 2011. 10.1007/978-0-85729-799-0_20. [ bib | http ]
[4] Frank Eichinger, Matthias Huber, and Klemens Böhm. On the Usefulness of Weight-Based Constraints in Frequent Subgraph Mining. In Proceedings of the 30th BCS SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence (AI), Max Bramer, Miltos Petridis, and Adrian Hopgood, editors, December 2010. BCS SGAI, Springer London, UK, Cambridge, UK. December 2010. [ bib | .pdf | Abstract ]
Frequent subgraph mining is an important data-mining technique. In this paper we look at weighted graphs, which are ubiquitous in the real world. The analysis of weights in combination with mining for substructures might yield more precise results. In particular, we study frequent subgraph mining in the presence of weight-based constraints and explain how to integrate them into mining algorithms. While such constraints only yield approximate mining results in most cases, we demonstrate that such results are useful nevertheless and explain this effect. To do so, we both assess the completeness of the approximate result sets, and we carry out application-oriented studies with real-world data-analysis problems: software-defect localization and explorative mining in transportation logistics. Our results are that the runtime can improve by a factor of up to 3.5 in defect localization and 7 in explorative mining. At the same time, we obtain an even slightly increased defect-localization precision and obtain good explorative mining results.
[5] Matthias Huber. Towards secure services in an untrusted environment. In Proceedings of the Fifteenth International Workshop on Component-Oriented Programming (WCOP) 2010, Barbora Bühnová, Ralf H. Reussner, Clemens Szyperski, and Wolfgang Weck, editors, June 2010, volume 2010-14 of Interne Berichte, pages 39-46. Karlsruhe Institue of Technology, Faculty of Informatics, Karlsruhe, Germany. June 2010. [ bib | http | Abstract ]
Software services offer many opportunities like reducedcost for IT infrastructure. However, they also introducenew risks, for example losing control over data. While data canbe secured against external threats using standard techniques, theservice providers themselves have to be trusted to ensure privacy.Cryptographic methods combined with architectures adjustedto the client's protection requirements offer promising methodsto build services with a provable amount of security againstinternal adversaries without the need to fully trust the serviceprovider. We propose a reference architecture which separatesservices, restricts privilege of the parts and deploys them ondifferent servers. Assumptions about the servers' and adversary'scapabilities yield security guarantees which are weaker thanclassical cryptographic guarantees, yet can be sufficient.
[6] Clemens Heidinger, Erik Buchmann, Matthias Huber, Klemens Böhm, and Jörn Müller-Quade. Privacy-aware folksonomies. In ECDL, 2010, pages 156-167. [ bib ]
[7] Christian Henrich, Matthias Huber, Carmen Kempka, Jörn Müller-Quade, and Mario Strefler. Brief announcement: Towards secure cloud computing. In Stabilization, Safety, and Security of Distributed Systems, 11th International Symposium, SSS 2009, Lyon, France, November 3-6, 2009. Proceedings, 2009, pages 785-786. [ bib ]
[8] Frank Eichinger, Klemens Böhm, and Matthias Huber. Mining Edge-Weighted Call Graphs to Localise Software Bugs. In Proceedings of the 8th European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), Walter Daelemans, Bart Goethals, and Katharina Morik, editors, September 2008, volume 5211 of Lecture Notes in Computer Science, pages 333-348. Springer-Verlag Berlin Heidelberg, Germany. September 2008, Part I. [ bib | DOI | .pdf | Abstract ]
An important problem in software engineering is the automated discovery of noncrashing occasional bugs. In this work we address this problem and show that mining of weighted call graphs of program executions is a promising technique. We mine weighted graphs with a combination of structural and numerical techniques. More specifically, we propose a novel reduction technique for call graphs which introduces edge weights. Then we present an analysis technique for such weighted call graphs based on graph mining and on traditional feature selection schemes. The technique generalises previous graph mining approaches as it allows for an analysis of weights. Our evaluation shows that our approach finds bugs which previous approaches cannot detect so far. Our technique also doubles the precision of finding bugs which existing techniques can already localise in principle.
[9] Frank Eichinger, Klemens Böhm, and Matthias Huber. Improved Software Fault Detection with Graph Mining. In Proceedings of the 6th International Workshop on Mining and Learning with Graphs (MLG) at ICML, Samuel Kaski, S.V.N. Vishwanathan, and Stefan Wrobel, editors, July 2008. Helsinki, Finnland. [ bib | .pdf | Abstract ]
This work addresses the problem of discovering bugs in software development. We investigate the utilisation of call graphs of program executions and graph mining algorithms to approach this problem. We propose a novel reduction technique for call graphs which introduces edge weights. Then, we present an analysis technique for such weighted call graphs based on graph mining and on traditional feature selection. Our new approach finds bugs which could not be detected so far. With regard to bugs which can already be localised, our technique also doubles the precision of finding them.


[1] Matthias Huber. Approximatives und diskriminatives Mining von gewichteten Graphen. Master's thesis, Universität Karlsruhe (TH), 2009. [ bib | .pdf | Abstract ]
Graphen mit Kantengewichten treten in vielen Anwendungsdomanen auf, wie zum Beispiel in der Bildverarbeitung, der Transportlogistik, oder der Softwaretechnik. Die Analyse von solchen Graphen mittels Graph-Mining- Techniken ist eine lohnenswerte Aufgabe. Jedoch gibt es keinen Graph- Mining-Algorithmus, der in der Lage ist, kantengewichtete Graphen zu analysieren. Bisher wurden Kantengewichte diskretisiert, damit gewichtete Graphen analysiert werden konnten, oder Kantengewichte wurden erst in einem Postprocessing-Schritt betrachtet. In dieser Arbeit wird eine auf Constraints auf Kantengewichten basierende Erweiterung für die Graph-Mining-Algorithmen gSpan und CloseGraph vorgestellt, welche es ermoglicht, Kantengewichte direkt wahrend dem Mining zu betrachten und zu bewerten. Dadurch ergeben sich neue Pruningmoglichkeiten, welche zu Laufzeitgewinnen fuhren konnen. Es werden verschiedene Methoden vorgestellt, Kantengewichte zu bewerten. Des Weiteren werden diese Moglichkeiten bezuglich der Laufzeit und Ergebnisqualitat mit realen Daten aus den Domanen Transportlogistik und Softwaretechnik evaluiert und verglichen. Es wird gezeigt, dass die in dieser Arbeit vorgestellten Erweiterungen bei anlicher Ergebnisqualitat, zu einer Verbesserung der Laufzeit des Graph- Mining-Algorithmus' fuhren.