SPARCL: an effective and efficient algorithm for mining arbitrary shape-based clusters

Clustering is one of the fundamental data mining tasks. Many different clustering paradigms have been developed over the years, which include partitional, hierarchical, mixture model based, density-based, spectral, subspace, and so on. The focus of this paper is on full-dimensional, arbitrary shaped clusters. Existing methods for this problem suffer either in terms of the memory or time complexity (quadratic or even cubic). This shortcoming has restricted these algorithms to datasets of moderate sizes. In this paper we propose SPARCL, a simple and scalable algorithm for finding clusters with arbitrary shapes and sizes, and it has linear space and time complexity. SPARCL consists of two stages—the first stage runs a carefully initialized version of the Kmeans algorithm to generate many small seed clusters. The second stage iteratively merges the generated clusters to obtain the final shape-based clusters. Experiments were conducted on a variety of datasets to highlight the effectiveness, efficiency, and scalability of our approach. On the large datasets SPARCL is an order of magnitude faster than the best existing approaches.     

Immobilization of recombinant strains of Saccharomyces cerevisiae for the hydrolysis of lactose in salted Domiati cheese whey

Intergeneric protoplast fusants SK-26 and SK-35 between Saccharomyces cerevisiae ATCC 4126 and Kluyveromyces lactis CBS 683 produced 2.70 ml dl^-1 and 1.52 ml dl^-1 (v/v) of ethanol during fermentation of lactose at 25vv°C in salted Domiati cheese whey containing 6.1 g dl^-1 (w/v) NaCl when entrapped in alginate spheres, whereas the free recombinant cells produced 2.36 ml dl^-1 and 1.09 ml dl^-1 (v/v) of ethanol. Yeast hybrids spheres can be used nine times with accumulated increases of ethanol production of 9.1 ml dl^-1 and 8.98 ml dl^-1 (v/v), respectively.     

Guest Editorial: Flexible-Joint and Flexible-Link Robots

Editorial Introduction

Guest Editorial: Flexible-Joint and Flexible-Link Robots     

Efficiently mining frequent trees in a forest: algorithms and applications

Mining frequent trees is very useful in domains like bioinformatics, Web mining, mining semistructured data, etc. We formulate the problem of mining (embedded) subtrees in a forest of rooted, labeled, and ordered trees. We present TREEMINER, a novel algorithm to discover all frequent subtrees in a forest, using a new data structure called scope-list. We contrast TREEMINER with a pattern matching tree mining algorithm (PATTERNMATCHER), and we also compare it with TREEMINERD, which counts only distinct occurrences of a pattern. We conduct detailed experiments to test the performance and scalability of these methods. We also use tree mining to analyze RNA structure and phylogenetics data sets from bioinformatics domain.     

Group-Scheduling for Multi-Service Optical Burst Switching (OBS) Networks

We propose a new burst scheduling mechanism for Optical Burst-Switched (OBS) networks. The new approach is made possible by gathering data bursts into groups and performing the scheduling decision for each group collectively. In OBS group-scheduling, bursts will not be considered for scheduling until a pre-defined time period elapses, during which the group of burst header packets would be gathered. By transforming a set of data bursts into a set of corresponding time intervals, the problem of scheduling these bursts is transformed into a combinatorial optimization problem. Graph algorithms are applied to obtain the maximum number of non-overlapping bursts. The proposed OBS group-scheduling scheme is shown to improve the performance of OBS networks over existing scheduling schemes in terms of burst loss probability and channel utilization. With an extension through a sequential optimization, using a Branch-and-Bound technique, the proposed scheme can support multiple classes of service. It is shown that the new scheduling approach has several desired characteristics including fairness and service differentiability among classes in terms of burst loss probability and channel utilization.     

RISCBOT: A WWW-Enabled Mobile Surveillance and Identification Robot

This article describes RISCBOT (RISCBOT name has been derived from RISC lab and ‘bot’ of robot, a modular 802.11 b-enabled mobile autonomous robot built at the RISC lab of the University of Bridgeport. RISCBOT localizes itself and successfully fulfills www – enabled online user requests and navigates to various rooms, employing a visual recognition algorithm. This article describes the mechanical design, hardware and software algorithms of the robot, and the web-based interface for communicating with the robot. ^⋆RISC lab: Interdisciplinary Robotics, Intelligent Sensing and Control laboratory at the University of Bridgeport.     

Robust Linear Control of Flexible Joint Robot Systems

This paper presents the development and design of a robust linear controller for a flexible joint robot. First, a realistic model is developed taking into account the effects of stick-slip friction, nonlinear spring force characteristic, the rotors' Coriolis and centrifugal acceleration due to the velocities of the precedent links and rotors, and measurement noise. Second, a robust controller is designed using LQG/LTR techniques. Extensive simulation was performed on an experimental robot system using several trajectories. The results demonstrate that the proposed model/controller provides excellent tracking and regulation performance.     

A Computational Model for Interfacial Damage Through Microstructural Cohesive Zone Relaxation

A model introducing cohesive zones around material interfaces to simulate interfacial damage in microheterogeneous materials is developed. The material behavior within the cohesive zones is unknown a-priori, and is weakened, or "relaxed", on the continuum level from an initially undamaged state, by a reduction of the spatially variable elasticity tensor's eigenvalues. This reduction is initiated if constraints placed on the microstress fields, for example critical levels of pressure or deviatoric stresses, are violated. Outside of the cohesive zones the material is unaltered. Numerical computations are performed, employing the finite element method, to illustrate the approach in three dimensional applications.     

High nitrogen steel powder for near net shape products

This study has been performed to relate the powder properties to the process parameters of the plasma-rotating-electrode-process (Prep). A regression equation for average powder diameter was acquired from the experimental data as . A model is developed to describe the nitrogen pick-up by the Prep. On the basis of this model an expression for nitrogen content of the powder produced is deduced as: The powder has a solidification structure of austenitic dendrites. The CrN precipitates, the amount of which is dependent upon the nitrogen content, can be found in the interdendritic areas. During the solution annealing the Cr₂N precipitation reaction occurs in both of the inter- and intradendrite.