Clustering. Cluster Analysis of Microarray Data. Microarray Data for Clustering. Data for Clustering

Size: px
Start display at page:

Download "Clustering. Cluster Analysis of Microarray Data. Microarray Data for Clustering. Data for Clustering"

Transcription

1 Clustering Cluster Analysis of Microarray Data 4/3/009 Copyright 009 Dan Nettleton Group obects that are siilar to one another together in a cluster. Separate obects that are dissiilar fro each other into different clusters. The siilarity or dissiilarity of two obects is deterined by coparing the obects with respect to one or ore attributes that can be easured for each obect. Data for Clustering Microarray Data for Clustering attribute obect n attribute tie points obect n estiated expression levels genes 4 Microarray Data for Clustering attribute tissue types obect n estiated expression levels genes 5 Microarray Data for Clustering treatent attribute conditions obect n estiated expression levels genes 6

2 Microarray Data for Clustering attribute genes obect n estiated expression levels saples 7 Clustering: An Exaple Experient Researchers were interested in studying gene expression patterns in developing soybean seeds. Seeds were harvested fro soybean plants at 5, 30, 40, 45, and 50 days after flowering (daf). One RNA saple was obtained for each level of daf. 8 An Exaple Experient (continued) Diagra Illustrating the Experiental Design Each of the 5 saples was easured on two two-color cdna icroarray slides using a loop design. The entire process we repeated on a second occasion to obtain a total of two independent biological replications. Rep Rep The daf eans estiated for each gene fro a ixed linear odel analysis provide a useful suary of the data for cluster analysis. 400 genes exhibited significant evidence of differential expression across tie (p-value<0.0, FDR=3.%). We will focus on clustering their estiated ean profiles. Noralized Data for One Exaple Gene Estiated Means + or SE Noralized Data for One Exaple Gene Estiated Means + or SE Noralized Log Signal Noralized Log Signal daf daf daf daf

3 We build clusters based on the ost significant genes rather than on all genes because... Estiated Mean Profiles for Top 36 Genes Much of the variation in expression is noise rather than biological signal, and we would rather not build clusters on the basis of noise. Soe clustering algoriths will becoe coputationally expensive if there are a large nuber of obects (gene expression profiles in this case) to cluster. 3 4 Dissiilarity Measures Parallel Coordinate Plots When clustering obects, we try to put siilar obects in the sae cluster and dissiilar obects in different clusters. Scatterplot Parallel Coordinate Plot We ust define what we ean by dissiilar. There are any choices. Let x and y denote diensional obects: x Value x=(x, x,..., x ) y=(y,y,..., y ) e.g., estiated eans at =5 five tie points for a given gene. 5 x Coordinate 6 These are parallel coordinate plots that each show one point in 5-diensional space. Euclidean Distance d E (x,y) = x - y = (x - y ) = -Correlation d cor (x,y) = - r xy = - = = (x - x) (x - x)(y - y) = (y - y) 7 8 3

4 Euclidean Distance -Correlation Dissiilarity Scatterplot Parallel Coordinate Plot d E (black,green) x d E (black,red) d E (red,green) Value The black and green obects are close together and far fro the red obect. x Coordinate 9 0 Relationship between Euclidean Distance and -Correlation Dissiilarity Let x~ = Let y~ = ~ x - ~ y = = = x - x s s x y - y y = ~ x = and let x~ and let y~ ( ~ x - ~ y ) = + ( - ) - r y = xy = (x ~,x ~,...,x ~ = (y ~,y ~,...,y ~ ~ - = ~ ~ y x = ). ( ~ x + ~ y - ~ x ~ y ) ). Thus Euclidean distance for standardized obects is proportional to the square root of the -correlation dissiilarity. We will standardize our ean profiles so that each profile has ean 0 and standard deviation (i.e., we will convert each x to ~ x). We will cluster based on the Euclidean distance between standardized profiles. Original ean profiles with siilar patterns are close to one another using this approach. Clustering ethods are often divided into two ain groups.. Partitioning ethods that attept to optially separate n obects into K clusters.. Hierarchical ethods that produce a nested sequence of clusters. Soe Partitioning Methods. K-Means. K-Medoids 3. Self-Organizing Maps (SOM) (Kohonen, 990; Toayo, P. et al., 998) 3 4 4

5 K Medoids Clustering Exaple of K Medoids Clustering 0. Choose K of the n obects to represent K cluster centers (a.k.a., edoids).. Given a current set of K edoids, assign each obect to the nearest edoid to produce an assignent of obects to K clusters.. For a given assignent of obects to K clusters, find the new edoid for each cluster by finding the obect in the cluster that is the closest on average to all other obects in its cluster. 3. Repeat steps and until the cluster assignents do not change. 5 6 Start with K Medoids Assign Each Point to Closest Medoid 7 8 Assign Each Point to Closest Medoid Assign Each Point to Closest Medoid

6 Assign Each Point to Closest Medoid Find New Medoid for Each Cluster New edoids have sallest average distance to other points in their cluster. 3 3 Reassign Each Point to Closest Medoid Reassign Each Point to Closest Medoid Find New Medoid for Each Cluster Reassign Each Point to Closest Medoid No reassignent is needed, so the procedure stops

7 Cluster of 3 fro K-Medoids Algorith Applied to the Top 400 Genes fro Public the Cluster Two-Color of 3 Array Data Cluster of 3 fro K-Medoids Algorith Applied to the Top 400 Genes fro Public the Cluster Two-Color of 3 Array Data Standardized Mean Profile Standardized Mean Profile DAF 37 DAF 38 Cluster 3 of 3 fro K-Medoids Algorith Applied to the Top 400 Genes fro Public the Cluster Two-Color 3 of 3 Array Data Choosing the Nuber of Clusters K Standardized Mean Profile Choose K that axiizes the average silhouette width. Rousseeuw, P.J. (987). Journal of Coputational and Applied Matheatics, 0, Kaufan, L. and Rousseeuw, P.J. (990). Finding Groups in Data: An Introduction to Cluster Analysis. Wiley, New York. Choose K according to the gap statistic. DAF 39 Tibshirani, R., Walther, G., Hastie, T. (00). Journal of the Royal Statistics Society, Series B-Statistical Methodology, 63, Silhouette Width Silhouette Width = (B-W)/ax(B,W) The silhouette width of an obect is (B-W)/ax(B,W) where W=average distance of the obect to all other obects within its cluster and B=average distance of the obect to all obects in its nearest neighboring cluster. Values near indicate that an obect is near the center of a tight cluster. Values near 0 indicate that an obect is between clusters. Negative values indicate that an obect ay be in the wrong cluster. The silhouette width will be between - and

8 Silhouette Width = (B-W)/ax(B,W) For a Given K Copute Silhouette Width for Each Point The silhouette widths of clustered obects can be averaged. A clustering with a high average silhouette width is preferred. For a given ethod of clustering, we ay wish to choose the value of K that axiizes the average silhouette width. Find W=average distance fro point to all others within its cluster. Find B=average distance fro point to all others in its nearest neighboring cluster. 43 Silhouette width is B-W ax(b,w) 44 Choice of K Silhouette width is coputed for all points and averaged. Average Silhouette Width vs. K for the K-Medoids Algorith Applied to the Top 400 Genes fro the Two-Color Array Data K with largest average silhouette width is preferred. K=3: Average Silhouette Width=0.640 K=: Average Silhouette Width=0.646 Slight preference for K= in this case Cluster of fro K-Medoids Algorith Applied to the Top 400 Genes fro the Two-Color Array Data Cluster of fro K-Medoids Algorith Applied to the Top 400 Genes fro the Two-Color Array Data Standardized Mean Profile Standardized Mean Profile DAF 47 DAF 48 8

