An evaluation of the pedestrian classification in a multi-domain multi-modality setup

The objective of this article is to study the problem of pedestrian classification across different light spectrum domains (visible and far-infrared (FIR)) and modalities (intensity, depth and motion). In recent years, there has been a number of approaches for classifying and detecting pedestrians in both FIR and visible images, but the methods are difficult to compare, because either the datasets are not publicly available or they do not offer a comparison between the two domains. Our two primary contributions are the following: (1) we propose a public dataset, named RIFIR , containing both FIR and visible images collected in an urban environment from a moving vehicle during daytime; and (2) we compare the state-of-the-art features in a multi-modality setup: intensity, depth and flow, in far-infrared over visible domains. The experiments show that features families, intensity self-similarity (ISS), local binary patterns (LBP), local gradient patterns (LGP) and histogram of oriented gradients (HOG), computed from FIR and visible domains are highly complementary, but their relative performance varies across different modalities. In our experiments, the FIR domain has proven superior to the visible one for the task of pedestrian classification, but the overall best results are obtained by a multi-domain multi-modality multi-feature fusion.

Dynamic load balancing for the distributed mining of molecular structures

In molecular biology, it is often desirable to find common properties in large numbers of drug candidates. One family of methods stems from the data mining community, where algorithms to find frequent graphs have received increasing attention over the past years. However, the computational complexity of the underlying problem and the large amount of data to be explored essentially render sequential algorithms useless. In this paper, we present a distributed approach to the frequent subgraph mining problem to discover interesting patterns in molecular compounds. This problem is characterized by a highly irregular search tree, whereby no reliable workload prediction is available. We describe the three main aspects of the proposed distributed algorithm, namely, a dynamic partitioning of the search space, a distribution process based on a peer-to-peer communication framework, and a novel receiverinitiated load balancing algorithm. The effectiveness of the distributed method has been evaluated on the well-known National Cancer Institute’s HIV-screening data set, where we were able to show close-to linear speedup in a network of workstations. The proposed approach also allows for dynamic resource aggregation in a non dedicated computational environment. These features make it suitable for large-scale, multi-domain, heterogeneous environments, such as computational grids.

Decentralized load balancing for highly irregular search problems

In this paper, we present a distributed computing framework for problems characterized by a highly irregular search tree, whereby no reliable workload prediction is available. The framework is based on a peer-to-peer computing environment and dynamic load balancing. The system allows for dynamic resource aggregation, does not depend on any specific meta-computing middleware and is suitable for large-scale, multi-domain, heterogeneous environments, such as computational Grids. Dynamic load balancing policies based on global statistics are known to provide optimal load balancing performance, while randomized techniques provide high scalability. The proposed method combines both advantages and adopts distributed job-pools and a randomized polling technique. The framework has been successfully adopted in a parallel search algorithm for subgraph mining and evaluated on a molecular compounds dataset. The parallel application has shown good calability and close-to linear speedup in a distributed network of workstations.

A hierarchical distributed approach for mining molecular fragments

Recently, two approaches have been introduced that distribute the molecular fragment mining problem. The first approach applies a master/worker topology, the second approach, a completely distributed peer-to-peer system, solves the scalability problem due to the bottleneck at the master node. However, in many real world scenarios the participating computing nodes cannot communicate directly due to administrative policies such as security restrictions. Thus, potential computing power is not accessible to accelerate the mining run. To solve this shortcoming, this work introduces a hierarchical topology of computing resources, which distributes the management over several levels and adapts to the natural structure of those multi-domain architectures. The most important aspect is the load balancing scheme, which has been designed and optimized for the hierarchical structure. The approach allows dynamic aggregation of heterogenous computing resources and is applied to wide area network scenarios.

