Feature selection using expected attainable discrimination

We propose expected attainable discrimination (EAD) as a measure to select discrete valued features for reliable discrimination between two classes of data. EAD is an average of the area under the ROC curves obtained when a simple histogram probability density model is trained and tested on many random partitions of a data set. EAD can be incorporated into various stepwise search methods to determine promising subsets of features, particularly when misclassification costs are difficult or impossible to specify. Experimental application to the problem of risk prediction in pregnancy is described.

A new model for airport passenger segmentation

Recent changes in the aviation industry and in the expectations of travellers have begun to alter the way we approach our understanding, and thus the segmentation, of airport passengers. The key to successful segmentation of any population lies in the selection of the criteria on which the partitions are based. Increasingly, the basic criteria used to segment passengers (purpose of trip and frequency of travel) no longer provide adequate insights into the passenger experience. In this paper, we propose a new model for passenger segmentation based on the passenger core value, time. The results are based on qualitative research conducted in-situ at Brisbane International Terminal during 2012-2013. Based on our research, a relationship between time sensitivity and degree of passenger engagement was identified. This relationship was used as the basis for a new passenger segmentation model, namely: Airport Enthusiast (engaged, non time sensitive); Time Filler (non engaged, non time sensitive); Efficiency Lover (non engaged, time sensitive) and Efficient Enthusiast (engaged, time sensitive). The outcomes of this research extend the theoretical knowledge about passenger experience in the terminal environment. These new insights can ultimately be used to optimise the allocation of space for future terminal planning and design.

Investigation of the effect of balconies on natural ventilation of dwellings in high-rise residential buildings in subtropical climate

Balconies, as one of the main architectural features in subtropical climates, are assumed to enhance the ventilation performance of buildings by redirecting the wind. Although there are some studies on the effect of balconies on natural ventilation inside buildings, the majority have been conducted on single zone buildings with simple geometries. The purpose of this study is to explore the effect of balconies on the internal air flow pattern and ventilation performance of multi-storey residential buildings with internal partitions. To this end, a sample residential unit was selected for investigation and three different conditions tested, base case (no balcony), an open balcony and a semi-enclosed balcony. Computational Fluid Dynamics is used as an analysis method due to its accuracy and ability to provide detailed results. The cases are analysed in terms of average velocity, flow uniformity and number of Air Changes per Hour (ACH). The results suggest the introduction of a semi-enclosed balcony into high-rise dwellings improves the average velocity and flow uniformity. Integrating an open balcony results in reduction of the aforementioned parameters at 0° wind incidence.

Turbulent mixed convection in a shallow enclosure with a series of heat generating components

Turbulent mixed convection flow and heat transfer in a shallow enclosure with and without partitions and with a series of block-like heat generating components is studied numerically for a range of Reynolds and Grashof numbers with a time-dependent formulation. The flow and temperature distributions are taken to be two-dimensional. Regions with the same velocity and temperature distributions can be identified assuming repeated placement of the blocks and fluid entry and exit openings at regular distances, neglecting the end wall effects. One half of such module is chosen as the computational domain taking into account the symmetry about the vertical centreline. The mixed convection inlet velocity is treated as the sum of forced and natural convection components, with the individual components delineated based on pressure drop across the enclosure. The Reynolds number is based on forced convection velocity. Turbulence computations are performed using the standard k– model and the Launder–Sharma low-Reynolds number k– model. The results show that higher Reynolds numbers tend to create a recirculation region of increasing strength in the core region and that the effect of buoyancy becomes insignificant beyond a Reynolds number of typically 5×105. The Euler number in turbulent flows is higher by about 30 per cent than that in the laminar regime. The dimensionless inlet velocity in pure natural convection varies as Gr1/3. Results are also presented for a number of quantities of interest such as the flow and temperature distributions, Nusselt number, pressure drop and the maximum dimensionless temperature in the block, along with correlations.

A computationally efficient technique for data-clustering

A computationally efficient agglomerative clustering algorithm based on multilevel theory is presented. Here, the data set is divided randomly into a number of partitions. The samples of each such partition are clustered separately using hierarchical agglomerative clustering algorithm to form sub-clusters. These are merged at higher levels to get the final classification. This algorithm leads to the same classification as that of hierarchical agglomerative clustering algorithm when the clusters are well separated. The advantages of this algorithm are short run time and small storage requirement. It is observed that the savings, in storage space and computation time, increase nonlinearly with the sample size.