9 Gap Statistic Gap Statistic (continued) For a given clustering of n obects x,...,x n; the distance d(x i,x ) between obects x i and x is called a within-cluster distance if x i and x are within the sae cluster. Let D r = the su of all within-cluster distances in the r th cluster, and let n r denote the nuber of obects in the r th cluster. For a given clustering of n obects into k clusters, k let W k = Σ r= D r / n r. 49 For a given clustering ethod, copute log W, log W,..., log W K. Let in denote the iniu of the th coponent of all n obects clustered. Let ax denote the axiu of the th coponent of all n obects to be clustered. Generate n rando obects uniforly distributed on the diensional rectangle [in,ax ] x... x [in,ax ]. 50 Gap Statistic (continued) Estiate of Best K Using the Gap Statistic Using the rando unifor data, copute log W *, log W *,..., log W K *. Randoly generate new unifor data ultiple ties (0 or ore) and use the results to obtain log W *, log W *,..., log W K * and S, S,...,S K ; the averages and standard deviations of the siulated log W values. Let G(k) = log W k * log W k. An approxiate standard error for G(k) is S k +/N where N denotes the nuber of randoly generated data sets. An estiate of the best K is given by K ^ = in { k : G(k) G(k+) S k+ +/N }. 5 5 Siple Exaple Data Revisited Gap Analysis for the Siple Exaple (N=000) log W k * and log W k vs. k G(k)=log W k * log W k vs. k (+ or standard error) log W k G(k) k=nuber of Clusters k=nuber of Clusters

10 The Gap Statistic Suggests K=3 Clusters The Gap Statistic Suggests K=3 Clusters G(3) >= G(4)-SE G() less than G(3)-SE G() less than G()-SE Gap Analysis for Two-Color Array Data (N=00) Gap Analysis for Two-Color Array Data (N=00) log W k * and log W k vs. k G(k)=log W k * log W k vs. k (+ or standard error) zooed in version of previous plot Gap Analysis Estiates K= Clusters log W k G(k) G(k) k=nuber of Clusters k=nuber of Clusters 57 k=nuber of Clusters

11

12 67 68 Plot of Cluster Medoids Principal Coponents Principal coponents can be useful for providing low-diensional views of high-diensional data.... nuber of variables Principal Coponents (continued) Each principal coponent of a data set is a variable obtained by taking a linear cobination of the original variables in the data set. variable or attribute x x... x x Data x. Matrix. X =... x n or Data Set x n..... x n.. n observation or obect nuber of observations 7 A linear cobination of variables x, x,..., x is given by c x + c x c x. For the purpose of constructing principal coponents, the vector of coefficients is restricted to have unit length, i.e., c + c c =. 7

13 Principal Coponents (continued) The Siple Data Exaple The first principal coponent is the linear cobination of the variables that has axiu variation across the observations in the data set. The th principal coponent is the linear cobination of the variables that has axiu variation across the observations in the data set subect to the constraint that the vector of coefficients be orthogonal to coefficient vectors for principal coponents,..., -. x 73 x 74 The First Principal Coponent Axis The First Principal Coponents x x st PC for this point is signed distance between its proection onto the st PC axis and the origin. x 75 x 76 The Second Principal Coponent Axis The Second Principal Coponent x x nd PC for this point is signed distance between its proection onto the nd PC axis and the origin. x 77 x 78 3

14 Plot of PC vs. PC Copare the PC plot to the plot of the original data below. PC x Because there are only two variables here, the plot of PC vs. PC is ust a rotation of the original plot. PC 79 x 80 There is ore to be gained when the nuber of variables is greater than. Proection of Two-Color Array Data with -Medoid Clustering Consider the principal coponents for the 400 significant genes fro our two-color icroarray experient. Our data atrix has n=400 rows and =5 coluns. We have looked at this data using parallel coordinate plots. What would it look like if we proected the data points to -diensions? PC a= b= c=3 d=4 e=5 f=6 g=7 h=8 i=9 =0 k= 8 PC 8 Proection of Two-Color Array Data with -Medoid Clustering PC 3 a= b= c=3 d=4 e=5 f=6 g=7 h=8 i=9 =0 k= PC PC

15 Proection of Two-Color Array Data with -Medoid Clustering Hierarchical Clustering Methods PC 3 a= b= c=3 d=4 e=5 f=6 g=7 h=8 i=9 =0 k= Hierarchical clustering ethods build a nested sequence of clusters that can be displayed using a dendrogra. We will begin with soe siple illustrations and then ove on to a ore general discussion. PC The Siple Exaple Data with Observation Nubers Dendrogra for the Siple Exaple Data root node Tree Structure a parent node x nodes daughter nodes (daughter nodes with sae parent are sister nodes) x 87 terinal nodes or leaves corresponding to obects 88 A Hierarchical Clustering of the Siple Exaple Data Dendrogra for the Siple Exaple Data x Scatterplot of Data Dendrogra The height of a node represents the dissiilarity between the two clusters erged together at the node. x clusters within clusters within clusters These two clusters have a dissiilarity of about.75. 5

16 The appearance of a dendrogra is not unique. Dendrogra The appearance for of the a dendrogra Siple Exaple is not unique. Data Any two sister nodes could trade places without changing the eaning of the dendrogra. By convention, R dendrogras show the lower sister node on the left. Ties are broken by observation nuber. Thus 4 next to 7 does not iply that these obects are siilar. 9 e.g., 3 is to the left of 4 9 The appearance of a dendrogra is not unique. Cutting the tree at a given height will correspond to a partitioning of the data into k clusters. The lengths of the branches leading to terinal nodes have no particular eaning in R dendrogras. k= Clusters Cutting the tree at a given height will correspond to a partitioning of the data into k clusters. Cutting the tree at a given height will correspond to a partitioning of the data into k clusters. k=3 Clusters k=4 Clusters

17 Cutting the tree at a given height will correspond to a partitioning of the data into k clusters. Aggloerative (Botto-Up) Hierarchical Clustering Define a easure of distance between any two clusters. (An individual obect is considered a cluster of size one.) Find the two nearest clusters and erge the together to for a new cluster. k=0 Clusters Repeat until all obects have been erged into a single cluster Coon Measures of Between-Cluster Distance Coon Measures of Between-Cluster Distance Single Linkage a.k.a. Nearest Neighbor: the distance between any two clusters A and B is the iniu of all distances fro an obect in cluster A to an obect in cluster B. Coplete Linkage a.k.a Farthest Neighbor: the distance between any two clusters A and B is the axiu of all distances fro an obect in cluster A to an obect in cluster B. Average Linkage: the distance between any two clusters A and B is the average of all distances fro an obect in cluster A to an obect in cluster B. Centroid Linkage: the distance between any two clusters A and B is the distance between the centroids of cluster A and B. (The centroid of a cluster is the coponentwise average of the obects in a cluster.) x Aggloerative Clustering Using Average Linkage for the Siple Exaple Data Set Scatterplot of Data x N L J F Dendrogra P E D O K H A C M I G B 0 Aggloerative Clustering Using Average Linkage for the Siple Exaple Data Set P A. - N L J F O K H E D A C B. 9-0 C. 3-4 D. 5-6 E. 7-(5,6) F. 3-4 M G. - I G H. (,)-(3,4) B I. (9,0)-(,) etc

