Finding density-based subspace clusters in graphs with feature vectors

Data sources representing attribute information in combination with network information are widely available in today??s applications. To realize the full potential for knowledge extraction, mining techniques like clustering should consider both information types simultaneously. Recent clustering approaches combine subspace clustering with dense subgraph mining to identify groups of objects that are similar in subsets of their attributes as well as densely connected within the network. While those approaches successfully circumvent the problem of full-space clustering, their limited cluster definitions are restricted to clusters of certain shapes. In this work we introduce a density-based cluster definition, which takes into account the attribute similarity in subspaces as well as a local graph density and enables us to detect clusters of arbitrary shape and size. Furthermore, we avoid redundancy in the result by selecting only the most interesting non-redundant clusters. Based on this model, we introduce the clustering algorithm DB-CSC, which uses a fixed point iteration method to efficiently determine the clustering solution. We prove the correctness and complexity of this fixed point iteration analytically. In thorough experiments we demonstrate the strength of DB-CSC in comparison to related approaches.     

Fault-tolerant and maximally-permissive FMS controllers right out of the box—an automatic design approach

This paper presents a comprehensive, all-in-one approach to automatic generation of high-level coordination controllers for flexible manufacturing systems (FMS). The approach comprises an intuitive graphical world wide web (WWW) user interface to specify the FMS layout and the desired production sequences. It also includes algorithms (a) to generate a proper Petri Net model for the specified plant, (b) to set up optimal (i.e. minimally restrictive) collision and deadlock avoidance policies together with a proper supervisory controller to enforce these policies, and (c) to react to disturbances such as machine breakdowns. These algorithms run in the design (offline) phase without user interaction. During online operation the operator only needs to indicate the arrival of new raw materials and the removal of finished products using the same user interface as mentioned before. An example implementation demonstrates the applicability and efficiency of the proposed approach.     

Verbundbrücke Pöcking – Innovative VFT‐Träger mit Betondübeln

The new road bridge over the railway line next to Pöcking applied to a new construction method for composite bridges. The VFT®‐construction method, which has been established in Germany within the last years combined with concrete dowel technology, generated a new type of prefabricated composite girders. Concrete dowels, which are manufactured by special steel web cuts connect rolled steel profiles with pre‐cast concrete members. An increasing degree of standardisation and prefabrication enables the application with an economic use of resources and a high quality level. This article describes design and manufacturing aspects of the new bridge, which is already under traffic use.     

Über die Perlcopolymerisation von Acrylsäureethylester — Acrylnitril — Divinylbenzol im thermisch isolierten System

The Kinetics of the suspension copolymerization of ethyl acrylate, acrylonitrile, and divinylbenzene were studied and described by means of the time dependences of conversion α and of the momentary polymerization rate R_p. The appearance of the Trommsdorff effect already in the range beginning with α = 0,02 is a characteristic feature of this copolymerization. The dependence of the polymerization rate on temperature at the beginning of the polymerization was calculated as well as its dependence on the initial concentration of initiator. The magnitude of the activation energy of the copolymerization was determined from the above values which is E = 125 KJ · mol^?1.     

Fabrication and microstructure characterization of inert matrix fuel based on yttria stabilized zirconia

The deployment of a suitable, Pu-bearing inert matrix fuel (IMF) could offer an attractive option as a single-recycling LWR strategy aimed at reducing the currently growing plutonium stockpiles. A development programme focusing on yttria stabilized zirconia (YSZ)-based IMF is conducted at PSI. YSZ-based IMF has so far been irradiated in two test reactors. The fabrication routes as well as the characterization of the irradiated material by ceramography, electronprobe microanalysis, and X-ray diffraction are presented. IMF fabrication by attrition milling of the oxide constituents is possible, but high sintering temperatures are required to achieve homogeneity. X-ray diffraction is a suitable tool to monitor the homogeneity. Extra efforts are needed to increase the density.     

Voltage-Controlled Electron-Hole Interaction in a Single Quantum Dot

The ground state of neutral and negatively charged excitons confined to a single self-assembled InGaAs quantum dot is probed in a direct absorption experiment by high resolution laser spectroscopy. We show how the anisotropic electron-hole exchange interaction depends on the exciton charge and demonstrate how the interaction can be switched on and off with a small dc voltage. Furthermore, we report polarization sensitive analysis of the excitonic interband transition in a single quantum dot as a function of charge with and without magnetic field.     

Optical font recognition using typographical features

A new statistical approach based on global typographical features is proposed to the widely neglected problem of font recognition. It aims at the identification of the typeface, weight, slope and size of the text from an image block without any knowledge of the content of that text. The recognition is based on a multivariate Bayesian classifier and operates on a given set of known fonts. The effectiveness of the adopted approach has been experimented on a set of 280 fonts. Font recognition accuracies of about 97 percent were reached on high-quality images. In addition, rates higher than 99.9 percent were obtained for weight and slope detection. Experiments have also shown the system robustness to document language and text content and its sensitivity to text length