Prediction of Queue Waiting Times for Metascheduling on Parallel Batch Systems

Prediction of queue waiting times of jobs submitted to production parallel batch systems is important to provide overall estimates to users and can also help meta-schedulers make scheduling decisions. In this work, we have developed a framework for predicting ranges of queue waiting times for jobs by employing multi-class classification of similar jobs in history. Our hierarchical prediction strategy first predicts the point wait time of a job using dynamic k-Nearest Neighbor (kNN) method. It then performs a multi-class classification using Support Vector Machines (SVMs) among all the classes of the jobs. The probabilities given by the SVM for the class predicted using k-NN and its neighboring classes are used to provide a set of ranges of predicted wait times with probabilities. We have used these predictions and probabilities in a meta-scheduling strategy that distributes jobs to different queues/sites in a multi-queue/grid environment for minimizing wait times of the jobs. Experiments with different production supercomputer job traces show that our prediction strategies can give correct predictions for about 77-87% of the jobs, and also result in about 12% improved accuracy when compared to the next best existing method. Experiments with our meta-scheduling strategy using different production and synthetic job traces for various system sizes, partitioning schemes and different workloads, show that the meta-scheduling strategy gives much improved performance when compared to existing scheduling policies by reducing the overall average queue waiting times of the jobs by about 47%.

Variation of Electrical Resistivity with High Pressure in Ge-Te-Sn Glasses: A Composition Dependent Study

The variation of normalized electrical resistivity in the system of glasses Ge15Te85-xSnx with (1 <= x <= 5) has been studied as a function of high pressure for pressures up to 9.5 GPa. It is found that with the increase in pressure, the resistivity decreases initially and shows an abrupt fall at a particular pressure, indicating the phase transition from semiconductor to near metallic at these pressures, which lie in the range 1.5-2.5 GPa, and then continues being metallic up to 9.5 GPa. This transition pressure is seen to decrease with the increase in the percentage content of tin due to increasing metallicity of tin. The semiconductor to near metallic transition is exactly reversible and may have its origin in a reduction of the band gap due to high pressure.

Effect Of High Pressure On The Electrical Resistivity Of Ge-Te In Glasses

The variation in the electrical resistivity of the chalcogenide glasses Ge15Te85-x has been studied as a function of high pressure for pressures up to 8.5GPa. All the samples studied undergo a semi-conductor to metallic transition in a continuous manner at pressures between 1.5-2.5GPa. The transition pressure at which the samples turn metallic increases with increase in percentage of Indium. This increase is a direct consequence of the increase in network rigidity with the addition of Indium. At a constant pressure of 0.5GPa, the normalized resistivity shows some signature of the existence of the intermediate phase. Samples recovered after a pressure cycle remain amorphous suggesting that the semi-conductor to metallic transition arises from a reduction of the band gap due to pressure or the movement of the Fermi level into the conduction or valence band.

Preprocess and data analysis techniques for affymetrix DNA microarrays using bioconductor: a case study in Alzheimer disease

DNA microarray, or DNA chip, is a technology that allows us to obtain the expression level of many genes in a single experiment. The fact that numerical expression values can be easily obtained gives us the possibility to use multiple statistical techniques of data analysis. In this project microarray data is obtained from Gene Expression Omnibus, the repository of National Center for Biotechnology Information (NCBI). Then, the noise is removed and data is normalized, also we use hypothesis tests to find the most relevant genes that may be involved in a disease and use machine learning methods like KNN, Random Forest or Kmeans. For performing the analysis we use Bioconductor, packages in R for the analysis of biological data, and we conduct a case study in Alzheimer disease. The complete code can be found in https://github.com/alberto-poncelas/ bioc-alzheimer

Technology for Single Cell Protein Analysis in Immunology and Cancer Prognostics

The first chapter of this thesis deals with automating data gathering for single cell microfluidic tests. The programs developed saved significant amounts of time with no loss in accuracy. The technology from this chapter was applied to experiments in both Chapters 4 and 5.

The second chapter describes the use of statistical learning to prognose if an anti-angiogenic drug (Bevacizumab) would successfully treat a glioblastoma multiforme tumor. This was conducted by first measuring protein levels from 92 blood samples using the DNA-encoded antibody library platform. This allowed the measure of 35 different proteins per sample, with comparable sensitivity to ELISA. Two statistical learning models were developed in order to predict whether the treatment would succeed. The first, logistic regression, predicted with 85% accuracy and an AUC of 0.901 using a five protein panel. These five proteins were statistically significant predictors and gave insight into the mechanism behind anti-angiogenic success/failure. The second model, an ensemble model of logistic regression, kNN, and random forest, predicted with a slightly higher accuracy of 87%.