18 Aggloerative Clustering Using Single Linkage for the Siple Exaple Data Set Aggloerative Clustering Using Coplete Linkage for the Siple Exaple Data Set Aggloerative Clustering Using Centroid Linkage for the Siple Exaple Data Set Centroid linkage is not onotone in the sense that later cluster erges can involve clusters that are ore siilar to each other than earlier erges. Aggloerative Clustering Using Centroid Linkage for the Siple Exaple Data Set The erge between 4 and (,,3,5) creates a cluster whose centroid is closer to the (6,7) centroid than 4 was to the centroid of (,,3,5) Aggloerative Clustering Using Single Linkage for the Two-Color Microarray Data Set Aggloerative Clustering Using Coplete Linkage for the Two-Color Microarray Data Set

19 Aggloerative Clustering Using Average Linkage for the Two-Color Microarray Data Set Aggloerative Clustering Using Centroid Linkage for the Two-Color Microarray Data Set 09 0 Which Between-Cluster Distance is Best? Depends, of course, on what is eant by best. Single linkage tends to produce long stringy clusters.. Conduct aggloerative hierarchical clustering for this data using Euclidean distance and coplete linkage.. Display your results using a dendrogra. 3. Identify the k= clustering using your results. Coplete linkage produces copact spherical clusters but ight result in soe obects that are closer to obects in clusters other than their own. (See next exaple.) Average linkage is a coproise between single and coplete linkage. Centroid linkage is not onotone. Results of Coplete-Linkage Clustering Divisive (Top-Down) Hierarchical Clustering Results for k= Clusters Start with all data in one cluster and divide it into two clusters (using, e.g., -eans or -edoids clustering). At each subsequent step, choose one of the existing clusters and divide it into two clusters. Repeat until there are n clusters each containing a single obect

20 Potential Proble with Divisive Clustering 5 5 Macnaughton-Sith et al. (965). Start with obects in one cluster A.. Find the obect with the largest average dissiilarity to all other obects in A and ove that obect to a new cluster B. 3. Find the obect in cluster A whose average dissiilarity to other obects in cluster A inus its average dissiilarity to obects in cluster B is axiu. If this difference is positive, ove the obect to cluster B. 4. Repeat step 3 until no obects satisfying 3 are found. 5. Repeat steps through 4 to one of the existing clusters (e.g., the one with the largest average within-cluster dissiilarity) until n clusters of obect each are obtained. 6 Macnaughton-Sith Divisive Clustering Macnaughton-Sith Divisive Clustering A B A B Macnaughton-Sith Divisive Clustering Macnaughton-Sith Divisive Clustering A B A B B

21 Macnaughton-Sith Divisive Clustering Dendrogra for the Macnaughton-Sith Approach A B B 5 Next continue to split each of these clusters until each obect is in a cluster by itself. Aggloerative vs. Divisive Clustering Divisive clustering has not been studied as extensively as aggloerative clustering. Divisive clustering ay be preferred if only a sall nuber of large clusters is desired. Aggloerative clustering ay be preferred if a large nuber of sall clusters is desired. 3

Cluster Analysis for Microarray Data

Cluster Analysis for Microarray Data Cluster Analysis for Microarray Data Seventh International Long Oligonucleotide Microarray Workshop Tucson, Arizona January 7-12, 2007 Dan Nettleton IOWA STATE UNIVERSITY 1 Clustering Group objects that

More information

Detection of Outliers and Reduction of their Undesirable Effects for Improving the Accuracy of K-means Clustering Algorithm

Detection of Outliers and Reduction of their Undesirable Effects for Improving the Accuracy of K-means Clustering Algorithm Detection of Outliers and Reduction of their Undesirable Effects for Iproving the Accuracy of K-eans Clustering Algorith Bahan Askari Departent of Coputer Science and Research Branch, Islaic Azad University,

More information

Short Papers. Location- and Density-Based Hierarchical Clustering Using Similarity Analysis 1 INTRODUCTION

Short Papers. Location- and Density-Based Hierarchical Clustering Using Similarity Analysis 1 INTRODUCTION IEEE TRASACTIOS O PATTER AALYSIS AD MACHIE ITELLIGECE, VOL. 20, O. 9, SEPTEMBER 1998 1011 Short Papers Location- and Density-Based Hierarchical Clustering Using Siilarity Analysis Peter Bacsy and arendra

More information

Geo-activity Recommendations by using Improved Feature Combination

Geo-activity Recommendations by using Improved Feature Combination Geo-activity Recoendations by using Iproved Feature Cobination Masoud Sattari Middle East Technical University Ankara, Turkey e76326@ceng.etu.edu.tr Murat Manguoglu Middle East Technical University Ankara,

More information

Region Segmentation Region Segmentation

Region Segmentation Region Segmentation /7/ egion Segentation Lecture-7 Chapter 3, Fundaentals of Coputer Vision Alper Yilaz,, Mubarak Shah, Fall UCF egion Segentation Alper Yilaz,, Mubarak Shah, Fall UCF /7/ Laer epresentation Applications

More information

Solving the Damage Localization Problem in Structural Health Monitoring Using Techniques in Pattern Classification

Solving the Damage Localization Problem in Structural Health Monitoring Using Techniques in Pattern Classification Solving the Daage Localization Proble in Structural Health Monitoring Using Techniques in Pattern Classification CS 9 Final Project Due Dec. 4, 007 Hae Young Noh, Allen Cheung, Daxia Ge Introduction Structural

More information

Identifying Converging Pairs of Nodes on a Budget

Identifying Converging Pairs of Nodes on a Budget Identifying Converging Pairs of Nodes on a Budget Konstantina Lazaridou Departent of Inforatics Aristotle University, Thessaloniki, Greece konlaznik@csd.auth.gr Evaggelia Pitoura Coputer Science and Engineering

More information

Summary. Reconstruction of data from non-uniformly spaced samples

Summary. Reconstruction of data from non-uniformly spaced samples Is there always extra bandwidth in non-unifor spatial sapling? Ralf Ferber* and Massiiliano Vassallo, WesternGeco London Technology Center; Jon-Fredrik Hopperstad and Ali Özbek, Schluberger Cabridge Research

More information

Verdict Accuracy of Quick Reduct Algorithm using Clustering, Classification Techniques for Gene Expression Data

Verdict Accuracy of Quick Reduct Algorithm using Clustering, Classification Techniques for Gene Expression Data Verdict Accuracy of Quick Reduct Algorith using Clustering, Classification Techniques for Gene Expression Data T.Chandrasekhar 1, K.Thangavel 2 and E.N.Sathishkuar 3 1 Departent of Coputer Science, eriyar

More information

Reconstruction of Time Series using Optimal Ordering of ICA Components

Reconstruction of Time Series using Optimal Ordering of ICA Components Reconstruction of Tie Series using Optial Ordering of ICA Coponents Ar Goneid and Abear Kael Departent of Coputer Science & Engineering, The Aerican University in Cairo, Cairo, Egypt e-ail: goneid@aucegypt.edu

