Maschinenbau und Mechatronik
T. Gruber, Bernhard Sick, D. Fisch
SwiftRule: Mining Comprehensible Classication Rules for Time Series Analysis
IEEE Transactions on Knowledge and Data Engineering, vol. 23, no. 5, pp. 774-787
In this article, we provide a new technique for temporal data mining which is based on classification rules that can easily be understood by human domain experts. Basically, time series are decomposed into short segments, and short-term trends of the time series within the segments (e.g., average, slope, and curvature) are described by means of polynomial models. Then, the classifiers assess short sequences of trends in subsequent segments with their rule premises. The conclusions gradually assign an input to a class. As the classifier is a generative model of the processes from which the time series are assumed to originate, anomalies can be detected, too. Segmentation and piecewise polynomial modeling are done extremely fast in only one pass over the time series. Thus, the approach is applicable to problems with harsh timing constraints. We lay the theoretical foundations for this classifier, including a new distance measure for time series and a new technique to construct a dynamic classifier from a static one, and demonstrate its properties by means of various benchmark time series, for example, Lorenz attractor time series, energy consumption in a building, or ECG data.
Maschinenbau und Mechatronik
A. Hofmann, Bernhard Sick
On-Line Intrusion Alert Aggregation With Generative Data Stream Modeling
IEEE Transactions on Dependable and Secure Computing, vol. 8, no. 2, pp. 282-294
Alert aggregation is an important subtask of intrusion detection. The goal is to identify and to cluster different alerts—produced by low-level intrusion detection systems, firewalls, etc.—belonging to a specific attack instance which has been initiated by an attacker at a certain point in time. Thus, meta-alerts can be generated for the clusters that contain all the relevant information whereas the amount of data (i.e., alerts) can be reduced substantially. Meta-alerts may then be the basis for reporting to security experts or for communication within a distributed intrusion detection system. We propose a novel technique for online alert aggregation which is based on a dynamic, probabilistic model of the current attack situation. Basically, it can be regarded as a data stream version of a maximum likelihood approach for the estimation of the model parameters. With three benchmark data sets, we demonstrate that it is possible to achieve reduction rates of up to 99.96 percent while the number of missing meta-alerts is extremely low. In addition, meta-alerts are generated with a delay of typically only a few seconds after observing the first alert belonging to a new attack instance.
NachhaltigEuropan Campus Rottal-Inn
C.-H. Solterbeck, M. Mátéfi-Tempfli, M. Es-Souni, Stefan Mátéfi-Tempfli, S. Habouti
Self-standing corrugated Ag and Au-Nanorods as plasmonic substrates
Journal of Materials Chemistry, vol. 21, pp. 6269-6273
We use home-made Si-supported anodized alumina thin film templates for the electrodeposition of large area self-standing Ag- and Au-nanorod (Au-NR) arrays. The deposition conditions chosen, i.e. electrolyte composition and deposition voltage, lead to a corrugated rod morphology, particularly for Au-NRs. Instantaneous nucleation followed by diffusion-controlled growth are thought to be the dominating mechanism for the morphology observed. Diffuse reflectance spectra show specific behaviours of Ag- and Au-NRs with longitudinal and transverse plasmon resonance modes and additional modes for Ag-NRs. The activity of the NR arrays as substrates for molecular detection using Raman scattering and Rhodamine 6G (R6G) as a model dye strongly depends on noble metal. R6G concentrations down to 1 pM are detected on the corrugated arrays yielding an effective enhancement factor (EF) of approximately 2 × 1010 for Ag-NRs and 1 × 109 for Au-NRs. The latter is the highest ever obtained for Au-nanostructures. Both nanostructures provide an enhancement that is high enough to detect single molecules using Raman scattering. The results are rationalized in terms of morphology effects on electromagnetic field intensity.
Hochschulleitung und -einrichtungen
N. Venter, Robert Rossberger, D. Krause, T. Joubert, K. Dowdeswell
Assessment center practices in South Africa
International Journal of Selection and Assessment, vol. 19, no. 3, pp. 262-275
Despite the popularity of assessment centers (AC) in South Africa, no recent study exists that describes AC practices in that region. Given this research gap, we conducted a survey study that analyzes the development, execution, and evaluation of ACs in N=43 South African organizations. We report findings regarding AC design, job analysis and job requirements assessed, target groups and positions of the participants after the AC, number and kind of exercises used, additional diagnostic methods used, assessors and characteristics considered in constitution of the assessor pool, observational systems and rotation plan, characteristics, contents, and methods of assessor training, types of information provided to participants, data integration process, use of self- and peer-rating, characteristics of the feedback process, and features after the AC. Finally, we compare the results with professional suggestions to identify pros and cons in current South African AC practices and offer suggestions for improvement.