The third chapter details the development of a photocleavable conjugate that multiplexed cell surface detection in microfluidic devices. The method successfully detected streptavidin on coated beads with 92% positive predictive rate. Furthermore, chambers with 0, 1, 2, and 3+ beads were statistically distinguishable. The method was then used to detect CD3 on Jurkat T cells, yielding a positive predictive rate of 49% and false positive rate of 0%.

The fourth chapter talks about the use of measuring T cell polyfunctionality in order to predict whether a patient will succeed an adoptive T cells transfer therapy. In 15 patients, we measured 10 proteins from individual T cells (~300 cells per patient). The polyfunctional strength index was calculated, which was then correlated with the patient's progress free survival (PFS) time. 52 other parameters measured in the single cell test were correlated with the PFS. No statistical correlator has been determined, however, and more data is necessary to reach a conclusion.

Finally, the fifth chapter talks about the interactions between T cells and how that affects their protein secretion. It was observed that T cells in direct contact selectively enhance their protein secretion, in some cases by over 5 fold. This occurred for Granzyme B, Perforin, CCL4, TNFa, and IFNg. IL- 10 was shown to decrease slightly upon contact. This phenomenon held true for T cells from all patients tested (n=8). Using single cell data, the theoretical protein secretion frequency was calculated for two cells and then compared to the observed rate of secretion for both two cells not in contact, and two cells in contact. In over 90% of cases, the theoretical protein secretion rate matched that of two cells not in contact.

模式识别在Eu~(2+)光谱结构研究中的应用

由于Eu~(2+)离子在不同复合氟化物中存在不同的跃迁发射形式，主要有5d → 4f的宽带跃迁，位于365nm-650nm间和4f → 4f的窄带跃迁，中心位置在360nm附近。Eu~(2+)离子的跃迁形式决定于基质的化学组成。本工作就是用多种模式识别方法（KNN，ALKNN，BAYES，LLM，SIMCA和PCA）研究不同复合氟化物基质中Eu~(2+)离子的跃迁发射形式和基质晶体结构之间的关系，找出Eu~(2+)离子产生f → f跃迁其基质构成的一般规律性。收集了90个复合氟化物（AB_mF_n）作为样本集，根据其中Eu~(2+)离子跃迁形式的不同将它们分成两类，一类为具有f → f跃迁的基质45个；另一类为不具有f → f跃迁的基质45个。随机地选用63个基质作为训练集，其余的为验证集。每个基质样本利用其12个晶体结构参数作为描述。由于各参数间差别不大，对原始数据未进行标度化。特征提取是模式识别分析的一个重要步骤，本工作结合变化权重法，BAYES特征量评价法和SIMCA变量相关性评价法的特点，建立了一个以验评价判据式：d(i) = -5.0 + 2.3V(i) + 0.89f(i) + 7.2W(i)根据经验式，选取了变量Z_B/r_(kB)，r_(covA)/r_(covB)和Z_B/r_(covB)，并删除了变量Xσ_A，Xσ_B，r_(covA)。其它变量由于其D值接近，利用穷举法对它们进行选取，结果M，Z'_A和r_(covB)被选中。这样把这6个被选的变量作为对跃迁发射问题最相关的变量进行进一步分析。采用被选的6维变量对训练集样本施行主成份分析，结果表示前三个主成份已可解释原数据信息量的99%以上。所以分别以主成份1-3及主成份1和主成份3作了三维和二维的映射图。结果表示两类基质样本基本上分在不同区域。进一步分别用12维和6维变量对样本系进行了其它几种模式识别分析。所有这些方法对训练集的分类效果都比较理想。采取6维特征时，其正确分类率达79.4-96.8%，这说明与跃迁问题相关的大部分变量已被选入。但是结果显示，各种方法对训练集的分类有一定的差别。我们认为这是由于各种不同的方法对数据结构要求不同引起的。实验证明Bayes线性判别方法对该样本集数据的分类效果最佳。根据Bayes线性差别方法的执行得到了对基质样本分类模式，由此模式讨论了各结构参数对Eu~(2+)离子光谱结构的影响，并对七个未知基质中Eu~(2+)离子的光谱结构进行了计算机预报，结果表示KTbF_4，KBF_4，NaIn_2F_7和KLu_2F_7为具有f → f跃迁发射的基质，而NaCaF_3，MgBeF_4和MgAlF_5为不具有f → f跃迁发射的基质。