More information

Distributed Multicast Tree Construction in Wireless Sensor Networks

Distributed Multicast Tree Construction in Wireless Sensor Networks Distributed Multicast Tree Construction in Wireless Sensor Networks Hongyu Gong, Luoyi Fu, Xinzhe Fu, Lutian Zhao 3, Kainan Wang, and Xinbing Wang Dept. of Electronic Engineering, Shanghai Jiao Tong University,

More information

Genetic-Based EM Algorithm for Learning Gaussian Mixture Models

Genetic-Based EM Algorithm for Learning Gaussian Mixture Models 1344 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 27, NO. 8, AUGUST 2005 Genetic-Based EM Algorith for Learning Gaussian Mixture Models Franz Pernkopf, Meber, IEEE, and Djael Bouchaffra,

More information

On the Computation and Application of Prototype Point Patterns

On the Computation and Application of Prototype Point Patterns On the Coputation and Application of Prototype Point Patterns Katherine E. Tranbarger Freier 1 and Frederic Paik Schoenberg 2 Abstract This work addresses coputational probles related to the ipleentation

More information

A Trajectory Splitting Model for Efficient Spatio-Temporal Indexing

A Trajectory Splitting Model for Efficient Spatio-Temporal Indexing A Trajectory Splitting Model for Efficient Spatio-Teporal Indexing Slobodan Rasetic Jörg Sander Jaes Elding Mario A. Nasciento Departent of Coputing Science University of Alberta Edonton, Alberta, Canada

More information

A Novel Fast Constructive Algorithm for Neural Classifier

A Novel Fast Constructive Algorithm for Neural Classifier A Novel Fast Constructive Algorith for Neural Classifier Xudong Jiang Centre for Signal Processing, School of Electrical and Electronic Engineering Nanyang Technological University Nanyang Avenue, Singapore

More information

Novel Image Representation and Description Technique using Density Histogram of Feature Points

Novel Image Representation and Description Technique using Density Histogram of Feature Points Novel Iage Representation and Description Technique using Density Histogra of Feature Points Keneilwe ZUVA Departent of Coputer Science, University of Botswana, P/Bag 00704 UB, Gaborone, Botswana and Tranos

More information

Efficient Estimation of Inclusion Coefficient using HyperLogLog Sketches

Efficient Estimation of Inclusion Coefficient using HyperLogLog Sketches Efficient Estiation of Inclusion Coefficient using HyperLogLog Sketches Azade Nazi, Bolin Ding, Vivek Narasayya, Surajit Chaudhuri Microsoft Research {aznazi, bolind, viveknar, surajitc}@icrosoft.co ABSTRACT

More information

Structural Balance in Networks. An Optimizational Approach. Andrej Mrvar. Faculty of Social Sciences. University of Ljubljana. Kardeljeva pl.

Structural Balance in Networks. An Optimizational Approach. Andrej Mrvar. Faculty of Social Sciences. University of Ljubljana. Kardeljeva pl. Structural Balance in Networks An Optiizational Approach Andrej Mrvar Faculty of Social Sciences University of Ljubljana Kardeljeva pl. 5 61109 Ljubljana March 23 1994 Contents 1 Balanced and clusterable

More information

Brian Noguchi CS 229 (Fall 05) Project Final Writeup A Hierarchical Application of ICA-based Feature Extraction to Image Classification Brian Noguchi

Brian Noguchi CS 229 (Fall 05) Project Final Writeup A Hierarchical Application of ICA-based Feature Extraction to Image Classification Brian Noguchi A Hierarchical Application of ICA-based Feature Etraction to Iage Classification Introduction Iage classification poses one of the greatest challenges in the achine vision and achine learning counities.

More information

Automatic Graph Drawing Algorithms

Automatic Graph Drawing Algorithms Autoatic Graph Drawing Algoriths Susan Si sisuz@turing.utoronto.ca Deceber 7, 996. Ebeddings of graphs have been of interest to theoreticians for soe tie, in particular those of planar graphs and graphs

More information

Efficient Learning of Generalized Linear and Single Index Models with Isotonic Regression

Efficient Learning of Generalized Linear and Single Index Models with Isotonic Regression Efficient Learning of Generalized Linear and Single Index Models with Isotonic Regression Sha M. Kakade Microsoft Research and Wharton, U Penn skakade@icrosoft.co Varun Kanade SEAS, Harvard University

More information

Image Filter Using with Gaussian Curvature and Total Variation Model

Image Filter Using with Gaussian Curvature and Total Variation Model IJECT Vo l. 7, Is s u e 3, Ju l y - Se p t 016 ISSN : 30-7109 (Online) ISSN : 30-9543 (Print) Iage Using with Gaussian Curvature and Total Variation Model 1 Deepak Kuar Gour, Sanjay Kuar Shara 1, Dept.

More information

COMPUTER GENERATED HOLOGRAMS Optical Sciences 627 W.J. Dallas (Monday, August 23, 2004, 12:38 PM) PART III: CHAPTER ONE DIFFUSERS FOR CGH S

COMPUTER GENERATED HOLOGRAMS Optical Sciences 627 W.J. Dallas (Monday, August 23, 2004, 12:38 PM) PART III: CHAPTER ONE DIFFUSERS FOR CGH S COPUTER GEERATED HOLOGRAS Optical Sciences 67 W.J. Dallas (onday, August 3, 004, 1:38 P) PART III: CHAPTER OE DIFFUSERS FOR CGH S Part III: Chapter One Page 1 of 8 Introduction Hologras for display purposes

More information

Using Imperialist Competitive Algorithm in Optimization of Nonlinear Multiple Responses

Using Imperialist Competitive Algorithm in Optimization of Nonlinear Multiple Responses International Journal of Industrial Engineering & Production Research Septeber 03, Volue 4, Nuber 3 pp. 9-35 ISSN: 008-4889 http://ijiepr.iust.ac.ir/ Using Iperialist Copetitive Algorith in Optiization

More information

Evaluation of a multi-frame blind deconvolution algorithm using Cramér-Rao bounds

Evaluation of a multi-frame blind deconvolution algorithm using Cramér-Rao bounds Evaluation of a ulti-frae blind deconvolution algorith using Craér-Rao bounds Charles C. Beckner, Jr. Air Force Research Laboratory, 3550 Aberdeen Ave SE, Kirtland AFB, New Mexico, USA 87117-5776 Charles

More information

EE 364B Convex Optimization An ADMM Solution to the Sparse Coding Problem. Sonia Bhaskar, Will Zou Final Project Spring 2011

EE 364B Convex Optimization An ADMM Solution to the Sparse Coding Problem. Sonia Bhaskar, Will Zou Final Project Spring 2011 EE 364B Convex Optiization An ADMM Solution to the Sparse Coding Proble Sonia Bhaskar, Will Zou Final Project Spring 20 I. INTRODUCTION For our project, we apply the ethod of the alternating direction

More information

TensorFlow and Keras-based Convolutional Neural Network in CAT Image Recognition Ang LI 1,*, Yi-xiang LI 2 and Xue-hui LI 3

TensorFlow and Keras-based Convolutional Neural Network in CAT Image Recognition Ang LI 1,*, Yi-xiang LI 2 and Xue-hui LI 3 2017 2nd International Conference on Coputational Modeling, Siulation and Applied Matheatics (CMSAM 2017) ISBN: 978-1-60595-499-8 TensorFlow and Keras-based Convolutional Neural Network in CAT Iage Recognition