Efficient mining of discriminative molecular fragments

Frequent pattern discovery in structured data is receiving an increasing attention in many application areas of sciences. However, the computational complexity and the large amount of data to be explored often make the sequential algorithms unsuitable. In this context high performance distributed computing becomes a very interesting and promising approach. In this paper we present a parallel formulation of the frequent subgraph mining problem to discover interesting patterns in molecular compounds. The application is characterized by a highly irregular tree-structured computation. No estimation is available for task workloads, which show a power-law distribution in a wide range. The proposed approach allows dynamic resource aggregation and provides fault and latency tolerance. These features make the distributed application suitable for multi-domain heterogeneous environments, such as computational Grids. The distributed application has been evaluated on the well known National Cancer Institute’s HIV-screening dataset.

PDM 2006: Proceedings of the International Workshop on Data Mining

Recently major processor manufacturers have announced a dramatic shift in their paradigm to increase computing power over the coming years. Instead of focusing on faster clock speeds and more powerful single core CPUs, the trend clearly goes towards multi core systems. This will also result in a paradigm shift for the development of algorithms for computationally expensive tasks, such as data mining applications. Obviously, work on parallel algorithms is not new per se but concentrated efforts in the many application domains are still missing. Multi-core systems, but also clusters of workstations and even large-scale distributed computing infrastructures provide new opportunities and pose new challenges for the design of parallel and distributed algorithms. Since data mining and machine learning systems rely on high performance computing systems, research on the corresponding algorithms must be on the forefront of parallel algorithm research in order to keep pushing data mining and machine learning applications to be more powerful and, especially for the former, interactive. To bring together researchers and practitioners working in this exciting field, a workshop on parallel data mining was organized as part of PKDD/ECML 2006 (Berlin, Germany). The six contributions selected for the program describe various aspects of data mining and machine learning approaches featuring low to high degrees of parallelism: The first contribution focuses the classic problem of distributed association rule mining and focuses on communication efficiency to improve the state of the art. After this a parallelization technique for speeding up decision tree construction by means of thread-level parallelism for shared memory systems is presented. The next paper discusses the design of a parallel approach for dis- tributed memory systems of the frequent subgraphs mining problem. This approach is based on a hierarchical communication topology to solve issues related to multi-domain computational envi- ronments. The forth paper describes the combined use and the customization of software packages to facilitate a top down parallelism in the tuning of Support Vector Machines (SVM) and the next contribution presents an interesting idea concerning parallel training of Conditional Random Fields (CRFs) and motivates their use in labeling sequential data. The last contribution finally focuses on very efficient feature selection. It describes a parallel algorithm for feature selection from random subsets. Selecting the papers included in this volume would not have been possible without the help of an international Program Committee that has provided detailed reviews for each paper. We would like to also thank Matthew Otey who helped with publicity for the workshop.

Optimisation of recombinant production of active human cardiac SERCA2a ATPase

Methods for recombinant production of eukaryotic membrane proteins, yielding sufficient quantity and quality of protein for structural biology, remain a challenge. We describe here, expression and purification optimisation of the human SERCA2a cardiac isoform of Ca2+ translocating ATPase, using Saccharomyces cerevisiae as the heterologous expression system of choice. Two different expression vectors were utilised, allowing expression of C-terminal fusion proteins with a biotinylation domain or a GFP- His8 tag. Solubilised membrane fractions containing the protein of interest were purified onto Streptavidin-Sepharose, Ni-NTA or Talon resin, depending on the fusion tag present. Biotinylated protein was detected using specific antibody directed against SERCA2 and, advantageously, GFP-His8 fusion protein was easily traced during the purification steps using in-gel fluorescence. Importantly, talon resin affinity purification proved more specific than Ni-NTA resin for the GFP-His8 tagged protein, providing better separation of oligomers present, during size exclusion chromatography. The optimised method for expression and purification of human cardiac SERCA2a reported herein, yields purified protein (> 90%) that displays a calcium-dependent thapsigargin-sensitive activity and is suitable for further biophysical, structural and physiological studies. This work provides support for the use of Saccharomyces cerevisiae as a suitable expression system for recombinant production of multi-domain eukaryotic membrane proteins.