Classification of valence changes of trivalent rare earth ions in alkaline earth borates using artificial neural networks

The investigations of classification on the valence changes from RE3+ to RE2+ (RE = Eu, Sm, Yb, Tm) in host compounds of alkaline earth berate were performed using artificial neural networks (ANNs). For comparison, the common methods of pattern recognition, such as SIMCA, KNN, Fisher discriminant analysis and stepwise discriminant analysis were adopted. A learning set consisting of 24 host compounds and a test set consisting of 12 host compounds were characterized by eight crystal structure parameters. These parameters were reduced from 8 to 4 by leaps and bounds algorithm. The recognition rates from 87.5 to 95.8% and prediction capabilities from 75.0 to 91.7% were obtained. The results provided by ANN method were better than that achieved by the other four methods. (C) 1999 Elsevier Science B.V. All rights reserved.

SEARCH FOR THE BASIC REGULARITIES OF THE TRANSITION EMISSION OF EUROPIUM(II) ION IN COMPLEX FLUORIDES USING PATTERN-RECOGNITION

The relationship between structures of complex fluorides and spectral structure of Eu(II) ion in complex fluorides (AB(m)F(n)) is investigated by means of pattern recognition methods, such as KNN, ALKNN, BAYES, LLM, SIMCA and PCA. A learning set consisting of 32 f-f transition emission host compounds and 31 d-f transition emission host compounds and a test set consisting of 27 host compounds were characterized by 12 crystal structural parameters. These parameters, i.e. features, were reduced from 12 to 6 by multiple criteria for the classification of these host compounds as f-f transition emission or d-f transition emission. A recognition rate from 79.4 to 96.8% and prediction capabilities from 85.2 to 92.6% were obtained. According to the above results, the spectral structures of Eu(II) ion in seven unknown host lattices were predicted.

模式识别在有机化合物碳—13核磁共振波谱中的应用

本文介绍了以碳—13 NMR 谱为基础,运用模式识别方法对于取代苯类有机化合物的分类情况。数据源为 CIAC-碳-13数据库。特征选择为简单的机率比率法。模式识别方法为Fisher 意义下的判别函数、KNN 及非线性映射。所得结果比较满意。

Aiding classification of gene expression data with feature selection: a comparative study

C. Shang and Q. Shen. Aiding classification of gene expression data with feature selection: a comparative study. Computational Intelligence Research, 1(1):68-76.

Speaker Normalization Methods for Vowel Cognition: Comparative Analysis Using Neural Network and Nearest Neighbor Classifiers

Intrinsic and extrinsic speaker normalization methods are systematically compared using a neural network (fuzzy ARTMAP) and L1 and L2 K-Nearest Neighbor (K-NN) categorizers trained and tested on disjoint sets of speakers of the Peterson-Barney vowel database. Intrinsic methods include one nonscaled, four psychophysical scales (bark, bark with endcorrection, mel, ERB), and three log scales, each tested on four combinations of F0 , F1, F2, F3. Extrinsic methods include four speaker adaptation schemes, each combined with the 32 intrinsic methods: centroid subtraction across all frequencies (CS), centroid subtraction for each frequency (CSi), linear scale (LS), and linear transformation (LT). ARTMAP and KNN show similar trends, with K-NN performing better, but requiring about ten times as much memory. The optimal intrinsic normalization method is bark scale, or bark with endcorrection, using the differences between all frequencies (Diff All). The order of performance for the extrinsic methods is LT, CSi, LS, and CS, with fuzzy ARTMAP performing best using bark scale with Diff All; and K-NN choosing psychophysical measures for all except CSi.

SVM Training Phase Reduction Using Dataset Feature Filtering for Malware Detection

N-gram analysis is an approach that investigates the structure of a program using bytes, characters, or text strings. A key issue with N-gram analysis is feature selection amidst the explosion of features that occurs when N is increased. The experiments within this paper represent programs as operational code (opcode) density histograms gained through dynamic analysis. A support vector machine is used to create a reference model, which is used to evaluate two methods of feature reduction, which are 'area of intersect' and 'subspace analysis using eigenvectors.' The findings show that the relationships between features are complex and simple statistics filtering approaches do not provide a viable approach. However, eigenvector subspace analysis produces a suitable filter.