More information

Weeks 1 3 Weeks 4 6 Unit/Topic Number and Operations in Base 10

Weeks 1 3 Weeks 4 6 Unit/Topic Number and Operations in Base 10 Weeks 1 3 Weeks 4 6 Unit/Topic Nuber and Operations in Base 10 FLOYD COUNTY SCHOOLS CURRICULUM RESOURCES Building a Better Future for Every Child - Every Day! Suer 2013 Subject Content: Math Grade 3rd

More information

A Fast Multi-Objective Genetic Algorithm for Hardware-Software Partitioning In Embedded System Design

A Fast Multi-Objective Genetic Algorithm for Hardware-Software Partitioning In Embedded System Design A Fast Multi-Obective Genetic Algorith for Hardware-Software Partitioning In Ebedded Syste Design 1 M.Jagadeeswari, 2 M.C.Bhuvaneswari 1 Research Scholar, P.S.G College of Technology, Coibatore, India

More information

Efficient Constraint Evaluation Algorithms for Hierarchical Next-Best-View Planning

Efficient Constraint Evaluation Algorithms for Hierarchical Next-Best-View Planning Efficient Constraint Evaluation Algoriths for Hierarchical Next-Best-View Planning Ko-Li Low National University of Singapore lowl@cop.nus.edu.sg Anselo Lastra University of North Carolina at Chapel Hill

More information

The Internal Conflict of a Belief Function

The Internal Conflict of a Belief Function The Internal Conflict of a Belief Function Johan Schubert Abstract In this paper we define and derive an internal conflict of a belief function We decopose the belief function in question into a set of

More information

Colorado School of Mines. Computer Vision. Professor William Hoff Dept of Electrical Engineering &Computer Science.

Colorado School of Mines. Computer Vision. Professor William Hoff Dept of Electrical Engineering &Computer Science. Professor Willia Hoff Dept of Electrical Engineering &Coputer Science http://inside.ines.edu/~whoff/ 1 Caera Calibration 2 Caera Calibration Needed for ost achine vision and photograetry tasks (object

More information

Research on a Kind of QoS-Sensitive Semantic Web Services Composition Method Based on Genetic Algorithm

Research on a Kind of QoS-Sensitive Semantic Web Services Composition Method Based on Genetic Algorithm roceedings of the 7th International Conference on Innovation & Manageent 893 Research on a Kind of QoS-Sensitive Seantic Web Services Coposition Method Based on Genetic Algorith Cao Hongjiang, Nie Guihua,

More information

IN many interactive multimedia streaming applications, such

IN many interactive multimedia streaming applications, such 840 IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 8, NO. 5, MAY 06 A Geoetric Approach to Server Selection for Interactive Video Streaing Yaochen Hu, Student Meber, IEEE, DiNiu, Meber, IEEE, and Zongpeng Li, Senior

More information

KS3 Maths Progress Delta 2-year Scheme of Work

KS3 Maths Progress Delta 2-year Scheme of Work KS3 Maths Progress Delta 2-year Schee of Work See the spreadsheet for the detail. Essentially this is an express route through our higher-ability KS3 Maths Progress student books Delta 1, 2 and 3 as follows:

More information

TALLINN UNIVERSITY OF TECHNOLOGY, INSTITUTE OF PHYSICS 17. FRESNEL DIFFRACTION ON A ROUND APERTURE

TALLINN UNIVERSITY OF TECHNOLOGY, INSTITUTE OF PHYSICS 17. FRESNEL DIFFRACTION ON A ROUND APERTURE 7. FRESNEL DIFFRACTION ON A ROUND APERTURE. Objective Exaining diffraction pattern on a round aperture, deterining wavelength of light source.. Equipent needed Optical workbench, light source, color filters,

More information

NON-RIGID OBJECT TRACKING: A PREDICTIVE VECTORIAL MODEL APPROACH

NON-RIGID OBJECT TRACKING: A PREDICTIVE VECTORIAL MODEL APPROACH NON-RIGID OBJECT TRACKING: A PREDICTIVE VECTORIAL MODEL APPROACH V. Atienza; J.M. Valiente and G. Andreu Departaento de Ingeniería de Sisteas, Coputadores y Autoática Universidad Politécnica de Valencia.

More information

A Broadband Spectrum Sensing Algorithm in TDCS Based on ICoSaMP Reconstruction

A Broadband Spectrum Sensing Algorithm in TDCS Based on ICoSaMP Reconstruction MATEC Web of Conferences 73, 03073(08) https://doi.org/0.05/atecconf/087303073 SMIMA 08 A roadband Spectru Sensing Algorith in TDCS ased on I Reconstruction Liu Yang, Ren Qinghua, Xu ingzheng and Li Xiazhao

More information

ELEVATION SURFACE INTERPOLATION OF POINT DATA USING DIFFERENT TECHNIQUES A GIS APPROACH

ELEVATION SURFACE INTERPOLATION OF POINT DATA USING DIFFERENT TECHNIQUES A GIS APPROACH ELEVATION SURFACE INTERPOLATION OF POINT DATA USING DIFFERENT TECHNIQUES A GIS APPROACH Kulapraote Prathuchai Geoinforatics Center, Asian Institute of Technology, 58 Moo9, Klong Luang, Pathuthani, Thailand.

More information

2013 IEEE Conference on Computer Vision and Pattern Recognition. Compressed Hashing

2013 IEEE Conference on Computer Vision and Pattern Recognition. Compressed Hashing 203 IEEE Conference on Coputer Vision and Pattern Recognition Copressed Hashing Yue Lin Rong Jin Deng Cai Shuicheng Yan Xuelong Li State Key Lab of CAD&CG, College of Coputer Science, Zhejiang University,

More information

Guillotine subdivisions approximate polygonal subdivisions: Part III { Faster polynomial-time approximation schemes for

Guillotine subdivisions approximate polygonal subdivisions: Part III { Faster polynomial-time approximation schemes for Guillotine subdivisions approxiate polygonal subdivisions: Part III { Faster polynoial-tie approxiation schees for geoetric network optiization Joseph S. B. Mitchell y April 19, 1997; Last revision: May

More information

Oblivious Routing for Fat-Tree Based System Area Networks with Uncertain Traffic Demands

Oblivious Routing for Fat-Tree Based System Area Networks with Uncertain Traffic Demands Oblivious Routing for Fat-Tree Based Syste Area Networks with Uncertain Traffic Deands Xin Yuan Wickus Nienaber Zhenhai Duan Departent of Coputer Science Florida State University Tallahassee, FL 3306 {xyuan,nienaber,duan}@cs.fsu.edu

More information

Minimax Sensor Location to Monitor a Piecewise Linear Curve

Minimax Sensor Location to Monitor a Piecewise Linear Curve NSF GRANT #040040 NSF PROGRAM NAME: Operations Research Miniax Sensor Location to Monitor a Piecewise Linear Curve To M. Cavalier The Pennsylvania State University University Par, PA 680 Whitney A. Conner

More information

Distributed Multi-robot Work Load Partition In Manufacturing Automation

Distributed Multi-robot Work Load Partition In Manufacturing Automation th IEEE Conference on Autoation Science and Engineering Key Bridge Marriott, Washington DC, USA August 3-, Distributed Multi-robot Work Load Partition In Manufacturing Autoation Gira S Tewolde, Changhua