GTP binding controls complex formation by the human ROCO protein MASL1

The human ROCO proteins are a family of multi-domain proteins sharing a conserved ROC-COR supra-domain. The family has four members: leu- cine-rich repeat kinase 1 (LRRK1), leucine-rich repeat kinase 2 (LRRK2), death-associated protein kinase 1 (DAPK1) and malignant fibrous histiocy- toma amplified sequences with leucine-rich tandem repeats 1 (MASL1). Previous studies of LRRK1/2 and DAPK1 have shown that the ROC (Ras of complex proteins) domain can bind and hydrolyse GTP, but the cellular consequences of this activity are still unclear. Here, the first biochemical characterization of MASL1 and the impact of GTP binding on MASL1 complex formation are reported. The results demonstrate that MASL1, similar to other ROCO proteins, can bind guanosine nucleotides via its ROC domain. Furthermore, MASL1 exists in two distinct cellular com- plexes associated with heat shock protein 60, and the formation of a low molecular weight pool of MASL1 is modulated by GTP binding. Finally, loss of GTP enhances MASL1 toxicity in cells. Taken together, these data point to a central role for the ROC/GTPase domain of MASL1 in the reg- ulation of its cellular function.

Multi-agent collaboration for decision support in shared zones of intelligent buildings

In domain of intelligent buildings, saving energy in buildings and increasing preferences of occupants are two important factors. These factors are the important keys for evaluating the performance of work environment. In recent years, many researchers combine these areas to create the system that can change from original to the modern work environment called intelligent work environment. Due to advance of agent technology, it has received increasing attention in the area of intelligent pervasive environments. In this paper, we review several issues in intelligent buildings, with respect to the implementation of control system for intelligent buildings via multi-agent systems. Furthermore, we present the MASBO (Multi-Agent System for Building cOntrol) that has been implemented for controlling the building facilities to reach the balancing between energy efficiency and occupant’s comfort. In addition to enhance the MASBO system, the collaboration through negotiation among agents is presented.

The point source method for inverse scattering in the time domain

Many recent inverse scattering techniques have been designed for single frequency scattered fields in the frequency domain. In practice, however, the data is collected in the time domain. Frequency domain inverse scattering algorithms obviously apply to time-harmonic scattering, or nearly time-harmonic scattering, through application of the Fourier transform. Fourier transform techniques can also be applied to non-time-harmonic scattering from pulses. Our goal here is twofold: first, to establish conditions on the time-dependent waves that provide a correspondence between time domain and frequency domain inverse scattering via Fourier transforms without recourse to the conventional limiting amplitude principle; secondly, we apply the analysis in the first part of this work toward the extension of a particular scattering technique, namely the point source method, to scattering from the requisite pulses. Numerical examples illustrate the method and suggest that reconstructions from admissible pulses deliver superior reconstructions compared to straight averaging of multi-frequency data. Copyright (C) 2006 John Wiley & Sons, Ltd.

Typification of farming systems for constructing representative farm models: two illustrations of the application of multi-variate analyses in Chile and Pakistan

If the fundamental precepts of Farming Systems Research were to be taken literally then it would imply that for each farm 'unique' solutions should be sought. This is an unrealistic expectation, but it has led to the idea of a recommendation domain, implying creating a taxonomy of farms, in order to increase the general applicability of recommendations. Mathematical programming models are an established means of generating recommended solutions, but for such models to be effective they have to be constructed for 'truly' typical or representative situations. The multi-variate statistical techniques provide a means of creating the required typologies, particularly when an exhaustive database is available. This paper illustrates the application of this methodology in two different studies that shared the common purpose of identifying types of farming systems in their respective study areas. The issues related with the use of factor and cluster analyses for farm typification prior to building representative mathematical programming models for Chile and Pakistan are highlighted. (C) 2003 Elsevier Science Ltd. All rights reserved.

A normative approach to multi-agent systems for intelligent buildings

Building Management Systems (BMS) are widely adopted in modern buildings around the world in order to provide high-quality building services, and reduce the running cost of the building. However, most BMS are functionality-oriented and do not consider user personalization. The aim of this research is to capture and represent building management rules using organizational semiotics methods. We implement Semantic Analysis, which determines semantic units in building management and their relationship patterns of behaviour, and Norm Analysis, which extracts and specifies the norms that establish how and when these management actions occur. Finally, we propose a multi-agent framework for norm based building management. This framework contributes to the design domain of intelligent building management system by defining a set of behaviour patterns, and the norms that govern the real-time behaviour in a building.