ZIL: An Energy-Efficient Indoor Localization System Using ZigBee Radio to Detect WiFi Fingerprints

In existing WiFi-based localization methods, smart mobile devices consume quite a lot of power as WiFi interfaces need to be used for frequent AP scanning during the localization process. In this work, we design an energy-efficient indoor localization system called ZigBee assisted indoor localization (ZIL) based on WiFi fingerprints via ZigBee interference signatures. ZIL uses ZigBee interfaces to collect mixed WiFi signals, which include non-periodic WiFi data and periodic beacon signals. However, WiFi APs cannot be identified from these WiFi signals by ZigBee interfaces directly. To address this issue, we propose a method for detecting WiFi APs to form WiFi fingerprints from the signals collected by ZigBee interfaces. We propose a novel fingerprint matching algorithm to align a pair of fingerprints effectively. To improve the localization accuracy, we design the K-nearest neighbor (KNN) method with three different weighted distances and find that the KNN algorithm with the Manhattan distance performs best. Experiments show that ZIL can achieve the localization accuracy of 87%, which is competitive compared to state-of-the-art WiFi fingerprint-based approaches, and save energy by 68% on average compared to the approach based on WiFi interface.

Electromechanical properties of engineered lead free potassium sodium niobate based materials

K0.5Na0.5NbO3 (KNN), is the most promising lead free material for substituting lead zirconate titanate (PZT) which is still the market leader used for sensors and actuators. To make KNN a real competitor, it is necessary to understand and to improve its properties. This goal is pursued in the present work via different approaches aiming to study KNN intrinsic properties and then to identify appropriate strategies like doping and texturing for designing better KNN materials for an intended application. Hence, polycrystalline KNN ceramics (undoped, non-stoichiometric; NST and doped), high-quality KNN single crystals and textured KNN based ceramics were successfully synthesized and characterized in this work. Polycrystalline undoped, non-stoichiometric (NST) and Mn doped KNN ceramics were prepared by conventional ceramic processing. Structure, microstructure and electrical properties were measured. It was observed that the window for mono-phasic compositions was very narrow for both NST ceramics and Mn doped ceramics. For NST ceramics the variation of A/B ratio influenced the polarization (P-E) hysteresis loop and better piezoelectric and dielectric responses could be found for small stoichiometry deviations (A/B = 0.97). Regarding Mn doping, as compared to undoped KNN which showed leaky polarization (P-E) hysteresis loops, B-site Mn doped ceramics showed a well saturated, less-leaky hysteresis loop and a significant properties improvement. Impedance spectroscopy was used to assess the role of Mn and a relation between charge transport – defects and ferroelectric response in K0.5Na0.5NbO3 (KNN) and Mn doped KNN ceramics could be established. At room temperature the conduction in KNN which is associated with holes transport is suppressed by Mn doping. Hence Mn addition increases the resistivity of the ceramic, which proved to be very helpful for improving the saturation of the P-E loop. At high temperatures the conduction is dominated by the motion of ionized oxygen vacancies whose concentration increases with Mn doping. Single crystals of potassium sodium niobate (KNN) were grown by a modified high temperature flux method. A boron-modified flux was used to obtain the crystals at a relatively low temperature. XRD, EDS and ICP analysis proved the chemical and crystallographic quality of the crystals. The grown KNN crystals exhibit higher dielectric permittivity (29,100) at the tetragonal-to-cubic phase transition temperature, higher remnant polarization (19.4 μC/cm2) and piezoelectric coefficient (160 pC/N) when compared with the standard KNN ceramics. KNN single crystals domain structure was characterized for the first time by piezoforce response microscopy. It could be observed that <001> - oriented potassium sodium niobate (KNN) single crystals reveal a long range ordered domain pattern of parallel 180° domains with zig-zag 90° domains. From the comparison of KNN Single crystals to ceramics, It is argued that the presence in KNN single crystal (and absence in KNN ceramics) of such a long range order specific domain pattern that is its fingerprint accounts for the improved properties of single crystals. These results have broad implications for the expanded use of KNN materials, by establishing a relation between the domain patterns and the dielectric and ferroelectric response of single crystals and ceramics and by indicating ways of achieving maximised properties in KNN materials. Polarized Raman analysis of ferroelectric potassium sodium niobate (K0.5Na0.5)NbO3 (KNN) single crystals was performed. For the first time, an evidence is provided that supports the assignment of KNN single crystals structure to the monoclinic symmetry at room temperature. Intensities of A′, A″ and mixed A′+A″ phonons have been theoretically calculated and compared with the experimental data in dependence