More information

Utility-Based Resource Allocation for Mixed Traffic in Wireless Networks

Utility-Based Resource Allocation for Mixed Traffic in Wireless Networks IEEE IFOCO 2 International Workshop on Future edia etworks and IP-based TV Utility-Based Resource Allocation for ixed Traffic in Wireless etworks Li Chen, Bin Wang, Xiaohang Chen, Xin Zhang, and Dacheng

More information

Collaborative Web Caching Based on Proxy Affinities

Collaborative Web Caching Based on Proxy Affinities Collaborative Web Caching Based on Proxy Affinities Jiong Yang T J Watson Research Center IBM jiyang@usibco Wei Wang T J Watson Research Center IBM ww1@usibco Richard Muntz Coputer Science Departent UCLA

More information

A Measurement-Based Model for Parallel Real-Time Tasks

A Measurement-Based Model for Parallel Real-Time Tasks A Measureent-Based Model for Parallel Real-Tie Tasks Kunal Agrawal 1 Washington University in St. Louis St. Louis, MO, USA kunal@wustl.edu https://orcid.org/0000-0001-5882-6647 Sanjoy Baruah 2 Washington

More information

Theoretical Analysis of Local Search and Simple Evolutionary Algorithms for the Generalized Travelling Salesperson Problem

Theoretical Analysis of Local Search and Simple Evolutionary Algorithms for the Generalized Travelling Salesperson Problem Theoretical Analysis of Local Search and Siple Evolutionary Algoriths for the Generalized Travelling Salesperson Proble Mojgan Pourhassan ojgan.pourhassan@adelaide.edu.au Optiisation and Logistics, The

More information

Minimax Sensor Location in the Plane

Minimax Sensor Location in the Plane NSF GRANT #040040 NSF PROGRAM NAME: Operations Research Miniax Sensor Location in the Plane To M. Cavalier The Pennsylvania State University University Par, PA 680 Enrique del Castillo The Pennsylvania

More information

The optimization design of microphone array layout for wideband noise sources

The optimization design of microphone array layout for wideband noise sources PROCEEDINGS of the 22 nd International Congress on Acoustics Acoustic Array Systes: Paper ICA2016-903 The optiization design of icrophone array layout for wideband noise sources Pengxiao Teng (a), Jun

More information

Mapping Data in Peer-to-Peer Systems: Semantics and Algorithmic Issues

Mapping Data in Peer-to-Peer Systems: Semantics and Algorithmic Issues Mapping Data in Peer-to-Peer Systes: Seantics and Algorithic Issues Anastasios Keentsietsidis Marcelo Arenas Renée J. Miller Departent of Coputer Science University of Toronto {tasos,arenas,iller}@cs.toronto.edu

More information

λ-harmonious Graph Colouring Lauren DeDieu

λ-harmonious Graph Colouring Lauren DeDieu λ-haronious Graph Colouring Lauren DeDieu June 12, 2012 ABSTRACT In 198, Hopcroft and Krishnaoorthy defined a new type of graph colouring called haronious colouring. Haronious colouring is a proper vertex

More information

Error estimation of bathymetric grid models derived from historic and contemporary datasets 1 Introduction

Error estimation of bathymetric grid models derived from historic and contemporary datasets 1 Introduction Error estiation of bathyetric grid odels derived fro historic and conteporary datasets Martin Jakobsson, Brian Calder, Larry Mayer and Andy Arstrong Center for Coastal and Ocean Mapping, University of

More information

MiPPS: A Generative Model for Multi-Manifold Clustering

MiPPS: A Generative Model for Multi-Manifold Clustering Manifold Learning and its Applications: Papers fro the AAAI Fall Syposiu (FS-9-) MiPPS: A Generative Model for Multi-Manifold Clustering Oluwasani Koyejo and Joydeep Ghosh Electrical and Coputer Engineering

More information

Medical Biophysics 302E/335G/ st1-07 page 1

Medical Biophysics 302E/335G/ st1-07 page 1 Medical Biophysics 302E/335G/500 20070109 st1-07 page 1 STEREOLOGICAL METHODS - CONCEPTS Upon copletion of this lesson, the student should be able to: -define the ter stereology -distinguish between quantitative

More information

An Efficient Approach for Content Delivery in Overlay Networks

An Efficient Approach for Content Delivery in Overlay Networks An Efficient Approach for Content Delivery in Overlay Networks Mohaad Malli, Chadi Barakat, Walid Dabbous Projet Planète, INRIA-Sophia Antipolis, France E-ail:{alli, cbarakat, dabbous}@sophia.inria.fr

More information

The Boundary Between Privacy and Utility in Data Publishing

The Boundary Between Privacy and Utility in Data Publishing The Boundary Between Privacy and Utility in Data Publishing Vibhor Rastogi Dan Suciu Sungho Hong ABSTRACT We consider the privacy proble in data publishing: given a database instance containing sensitive

More information

CS 361 Meeting 8 9/24/18

CS 361 Meeting 8 9/24/18 CS 36 Meeting 8 9/4/8 Announceents. Hoework 3 due Friday. Review. The closure properties of regular languages provide a way to describe regular languages by building the out of sipler regular languages

More information

Classification and Prediction of Motion Trajectories using Spatiotemporal Approximations

Classification and Prediction of Motion Trajectories using Spatiotemporal Approximations Classification and Prediction of Motion Trajectories using Spatioteporal Approxiations Andrew Naftel & Shehzad Khalid School of Inforatics University of Manchester Manchester, M6 1QD a.naftel@anchester.ac.uk;

More information

Different criteria of dynamic routing

Different criteria of dynamic routing Procedia Coputer Science Volue 66, 2015, Pages 166 173 YSC 2015. 4th International Young Scientists Conference on Coputational Science Different criteria of dynaic routing Kurochkin 1*, Grinberg 1 1 Kharkevich

More information

Scheduling Parallel Task Graphs on (Almost) Homogeneous Multi-cluster Platforms

Scheduling Parallel Task Graphs on (Almost) Homogeneous Multi-cluster Platforms 1 Scheduling Parallel Task Graphs on (Alost) Hoogeneous Multi-cluster Platfors Pierre-François Dutot, Tchiou N Takpé, Frédéric Suter and Henri Casanova Abstract Applications structured as parallel task

More information

1 Extended Boolean Model

1 Extended Boolean Model 1 EXTENDED BOOLEAN MODEL It has been well-known that the Boolean odel is too inflexible, requiring skilful use of Boolean operators to obtain good results. On the other hand, the vector space odel is flexible

More information

OPTIMAL COMPLEX SERVICES COMPOSITION IN SOA SYSTEMS

OPTIMAL COMPLEX SERVICES COMPOSITION IN SOA SYSTEMS Key words SOA, optial, coplex service, coposition, Quality of Service Piotr RYGIELSKI*, Paweł ŚWIĄTEK* OPTIMAL COMPLEX SERVICES COMPOSITION IN SOA SYSTEMS One of the ost iportant tasks in service oriented

More information

Wavelets for Computer Graphics: A Primer Part 1

Wavelets for Computer Graphics: A Primer Part 1 Wavelets for Coputer Graphics: A Prier Part Eric J. Stollnitz Tony D. DeRose David H. Salesin University of Washington Introduction Wavelets are a atheatical tool for hierarchically decoposing functions.