Problems and Algorithms for Sequence Segmentations

The analysis of sequential data is required in many diverse areas such as telecommunications, stock market analysis, and bioinformatics. A basic problem related to the analysis of sequential data is the sequence segmentation problem. A sequence segmentation is a partition of the sequence into a number of non-overlapping segments that cover all data points, such that each segment is as homogeneous as possible. This problem can be solved optimally using a standard dynamic programming algorithm. In the first part of the thesis, we present a new approximation algorithm for the sequence segmentation problem. This algorithm has smaller running time than the optimal dynamic programming algorithm, while it has bounded approximation ratio. The basic idea is to divide the input sequence into subsequences, solve the problem optimally in each subsequence, and then appropriately combine the solutions to the subproblems into one final solution. In the second part of the thesis, we study alternative segmentation models that are devised to better fit the data. More specifically, we focus on clustered segmentations and segmentations with rearrangements. While in the standard segmentation of a multidimensional sequence all dimensions share the same segment boundaries, in a clustered segmentation the multidimensional sequence is segmented in such a way that dimensions are allowed to form clusters. Each cluster of dimensions is then segmented separately. We formally define the problem of clustered segmentations and we experimentally show that segmenting sequences using this segmentation model, leads to solutions with smaller error for the same model cost. Segmentation with rearrangements is a novel variation to the segmentation problem: in addition to partitioning the sequence we also seek to apply a limited amount of reordering, so that the overall representation error is minimized. We formulate the problem of segmentation with rearrangements and we show that it is an NP-hard problem to solve or even to approximate. We devise effective algorithms for the proposed problem, combining ideas from dynamic programming and outlier detection algorithms in sequences. In the final part of the thesis, we discuss the problem of aggregating results of segmentation algorithms on the same set of data points. In this case, we are interested in producing a partitioning of the data that agrees as much as possible with the input partitions. We show that this problem can be solved optimally in polynomial time using dynamic programming. Furthermore, we show that not all data points are candidates for segment boundaries in the optimal solution.

Performance estimation of streaming applications for hierarchical MPSoCs

Parallel programming and effective partitioning of applications for embedded many-core architectures requires optimization algorithms. However, these algorithms have to quickly evaluate thousands of different partitions. We present a fast performance estimator embedded in a parallelizing compiler for streaming applications. The estimator combines a single execution-based simulation and an analytic approach. Experimental results demonstrate that the estimator has a mean error of 2.6% and computes its estimation 2848 times faster compared to a cycle accurate simulator.

Highlight: Yale University Art Gallery

Modern architecture, with its exposed concrete, glass, and steel expanses, does not age gracefully. The Yale University Art Gallery, designed by architecture great Louis Kahn, is no exception. As Yale's first modernist building, completed in 1953, and Kahn's first major commission and collaboration with pioneering lighting designer, Richard Kelly, the gallery is an important part of American architecture. Yet, despite its iconic status, the building suffered numerous architectural indignities in the years following its completion, including the insertion of permanent gallery partitions, which divided Kahn's open plan, and the enclosure of an exterior court, which blocked daylight to the lower galleries.

Evaluating Option Pricing Models

This study evaluates three different time units in option pricing: trading time, calendar time and continuous time using discrete approximations (CTDA). The CTDA-time model partitions the trading day into 30-minute intervals, where each interval is given a weight corresponding to the historical volatility in the respective interval. Furthermore, the non-trading volatility, both overnight and weekend volatility, is included in the first interval of the trading day in the CTDA model. The three models are tested on market prices. The results indicate that the trading-time model gives the best fit to market prices in line with the results of previous studies, but contrary to expectations under non-arbitrage option pricing. Under non-arbitrage pricing, the option premium should reflect the cost of hedging the expected volatility during the option’s remaining life. The study concludes that the historical patterns in volatility are not fully accounted for by the market, rather the market prices options closer to trading time.

Molecular Caches: A caching structure for dynamic creation of application-specific Heterogeneous cache regions

CMPs enable simultaneous execution of multiple applications on the same platforms that share cache resources. Diversity in the cache access patterns of these simultaneously executing applications can potentially trigger inter-application interference, leading to cache pollution. Whereas a large cache can ameliorate this problem, the issues of larger power consumption with increasing cache size, amplified at sub-100nm technologies, makes this solution prohibitive. In this paper in order to address the issues relating to power-aware performance of caches, we propose a caching structure that addresses the following: 1. Definition of application-specific cache partitions as an aggregation of caching units (molecules). The parameters of each molecule namely size, associativity and line size are chosen so that the power consumed by it and access time are optimal for the given technology. 2. Application-Specific resizing of cache partitions with variable and adaptive associativity per cache line, way size and variable line size. 3. A replacement policy that is transparent to the partition in terms of size, heterogeneity in associativity and line size. Through simulation studies we establish the superiority of molecular cache (caches built as aggregations of molecules) that offers a 29% power advantage over that of an equivalently performing traditional cache.