of crystal rotation, which allowed the precise determination of the Raman tensor coefficients for (non-leaking) modes in monoclinic KNN. In relation to the previous literature, this study clarifies that assigning monoclinic phase is more suitable than the orthorhombic one. In addition, this study is the basis for non-destructive assessments of domain distribution by Raman spectroscopy in KNN-based lead-free ferroelectrics with complex structures. Searching a deeper understanding of the electrical behaviour of both KNN single crystal and polycrystalline materials for the sake of designing optimized KNN materials, a comparative study at the level of charge transport and point defects was carried out by impedance spectroscopy. KNN single crystals showed lower conductivity than polycrystals from room temperature up to 200 ºC, but above this temperature polycrystalline KNN displays lower conductivity. The low temperature (T < 200 ºC) behaviour reflects the different processing conditions of both ceramics and single crystals, which account for less defects prone to charge transport in the case of single crystals. As temperature increases (T > 200 ºC) single crystals become more conductive than polycrystalline samples, in which grain boundaries act as barriers to charge transport. For even higher temperatures the conductivity difference between both is increased due to the contribution of ionic conduction in single crystals. Indeed the values of activation energy calculated to the high temperature range (T > 300 ºC) were 1.60 and 0.97 eV, confirming the charge transport due to ionic conduction and ionized oxygen vacancies in single crystals and polycrystalline KNN, respectively. It is suggested that single crystals with low defects content and improved electromechanical properties could be a better choice for room temperature applications, though at high temperatures less conductive ceramics may be the choice, depending on the targeted use. Aiming at engineering the properties of KNN polycrystals towards the performance of single crystals, the preparation and properties study of (001) – oriented (K0.5Na0.5)0.98Li0.02NbO3 (KNNL) ceramics obtained by templated grain growth (TGG) using KNN single crystals as templates was undertaken. The choice of KNN single crystals templates is related with their better properties and to their unique domain structure which were envisaged as a tool for templating better properties in KNN ceramics too. X-ray diffraction analysis revealed for the templated ceramics a monoclinic structure at room temperature and a Lotgering factor (f) of 40% which confirmed texture development. These textured ceramics exhibit a long range ordered domain pattern consisting of 90º and 180º domains, similar to the one observed in the single crystals. Enhanced dielectric (13017 at TC), ferroelectric (2Pr = 42.8 μC/cm2) and piezoelectric (d33 = 280 pC/N) properties are observed for textured KNNL ceramics as compared to the randomly oriented ones. This behaviour is suggested to be due to the long range ordered domain patterns observed in the textured ceramics. The obtained results as compared with the data previously reported on texture KNN based ceramics confirm that superior properties were found due to ordered repeated domain pattern. This study provides an useful approach towards properties improvement of KNN-based piezoelectric ceramics. Overall, the present results bring a significant contribution to the pool of knowledge on the properties of sodium potassium niobate materials: a relation between the domain patterns and di-, ferro-, and piezo-electric response of single crystals and ceramics was demonstrated and ways of engineering maximised properties in KNN materials, for example by texturing were established. This contribution is envisaged to have broad implications for the expanded use of KNN over the alternative lead-based materials.

A lightweight indoor localization model based on motley-keenan and COST

This paper presents a novel approach to WLAN propagation models for use in indoor localization. The major goal of this work is to eliminate the need for in situ data collection to generate the Fingerprinting map, instead, it is generated by using analytical propagation models such as: COST Multi-Wall, COST 231 average wall and Motley- Keenan. As Location Estimation Algorithms kNN (K-Nearest Neighbour) and WkNN (Weighted K-Nearest Neighbour) were used to determine the accuracy of the proposed technique. This work is based on analytical and measurement tools to determine which path loss propagation models are better for location estimation applications, based on Receive Signal Strength Indicator (RSSI).This study presents different proposals for choosing the most appropriate values for the models parameters, like obstacles attenuation and coefficients. Some adjustments to these models, particularly to Motley-Keenan, considering the thickness of walls, are proposed. The best found solution is based on the adjusted Motley-Keenan and COST models that allows to obtain the propagation loss estimation for several environments.Results obtained from two testing scenarios showed the reliability of the adjustments, providing smaller errors in the measured values values in comparison with the predicted values.