More information

Generating Mechanisms for Evolving Software Mirror Graph

Generating Mechanisms for Evolving Software Mirror Graph Journal of Modern Physics, 2012, 3, 1050-1059 http://dx.doi.org/10.4236/jp.2012.39139 Published Online Septeber 2012 (http://www.scirp.org/journal/jp) Generating Mechaniss for Evolving Software Mirror

More information

Modeling Parallel Applications Performance on Heterogeneous Systems

Modeling Parallel Applications Performance on Heterogeneous Systems Modeling Parallel Applications Perforance on Heterogeneous Systes Jaeela Al-Jaroodi, Nader Mohaed, Hong Jiang and David Swanson Departent of Coputer Science and Engineering University of Nebraska Lincoln

More information

Scheduling Parallel Real-Time Recurrent Tasks on Multicore Platforms

Scheduling Parallel Real-Time Recurrent Tasks on Multicore Platforms IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, VOL., NO., NOV 27 Scheduling Parallel Real-Tie Recurrent Tasks on Multicore Platfors Risat Pathan, Petros Voudouris, and Per Stenströ Abstract We

More information

Energy-Efficient Disk Replacement and File Placement Techniques for Mobile Systems with Hard Disks

Energy-Efficient Disk Replacement and File Placement Techniques for Mobile Systems with Hard Disks Energy-Efficient Disk Replaceent and File Placeent Techniques for Mobile Systes with Hard Disks Young-Jin Ki School of Coputer Science & Engineering Seoul National University Seoul 151-742, KOREA youngjk@davinci.snu.ac.kr

More information

Global Optimization of Transmitter Placement in Wireless Communication Systems

Global Optimization of Transmitter Placement in Wireless Communication Systems Global Optiization of Transitter Placeent in Wireless Counication Systes J. He, A. Verstak, L. T. Watson, T. S. Rappaport, C. R. Anderson, N. Raakrishnan, C. A. Shaffer, W. H. Tranter, K. Bae, J. Jiang

More information

Discrete Fourier Transform

Discrete Fourier Transform Discrete Fourier Transfor This is the first tutorial in our ongoing series on tie series spectral analysis. In this entry, we will closely exaine the discrete Fourier transfor (aa DFT) and its inverse,

More information

Ranking Spatial Data by Quality Preferences

Ranking Spatial Data by Quality Preferences 1 Ranking Spatial Data by Quality Preferences Man Lung Yiu, Hua Lu, Nikos Maoulis, and Michail Vaitis Abstract A spatial preference query ranks objects based on the qualities of features in their spatial

More information

Boosted Detection of Objects and Attributes

Boosted Detection of Objects and Attributes L M M Boosted Detection of Objects and Attributes Abstract We present a new fraework for detection of object and attributes in iages based on boosted cobination of priitive classifiers. The fraework directly

More information

Computer Aided Drafting, Design and Manufacturing Volume 26, Number 2, June 2016, Page 13

Computer Aided Drafting, Design and Manufacturing Volume 26, Number 2, June 2016, Page 13 Coputer Aided Drafting, Design and Manufacturing Volue 26, uber 2, June 2016, Page 13 CADDM 3D reconstruction of coplex curved objects fro line drawings Sun Yanling, Dong Lijun Institute of Mechanical

More information

Ascending order sort Descending order sort

Ascending order sort Descending order sort Scalable Binary Sorting Architecture Based on Rank Ordering With Linear Area-Tie Coplexity. Hatrnaz and Y. Leblebici Departent of Electrical and Coputer Engineering Worcester Polytechnic Institute Abstract

More information

A simplified approach to merging partial plane images

A simplified approach to merging partial plane images A siplified approach to erging partial plane iages Mária Kruláková 1 This paper introduces a ethod of iage recognition based on the gradual generating and analysis of data structure consisting of the 2D

More information

Smarter Balanced Assessment Consortium Claims, Targets, and Standard Alignment for Math

Smarter Balanced Assessment Consortium Claims, Targets, and Standard Alignment for Math Sarter Balanced Assessent Consortiu Clais, s, Stard Alignent for Math The Sarter Balanced Assessent Consortiu (SBAC) has created a hierarchy coprised of clais targets that together can be used to ake stateents

More information

POSITION-PATCH BASED FACE HALLUCINATION VIA LOCALITY-CONSTRAINED REPRESENTATION. Junjun Jiang, Ruimin Hu, Zhen Han, Tao Lu, and Kebin Huang

POSITION-PATCH BASED FACE HALLUCINATION VIA LOCALITY-CONSTRAINED REPRESENTATION. Junjun Jiang, Ruimin Hu, Zhen Han, Tao Lu, and Kebin Huang IEEE International Conference on ultiedia and Expo POSITION-PATCH BASED FACE HALLUCINATION VIA LOCALITY-CONSTRAINED REPRESENTATION Junjun Jiang, Ruiin Hu, Zhen Han, Tao Lu, and Kebin Huang National Engineering

More information

Joint Measurement- and Traffic Descriptor-based Admission Control at Real-Time Traffic Aggregation Points

Joint Measurement- and Traffic Descriptor-based Admission Control at Real-Time Traffic Aggregation Points Joint Measureent- and Traffic Descriptor-based Adission Control at Real-Tie Traffic Aggregation Points Stylianos Georgoulas, Panos Triintzios and George Pavlou Centre for Counication Systes Research, University

More information

Research on the Influence of Several Factors on Clustering Results using Different Algorithms

Research on the Influence of Several Factors on Clustering Results using Different Algorithms Send Orders for Reprints to reprints@benthascience.ae The Open Cybernetics & Systeics Journal, 2015, 9, 1099-1106 1099 Open Access Research on the Influence of Several Factors on Clustering Results using

More information

Modal Masses Estimation in OMA by a Consecutive Mass Change Method.

Modal Masses Estimation in OMA by a Consecutive Mass Change Method. Modal Masses Estiation in OMA by a Consecutive Mass Change Method. F. Pelayo University of Oviedo, Departent of Construction and Manufacturing Engineering, Gijón, Spain M. López-Aenlle University of Oviedo,

More information

Set Theoretic Estimation for Problems in Subtractive Color

Set Theoretic Estimation for Problems in Subtractive Color Set Theoretic Estiation for Probles in Subtractive Color Gaurav Shara Digital Iaging Technology Center, Xerox Corporation, MS0128-27E, 800 Phillips Rd, Webster, NY 14580 Received 25 March 1999; accepted

More information

A Low-Cost Multi-Failure Resilient Replication Scheme for High Data Availability in Cloud Storage

A Low-Cost Multi-Failure Resilient Replication Scheme for High Data Availability in Cloud Storage 216 IEEE 23rd International Conference on High Perforance Coputing A Low-Cost Multi-Failure Resilient Replication Schee for High Data Availability in Cloud Storage Jinwei Liu* and Haiying Shen *Departent

More information

Classification and Segmentation of Glaucomatous Image Using Probabilistic Neural Network (PNN), K-Means and Fuzzy C- Means(FCM)