Processing map for hot working of alpha-zirconium

The hot deformation characteristics of alpha-zirconium in the temperature range of 650 °C to 850 °C and in the strain-rate range of 10-3 to 102 s-1 are studied with the help of a power dissipation map developed on the basis of the Dynamic Materials Model.[7,8,9] The processing map describes the variation of the efficiency of power dissipation (η =2m/m + 1) calculated on the basis of the strain-rate sensitivity parameter (m), which partitions power dissipation between thermal and microstructural means. The processing map reveals a domain of dynamic recrystallization in the range of 730 °C to 850 °C and 10−2 to 1−1 with its peak efficiency of 40 pct at 800 °C and 0.1 s-1 which may be considered as optimum hot-working parameters. The characteristics of dynamic recrystallization are similar to those of static recrystallization regarding the sigmoidal variation of grain size (or hardness) with temperature, although the dynamic recrystallization temperature is much higher. When deformed at 650 °C and 10-3 s-1 texture-induced dynamic recovery occurred, while at strain rates higher than 1 s-1, alpha-zirconium exhibits microstructural instabilities in the form of localized shear bands which are to be avoided in processing.

Flow-Acoustic Performance of Perforated Element Mufflers With Application to Design

A fairly comprehensive computer program incorporating explicit expressions for the four-pole parameters of concentric-tube resonators, plug mufflers, and three-duct cross-flow perforated elements has been used for parametric studies. The parameters considered are hole diameter, the center-to-center distance between consecutive holes (which decides porosity), the incoming mean flow Mach number, the area expansion ratio, the number of partitions of chambers within a given overall shell length, and the relative lengths of these partitions or chambers, all normalized with respect to the exhaust pipe diameter. Transmission loss has been plotted as a function of a normalized frequency parameter. Additionally, the effect of the tail pipe length on insertion loss for an anechoic source has also been studied. These studies have been supplemented by empirical expressions for the normalized static pressure drop for different types of perforated-element mufflers developed from experimental observations.

Automatic decomposition of unstructured meshes employing genetic algorithms for parallel FEM computations

Parallel execution of computational mechanics codes requires efficient mesh-partitioning techniques. These mesh-partitioning techniques divide the mesh into specified number of submeshes of approximately the same size and at the same time, minimise the interface nodes of the submeshes. This paper describes a new mesh partitioning technique, employing Genetic Algorithms. The proposed algorithm operates on the deduced graph (dual or nodal graph) of the given finite element mesh rather than directly on the mesh itself. The algorithm works by first constructing a coarse graph approximation using an automatic graph coarsening method. The coarse graph is partitioned and the results are interpolated onto the original graph to initialise an optimisation of the graph partition problem. In practice, hierarchy of (usually more than two) graphs are used to obtain the final graph partition. The proposed partitioning algorithm is applied to graphs derived from unstructured finite element meshes describing practical engineering problems and also several example graphs related to finite element meshes given in the literature. The test results indicate that the proposed GA based graph partitioning algorithm generates high quality partitions and are superior to spectral and multilevel graph partitioning algorithms.

Memory Architecture Exploration Framework for Cache Based Embedded SoC

Today's feature-rich multimedia products require embedded system solution with complex System-on-Chip (SoC) to meet market expectations of high performance at a low cost and lower energy consumption. The memory architecture of the embedded system strongly influences critical system design objectives like area, power and performance. Hence the embedded system designer performs a complete memory architecture exploration to custom design a memory architecture for a given set of applications. Further, the designer would be interested in multiple optimal design points to address various market segments. However, tight time-to-market constraints enforces short design cycle time. In this paper we address the multi-level multi-objective memory architecture exploration problem through a combination of exhaustive-search based memory exploration at the outer level and a two step based integrated data layout for SPRAM-Cache based architectures at the inner level. We present a two step integrated approach for data layout for SPRAM-Cache based hybrid architectures with the first step as data-partitioning that partitions data between SPRAM and Cache, and the second step is the cache conscious data layout. We formulate the cache-conscious data layout as a graph partitioning problem and show that our approach gives up to 34% improvement over an existing approach and also optimizes the off-chip memory address space. We experimented our approach with 3 embedded multimedia applications and our approach explores several hundred memory configurations for each application, yielding several optimal design points in a few hours of computation on a standard desktop.

Streaming FFT on REDEFINE-v2: An Application-Architecture Design Space Exploration

In this paper we explore an implementation of a high-throughput, streaming application on REDEFINE-v2, which is an enhancement of REDEFINE. REDEFINE is a polymorphic ASIC combining the flexibility of a programmable solution with the execution speed of an ASIC. In REDEFINE Compute Elements are arranged in an 8x8 grid connected via a Network on Chip (NoC) called RECONNECT, to realize the various macrofunctional blocks of an equivalent ASIC. For a 1024-FFT we carry out an application-architecture design space exploration by examining the various characterizations of Compute Elements in terms of the size of the instruction store. We further study the impact by using application specific, vectorized FUs. By setting up different partitions of the FFT algorithm for persistent execution on REDEFINE-v2, we derive the benefits of setting up pipelined execution for higher performance. The impact of the REDEFINE-v2 micro-architecture for any arbitrary N-point FFT (N > 4096) FFT is also analyzed. We report the various algorithm-architecture tradeoffs in terms of area and execution speed with that of an ASIC implementation. In addition we compare the performance gain with respect to a GPP.