Analysis of frequency and intensity of European winter storm events from a multi-model perspective, at synoptic and regional scales

This study focuses on the analysis of winter (October-November-December-January-February-March; ONDJFM) storm events and their changes due to increased anthropogenic greenhouse gas concentrations over Europe. In order to assess uncertainties that are due to model formulation, 4 regional climate models (RCMs) with 5 high resolution experiments, and 4 global general circulation models (GCMs) are considered. Firstly, cyclone systems as synoptic scale processes in winter are investigated, as they are a principal cause of the occurrence of extreme, damage-causing wind speeds. This is achieved by use of an objective cyclone identification and tracking algorithm applied to GCMs. Secondly, changes in extreme near-surface wind speeds are analysed. Based on percentile thresholds, the studied extreme wind speed indices allow a consistent analysis over Europe that takes systematic deviations of the models into account. Relative changes in both intensity and frequency of extreme winds and their related uncertainties are assessed and related to changing patterns of extreme cyclones. A common feature of all investigated GCMs is a reduced track density over central Europe under climate change conditions, if all systems are considered. If only extreme (i.e. the strongest 5%) cyclones are taken into account, an increasing cyclone activity for western parts of central Europe is apparent; however, the climate change signal reveals a reduced spatial coherency when compared to all systems, which exposes partially contrary results. With respect to extreme wind speeds, significant positive changes in intensity and frequency are obtained over at least 3 and 20% of the European domain under study (35–72°N and 15°W–43°E), respectively. Location and extension of the affected areas (up to 60 and 50% of the domain for intensity and frequency, respectively), as well as levels of changes (up to +15 and +200% for intensity and frequency, respectively) are shown to be highly dependent on the driving GCM, whereas differences between RCMs when driven by the same GCM are relatively small.

The contributions of domain-general and numerical factors to third-grade arithmetic skills and mathematical learning disability

Explanations of the marked individual differences in elementary school mathematical achievement and mathematical learning disability (MLD or dyscalculia) have involved domain-general factors (working memory, reasoning, processing speed and oral language) and numerical factors that include single-digit processing efficiency and multi-digit skills such as number system knowledge and estimation. This study of third graders (N = 258) finds both domain-general and numerical factors contribute independently to explaining variation in three significant arithmetic skills: basic calculation fluency, written multi-digit computation, and arithmetic word problems. Estimation accuracy and number system knowledge show the strongest associations with every skill and their contributions are both independent of each other and other factors. Different domain-general factors independently account for variation in each skill. Numeral comparison, a single digit processing skill, uniquely accounts for variation in basic calculation. Subsamples of children with MLD (at or below 10th percentile, n = 29) are compared with low achievement (LA, 11th to 25th percentiles, n = 42) and typical achievement (above 25th percentile, n = 187). Examination of these and subsets with persistent difficulties supports a multiple deficits view of number difficulties: most children with number difficulties exhibit deficits in both domain-general and numerical factors. The only factor deficit common to all persistent MLD children is in multi-digit skills. These findings indicate that many factors matter but multi-digit skills matter most in third grade mathematical achievement.

Fine-grained multi-phase array designs

Hybrid multiprocessor architectures which combine re-configurable computing and multiprocessors on a chip are being proposed to transcend the performance of standard multi-core parallel systems. Both fine-grained and coarse-grained parallel algorithm implementations are feasible in such hybrid frameworks. A compositional strategy for designing fine-grained multi-phase regular processor arrays to target hybrid architectures is presented in this paper. The method is based on deriving component designs using classical regular array techniques and composing the components into a unified global design. Effective designs with phase-changes and data routing at run-time are characteristics of these designs. In order to describe the data transfer between phases, the concept of communication domain is introduced so that the producer–consumer relationship arising from multi-phase computation can be treated in a unified way as a data routing phase. This technique is applied to derive new designs of multi-phase regular arrays with different dataflow between phases of computation.