Classification and Segmentation of Glaucomatous Image Using Probabilistic Neural Network (PNN), K-Means and Fuzzy C- Means(FCM) IJSRD - International Journal for Scientific Research & Developent Vol., Issue 7, 03 ISSN (online): 3-063 Classification and Segentation of Glaucoatous Iage Using Probabilistic Neural Network (PNN), K-Means

More information

Promoting similarity of model sparsity structures in integrative analysis of cancer genetic data

Promoting similarity of model sparsity structures in integrative analysis of cancer genetic data Research Article Received 7 Deceber 2014, Accepted 2 Septeber 2016 Published online in Wiley Online Library (wileyonlinelibrary.co) DOI: 10.1002/si.7138 Prooting siilarity of odel sparsity structures in

More information

Cassia County School District #151. Expected Performance Assessment Students will: Instructional Strategies. Performance Standards

Cassia County School District #151. Expected Performance Assessment Students will: Instructional Strategies. Performance Standards Unit 1 Congruence, Proof, and Constructions Doain: Congruence (CO) Essential Question: How do properties of congruence help define and prove geoetric relationships? Matheatical Practices: 1. Make sense

More information

An Integrated Processing Method for Multiple Large-scale Point-Clouds Captured from Different Viewpoints

An Integrated Processing Method for Multiple Large-scale Point-Clouds Captured from Different Viewpoints 519 An Integrated Processing Method for Multiple Large-scale Point-Clouds Captured fro Different Viewpoints Yousuke Kawauchi 1, Shin Usuki, Kenjiro T. Miura 3, Hiroshi Masuda 4 and Ichiro Tanaka 5 1 Shizuoka

More information

A Model Free Automatic Tuning Method for a Restricted Structured Controller. by using Simultaneous Perturbation Stochastic Approximation

A Model Free Automatic Tuning Method for a Restricted Structured Controller. by using Simultaneous Perturbation Stochastic Approximation 28 Aerican Control Conference Westin Seattle Hotel, Seattle, Washington, USA June -3, 28 WeC9.5 A Model Free Autoatic Tuning Method for a Restricted Structured Controller by Using Siultaneous Perturbation

More information

AGV PATH PLANNING BASED ON SMOOTHING A* ALGORITHM

AGV PATH PLANNING BASED ON SMOOTHING A* ALGORITHM International Journal of Software Engineering & Applications (IJSEA), Vol.6, No.5, Septeber 205 AGV PATH PLANNING BASED ON SMOOTHING A* ALGORITHM Xie Yang and Cheng Wushan College of Mechanical Engineering,

More information

EFFICIENT VIDEO SEARCH USING IMAGE QUERIES A. Araujo1, M. Makar2, V. Chandrasekhar3, D. Chen1, S. Tsai1, H. Chen1, R. Angst1 and B.

EFFICIENT VIDEO SEARCH USING IMAGE QUERIES A. Araujo1, M. Makar2, V. Chandrasekhar3, D. Chen1, S. Tsai1, H. Chen1, R. Angst1 and B. EFFICIENT VIDEO SEARCH USING IMAGE QUERIES A. Araujo1, M. Makar2, V. Chandrasekhar3, D. Chen1, S. Tsai1, H. Chen1, R. Angst1 and B. Girod1 1 Stanford University, USA 2 Qualco Inc., USA ABSTRACT We study

More information

Data Caching for Enhancing Anonymity

Data Caching for Enhancing Anonymity Data Caching for Enhancing Anonyity Rajiv Bagai and Bin Tang Departent of Electrical Engineering and Coputer Science Wichita State University Wichita, Kansas 67260 0083, USA Eail: {rajiv.bagai, bin.tang}@wichita.edu

More information

Trees. Linear vs. Branching CSE 143. Branching Structures in CS. What s in a Node? A Tree. [Chapter 10]

Trees. Linear vs. Branching CSE 143. Branching Structures in CS. What s in a Node? A Tree. [Chapter 10] CSE 143 Trees [Chapter 10] Linear vs. Branching Our data structures so far are linear Have a beginning and an end Everything falls in order between the ends Arrays, lined lists, queues, stacs, priority

More information

Depth Estimation of 2-D Magnetic Anomalous Sources by Using Euler Deconvolution Method

Depth Estimation of 2-D Magnetic Anomalous Sources by Using Euler Deconvolution Method Aerican Journal of Applied Sciences 1 (3): 209-214, 2004 ISSN 1546-9239 Science Publications, 2004 Depth Estiation of 2-D Magnetic Anoalous Sources by Using Euler Deconvolution Method 1,3 M.G. El Dawi,

More information

Simulated Sensor Reading. Heading. Y [m] local minimum ; m. local maximum ; M. discontinuity ; D. connection ; c X [m]

Simulated Sensor Reading. Heading. Y [m] local minimum ; m. local maximum ; M. discontinuity ; D. connection ; c X [m] Localization based on Visibility Sectors using Range Sensors Sooyong Lee y Nancy. Aato Jaes Fellers Departent of Coputer Science Texas A& University College Station, TX 7784 fsooyong,aato,jpf938g@cs.tau.edu

More information

Collection Selection Based on Historical Performance for Efficient Processing

Collection Selection Based on Historical Performance for Efficient Processing Collection Selection Based on Historical Perforance for Efficient Processing Christopher T. Fallen and Gregory B. Newby Arctic Region Supercoputing Center University of Alaska Fairbanks Fairbanks, Alaska

More information

Relief shape inheritance and graphical editor for the landscape design

Relief shape inheritance and graphical editor for the landscape design Relief shape inheritance and graphical editor for the landscape design Egor A. Yusov Vadi E. Turlapov Nizhny Novgorod State University after N. I. Lobachevsky Nizhny Novgorod Russia yusov_egor@ail.ru vadi.turlapov@cs.vk.unn.ru

More information

Approximate String Matching with Reduced Alphabet

Approximate String Matching with Reduced Alphabet Approxiate String Matching with Reduced Alphabet Leena Salela 1 and Jora Tarhio 2 1 University of Helsinki, Departent of Coputer Science leena.salela@cs.helsinki.fi 2 Aalto University Deptartent of Coputer

More information

Optimal Route Queries with Arbitrary Order Constraints

Optimal Route Queries with Arbitrary Order Constraints IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL.?, NO.?,? 20?? 1 Optial Route Queries with Arbitrary Order Constraints Jing Li, Yin Yang, Nikos Maoulis Abstract Given a set of spatial points DS,

More information

3 Conference on Inforation Sciences and Systes, The Johns Hopkins University, March, 3 Sensitivity Characteristics of Cross-Correlation Distance Metric and Model Function F. Porikli Mitsubishi Electric

More information

A Spectral Framework for Detecting Inconsistency across Multi-Source Object Relationships

A Spectral Framework for Detecting Inconsistency across Multi-Source Object Relationships A Spectral Fraework for Detecting Inconsistency across Multi-Source Object Relationships Jing Gao, Wei Fan, Deepak Turaga, Srinivasan Parthasarathy, and Jiawei Han University of Illinois at Urbana-Chapaign,

More information

Visualization and Processing of Mesospheric Hydroxyl Infrared Emissions from SABER

Visualization and Processing of Mesospheric Hydroxyl Infrared Emissions from SABER Visualization and Processing of Mesospheric Hydroxyl Infrared Eissions fro SABER Kenneth B. Reese Departent of Electrical and Coputer Engineering Utah State University 440 Old Main Hill Logan, UT 843-440

More information